# Longview — Complete Public Research Corpus > Independent fundamental company research, 11-stage economic underwriting notebooks, and dated forward shareholder IRR models. Canonical Site: https://longview-investing.vercel.app/ Summary Index: https://longview-investing.vercel.app/llms.txt --- # Constellation Software (CSU.TO) - Legal Name: Constellation Software Inc. - Exchange: Toronto Stock Exchange (Canada) - Sector: Vertical-market software - Currency: CAD - Reviewed: 2026-11-15 - Model Framework: Software cash and acquisition compounding - Financial Period: TTM September 2026 and 30 September balances; post-September commitments disclosed 6 November; 26 October FX and exits unchanged (anchored 2026-10-27) - Research Status: 8 of 11 research checks complete. Still to complete: Security and evidence base; Opportunities and funded scenarios; Variant view and catalysts - Static Report URL: https://longview-investing.vercel.app/research/constellation-software.html - Interactive Model URL: https://longview-investing.vercel.app/#/company/constellation-software - JSON API: https://longview-investing.vercel.app/api/companies/constellation-software ## Executive Summary Recurring specialist software cash funds a decentralized acquisition engine. Customer durability, capital deployment, purchase yields and ownership claims determine per-share compounding. ## Key Operational Watch Items - Recurring organic growth and customer retention, especially Altera - Acquisition cash yields and deployment capacity at increasing scale - IRGA/Joday claim, minority ownership, debt service and subsidiary cash restrictions ## Valuation Scenario Assumptions ### Downside Case Stress 2% annual erosion of the existing cash base, deploy only half the available surplus, earn an 8% mature purchase cash yield and exit at 15x. This tests weaker retention, Altera pressure, poorer deal economics and a shorter runway; retained cash remains an asset. It is not a default scenario. - Key Drivers: Exit / common operating cash after financing: 15.00× | Annual share dilution: 0.0% | Available cash deployment fraction: 50.0% - Organic revenue growth: 0.0%, -2.0%, -2.0%, -2.0%, -2.0% - Owner-cash margin on organic revenue: 17.5%, 16.5%, 16.5%, 16.5%, 16.5% - Price paid per unit of acquired revenue: 1.50×, 1.50×, 1.50×, 1.50×, 1.50× - Owner-cash margin of acquired businesses: 12.0%, 12.0%, 12.0%, 12.0%, 12.0% - Acquired revenue growth after purchase: -2.0%, -2.0%, -2.0%, -2.0%, -2.0% - Deal budget (capital available for acquisitions): 1,879.35, 680.20, 680.20, 1,879.40, 1,879.40 ### Base Case Latest group organic growth is 1% after FX. Continue current ongoing spending and flat dividends; deploy surplus into cash-funded acquisitions at the group's achieved 2020–2025 aggregate acquisition cash yield. Exit at a chosen 20x operating equity cash multiple, with investments and all retained claims added/deducted separately. - Key Drivers: Exit / common operating cash after financing: 20.00× | Annual share dilution: 0.0% | Available cash deployment fraction: 100.0% - Organic revenue growth: 1.5%, 2.0%, 2.0%, 3.0%, 3.0% - Owner-cash margin on organic revenue: 17.5%, 17.5%, 17.5%, 17.5%, 17.5% - Price paid per unit of acquired revenue: 1.17×, 1.17×, 1.17×, 1.17×, 1.17× - Owner-cash margin of acquired businesses: 15.0%, 16.5%, 16.5%, 17.8%, 17.8% - Acquired revenue growth after purchase: 1.5%, 2.0%, 2.0%, 3.0%, 3.0% - Deal budget (capital available for acquisitions): 2,315.11, 2,384.60, 2,456.10, 2,529.80, 2,605.70 ### Upside Case Use 3% existing cash growth, full cash deployment, the achieved 15.1% aggregate purchase cash yield (no better-than-record deal prices) and a 25x exit. Requires resilient customer economics and continued sourcing discipline. No margin expansion, new leverage or rerating of retained investments is credited. - Key Drivers: Exit / common operating cash after financing: 25.00× | Annual share dilution: 0.0% | Available cash deployment fraction: 100.0% - Organic revenue growth: 2.5%, 4.0%, 4.0%, 4.0%, 4.0% - Owner-cash margin on organic revenue: 17.5%, 17.5%, 17.5%, 18.0%, 18.0% - Price paid per unit of acquired revenue: 1.17×, 1.17×, 1.17×, 1.17×, 1.17× - Owner-cash margin of acquired businesses: 16.0%, 17.8%, 17.8%, 17.8%, 17.8% - Acquired revenue growth after purchase: 2.5%, 4.0%, 4.0%, 4.0%, 4.0% - Deal budget (capital available for acquisitions): 3,465.58, 3,716.60, 3,716.60, 4,516.10, 4,516.10 ## Research Notebook # Business model USD millions unless stated. Facts are in `financials/facts.json`; named arithmetic (prefixes `snap`, `seg`, `ue`) is in `valuation/calculations.json` as audit steps that do not feed the model. Sources: FY2025 shareholder report (`csi-fy2025`), FY2022 and FY2023 reports, Q2 2026 report (`csi-q2-2026`), the 2015 and 2021 letters (`csi-2015-letter`, `csi-2021-capital`). Constellation (CSU) owns a large number of specialist software businesses (the reports read give no count) serving narrow workflows: municipalities, transit, utilities, hospitals, clubs, banks, media, construction and so on. Six operating groups (Volaris, Harris, Topicus, Vela, Jonas, Perseus) are aggregated into one reportable segment because each "operates essentially as a mini Constellation" (FY2025 note 23). Topicus (Europe) and Lumine (communications and media) are separately listed and consolidated through super-voting shares. Common shares trade in CAD; reports are in USD. ## Value-creation mechanism Buy, hold forever, and recycle the cash. A vertical-market software (VMS) business is bought for a price close to a small multiple of its after-tax cash; the customer keeps paying maintenance because replacing mission-critical software means data migration, retraining and operational risk; head office and the operating group impose cost discipline and best-practice benchmarking; the business's surplus cash is then sent back (or kept, under the "keep your capital" rule) and spent on the next purchase. The step where most value is created is **the purchase price**. CSU's own analysis of every acquisition since 2004 found that the revenue multiple paid was a far more obvious driver of IRR than post-acquisition organic growth (2015 letter, pp.5-6). Integration and margin repair matter (Altera below), but the edge is paying a price that a permanent, decentralized owner can justify and a leveraged buyer flipping in three to seven years usually cannot (2021 letter). The second step, recycling, works because the businesses need almost no tangible capital: customers prepay, so organic growth is funded by customers, and nearly all reinvestment goes into the next acquisition. ## Business-model profile Comparable firms are chosen by these traits, not by the industry label. **Growth vehicle: Acquisitions.** Revenue rose from 3,969 in FY2020 to 11,623 in FY2025, about 24% a year, while FX-adjusted organic growth was 3% in FY2025 (5% for recurring revenue). Acquisition cash was 1,847, 1,683 and 1,513 in FY2023–FY2025, 16.8% of revenue. So roughly three points of growth a year come from the installed base and the rest is bought. The model therefore buys revenue every year at a measured price and grows the existing businesses separately at the installed-base rate; organic growth alone would describe a low-growth software company that CSU is not. **Capital intensity: 0.6% of sales, asset-light.** Capital spending is about 0.6% of revenue: offices and hardware for staff who write and maintain vertical-market software. Tangible capital including leases is negative (−982 in FY2025) because customers prepay maintenance (deferred revenue of 70 days). The capital that drives growth is acquisition cash, about 17% of revenue a year, not capex. **Comparables.** A comparable firm buys many small businesses every year with internally generated cash, needs almost no capital spending, and earns mostly recurring maintenance revenue (75% of FY2025) from customers who rarely replace mission-critical systems. Its constraint at this scale is finding enough deals at its price, so comparables are serial acquirers of small asset-light businesses, not software product companies. Not the label: "Software" suggests product-led organic growth valued on software multiples; CSU buys its growth and grows about 3% a year organically. Declared in `valuation/model.json` (`businessProfile`). `npm run check` measures deal and capital intensity from the reported facts, and the outside view grades the forecast only against firms that share them. ## Economic snapshot FY2025, consolidated, reported definitions unless marked. EBITA here is revenue less operating expenses including depreciation (so leases are capitalized), before acquired-intangible amortization, FX, IRGA revaluation, finance items and impairments. | Measure | FY2025 | Definition / source step | | --- | ---: | --- | | Revenue | 11,623 | 75% maintenance and other recurring (8,700); licenses 415; services 2,126; hardware 382 | | Direct margin | 67.7% | Revenue less hardware cost, third-party costs and maintenance and services staff (CSU reports no gross margin) | | EBITA margin | 26.5% | `snapEbitaMargin2025`; 28.4% in FY2020 (`snapEbitaMargin2020`) | | Incremental EBITA margin | 36.6% FY2024→25; 25.5% FY2020→25 | `snapIncrementalMargin2025`; acquired businesses enter at their own (often lower) margins | | R&D staff / revenue | 13.9% | 14.3-14.6% in every year 2020-2024 (facts `csuRdStaff*`) | | Receivables + unbilled | 60 days of revenue | 1,920 | | Deferred revenue | 70 days of revenue | 2,214 | | Net operating working capital | (1,607) | `snapWorkingCapital2025`; customers fund the business | | PP&E + right-of-use assets | 625 | 241 + 384 | | Tangible capital incl. leases | **(982)** | `snapTangibleCapital2025` | | Intangibles incl. goodwill, net / at cost | 8,388 / 15,157 | goodwill at cost 1,815 | | Debt / leases / IRGA claim / holdbacks | 4,131 / 423 / 1,233 / 432 | debt 1,489 recourse to CSI, 2,642 non-recourse | | Cash | 3,089 | 1,172 held where upstream distribution is restricted | | NOPAT, leases expensed | 2,455 | `snapNopat2025`: EBITA less current tax (606) and lease interest (16) | | Return on tangible capital | not meaningful | tangible capital is negative | | ROCE on amortized capital | 36.9% | `snapRoceAmortized`, average capital 6,661 | | ROCE on acquisition cost | **19.3%** | `snapRoceGross`, average capital 12,688 | Normalization. Capex (68) roughly equals PP&E depreciation (total depreciation 201 less lease principal 131 ≈ 70), so no sustaining-capital adjustment is needed. Recently acquired businesses are only partly in FY2025: Cipal Schaubroeck alone was missing 49 of revenue (pro forma 11,672 vs 11,623). Minorities took 243 of the 1,926 pre-NCI FCFA2S (12.6%). The model's common-owner bridge (stage 7, `research/07-assumption-register.md`) handles these; this snapshot is the consolidated base it explains. What the snapshot says. Return on tangible capital is meaningless for CSU: the business runs on customers' money, so the only capital that matters is the price paid for each business. Measured against that price (intangibles at cost), the portfolio earns about 19% after tax, well below the 37% on amortized book that flatters any accounting ROIC. That 19% blends decades-old purchases earning far above cost with recent ones earning less; the marginal return is measured separately (stage 6: 15.1%). ## Economic segmentation The reported segment table is one line. Re-cut by how value is created, using the supplemental Topicus, Lumine and Altera disclosures (FY2025 MD&A, PDF pp.15-17) and CSU's economic ownership (`segLookThroughCfo`, operating cash before interest, at accounting ownership): | Economic unit | FY2025 revenue | Operating cash | CSU share | Look-through cash | Share | Demand driver and trend | | --- | ---: | ---: | ---: | ---: | ---: | --- | | Core small and mid VMS (CSU ex Topicus, Lumine, Altera) | 8,400 | 1,930 | ~100% | 1,930 | 78% | Installed-base maintenance and price; FX-adjusted organic 5% recurring, 3% total in FY2025. Growing slowly, acyclical | | Topicus (European VMS) | 1,746 | 448 | 60.65% deemed | 272 | 11% | Dutch and European public sector, education, health; organic 4% total, 7% recurring. Growing; CSU's direct stake is ~31% plus the IRGA purchase claim | | Lumine (communications and media) | 766 | 236 | 61.4% | 145 | 6% | Telecom and broadcast software; organic 0%. Larger carve-outs (WideOrbit 2023, Synchronoss 2026) | | Altera (hospital EHR carve-out, 2022) | 712 | 117 | 100% | 117 | 5% | Hospital IT budgets; revenue down 16% since purchase; recurring −3% FY2025, −19% Q2 and −8% Q3 2026; 9M 2026 revenue 445, 4% of group. Declining | | Financial stakes (Asseco, other) | — | dividends 69 in H1 2026 | — | outside | — | Valued separately at fair value (Asseco 1,092 group, September 2026) | So about four-fifths of CSU's owner cash comes from the small and mid VMS engine whose economics are described above. The large carve-outs (Altera, Optimal Blue, WideOrbit, Synchronoss) are a different mechanism: buy a neglected division cheaply relative to revenue, cut cost to group norms and accept shrinking revenue. They are a growing share of capital but not yet of cash. Profit by revenue type is even more concentrated (unit economics below): recurring maintenance earns roughly 85% of EBITA on 75% of revenue. Revenue share is not profit share here: licenses and services were 22% of revenue but about 13% of EBITA, and they are the cyclical, discretionary part (licenses −8% and services −4% FX-adjusted organic in FY2025, while recurring grew 5%). ## Unit economics The unit through which CSU adds value is **an acquired VMS business**. Two views: a revenue-type P&L that shows where the profit sits inside every business, and observed deals that show what CSU pays and earns. **Inside the unit: profit by revenue type, FY2025** (allocations are estimates; `ue*` steps). Third-party costs (1,114) are allocated by software revenue; all R&D staff (1,615) is charged to recurring revenue, the adverse choice for that line; shared overhead (S&M 702, G&A 1,181, occupancy, travel/telecom/software, professional fees, other and depreciation: 3,175 in total) is allocated by revenue. | | Recurring | Services | Licenses | Hardware | Total | | --- | ---: | ---: | ---: | ---: | ---: | | Revenue | 8,700 | 2,126 | 415 | 382 | 11,623 | | Direct staff (maintenance, services) | (1,223) | (1,202) | — | — | | | R&D staff | (1,615) | — | — | — | | | Third-party / hardware cost | (862) | (211) | (41) | (217) | | | Contribution | 5,000 (57%) | 713 (34%) | 374 (90%) | 165 (43%) | 6,252 | | Shared overhead by revenue | (2,377) | (581) | (113) | (104) | (3,175) | | EBITA | **2,623 (30%)** | 133 (6%) | 261 (63%) | 61 (16%) | 3,077 | | Share of EBITA | **85%** | 4% | 8% | 2% | | Capital per unit is the purchase price. Working capital is negative (deferred revenue exceeds receivables), PP&E is about 2% of revenue and capex matches depreciation. At the group's 2025 unlevered owner-cash margin (21.7% of revenue: operating cash after tax, leases and capex) and the measured 6.61x gross purchase multiple (stage 6; 6.57x at entry on net capital in the current model), a typical acquired business costs about 1.4x revenue and earns about 15% after tax on that price once integrated. **Observed units** (FY2022, FY2023 and FY2025 acquisition notes; annual revenue from the issuer's pro forma figures): | Deal | Price | Annual revenue at purchase | Price / revenue | Profit at purchase | Later evidence | | --- | ---: | ---: | ---: | --- | --- | | Altera (Allscripts hospitals carve-out, May 2022) | 727 | 846 | 0.86x | net income 0 in first 8 months | FY2025 FCFA2S 92 (12.7% of price; 12.9% in FY2023) on revenue 712 (−16%) | | WideOrbit (Feb 2023, via Lumine) | 504 incl. special shares | 172 | 2.9x | net income 1 in 2023 | not separately disclosed | | Optimal Blue (Sep 2023) | 705 | 187 | 3.8x | net income 0 in 2023 | not separately disclosed | | Cipal Schaubroeck (Jun 2025) | 258 | 122 | 2.1x | net loss 6 in 7 months | — | | Synchronoss (Feb 2026, via Lumine) | 309 cash | ~169 (Q2 x4) | 1.8x | post-close losses | `lumine-q2-2026-fs` | **Before and after the intervention.** Every disclosed large deal arrives at roughly zero profit and is bought at 0.9-3.8x revenue; the group's acquired cash yield of about 15% means CSU lifts these units toward a cash margin of 15-25% of revenue. Altera is the cleanest case: a carve-out that earned nothing for Allscripts became a business producing about 13% FCFA2S on its price within a year, by cost restructuring (expenses were 558 on 712 of revenue in FY2025, 78%, against 73% for the group). **The failure, or the limit of the levers.** The same Altera shows what cost discipline cannot do: revenue fell from about 846 to 712, professional services fell 16% organically in FY2025 and recurring revenue fell 19% in Q2 2026, partly against one-off prior-year contract revenue. A 12.7% levered yield on a shrinking base is below the group aggregate and below CSU's historical small-deal returns. The large-deal practice that the 2021 letter announced (lower hurdles for multi-hundred-million cheques) therefore produces lower and more fragile returns, as Leonard predicted. ## Independent checks 1. **Group accounts against the unit model.** Recurring revenue of 8,700 with maintenance and R&D staff of 2,838 implies a 33% staff ratio; group staff expense is 51% of revenue, and the gap is services, S&M and G&A, consistent with the allocation. The unit model's 30% recurring EBITA margin and 6% services margin sum exactly to the reported 3,077 EBITA (the allocation reconciles by construction; the test is that no line goes negative under reasonable keys). 2. **Adverse allocation.** Charge all 3,175 of shared overhead to recurring revenue: recurring EBITA falls to 1,825, still 59% of the total, and services and licenses together would be 35%. Charging part of R&D to licenses (new products) would raise recurring's share. The conclusion that the installed-base maintenance stream is the profit pool survives every allocation tried. 3. **Implied price against observed prices.** The aggregate 6.61x cash multiple on gross capital (6.57x at entry on net capital since the 3 November reconciliation) implies about 1.4x revenue at group margins. Observed large deals were 0.9-3.8x revenue at near-zero margins. CSU pays more per dollar of revenue for businesses whose margins it expects to raise, and less for shrinking ones (Altera). That is consistent with the 2015 letter's finding that the revenue multiple drives IRR. 4. **Impairment testing.** Goodwill and intangibles are tested against revenue multiples of recurring revenue for each business unit (FY2025 note 10). FY2025 impairments were 43 on 15,157 of intangibles at cost (0.3%; 28 in FY2024). Few acquired units have failed badly enough to be written down, although this test is lenient by design. 5. **Group ROCE against marginal return.** 19.3% on all acquisition cost against 15.1% measured on 2021-2025 deals (stage 6): the newer capital earns less than the stock, as expected when hurdles were lowered for larger deals. Where the consensus disagrees. The market's live debate (2025-2026 commentary) is whether AI-assisted development lets cheap competitors replace these products, which would hit the recurring line that carries 85% of profit. The unit model says that is the right place to worry and that services and license weakness, which the market sees first, matters little to cash. The observation that would settle it is **FX-adjusted recurring organic growth excluding Altera** (4% in Q2 2026) and attrition commentary in the Q3 2026 MD&A on 6 November and following quarters; a move below about 0% for several quarters would mean the profit pool itself is shrinking. ## Q3 2026 check — 6 November 2026 Q3 confirmed the economics above rather than changing them. Recurring revenue outside Altera grew 4% after FX (Q2: 4%), 5% outside US healthcare and Altera, 7% at Topicus and 3% in the CSU core. Licenses (−17%) and services (−5%) kept shrinking, so group organic growth was 2%. Altera's recurring decline narrowed to −8% (Q2 −19%), with sizable client departures expected to run until the end of 2027. CSU also bought two US healthcare businesses whose very high customer attrition it says it modeled into the price, a reminder that the purchase yield, not the target's growth, is the decision variable. Source: Q3 MD&A pp.4–5, 16–17. # History, demand and counterparties USD millions unless stated. CSU learned to compound through many small software acquisitions, then delegated capital allocation to its operating groups as it grew. The 2015 letter (`csi-2015-letter`, pp.2-9) reported ROIC rising from 21% (2006) to 38% (2015) with organic net revenue growth between −3% and 8% a year, and warned that larger deals at lower hurdles would pull returns down. In February 2021 (`csi-2021-capital`) Leonard stopped special dividends, lowered hurdles for "very large" VMS deals and began investing outside VMS; Altera (2022), Optimal Blue and WideOrbit (2023), the Asseco stake (2025) and Synchronoss (2026) followed. The Topicus (2021) and Lumine (2023) spin-outs created listed subsidiaries with minority holders, so per-share earning power needs an ownership bridge before comparing years. Leonard resigned as President on 25 September 2025 and Mark Miller replaced him (`csi-mic-2026`, p.14). No external CSU thesis was supplied; this study reconstructs reported definitions and builds its own current estimates. ## Demand drivers and cycle **Who uses, who pays, who decides.** The end user is a clerk, dispatcher, nurse, loan officer or club manager running a daily workflow. The payer is usually an institution's operating budget (a municipality, transit authority, hospital, bank or private SME), and the decision-maker is a department head or IT director who personally bears the risk of a failed replacement. That asymmetry matters: the person who could save money by switching is the person who would be blamed for an outage, so incumbency is protected beyond what price alone would suggest. No customer exceeds 5% of revenue (FY2025 note 23), and a large share of receivables is with public-sector governments (FY2025 note 21). **Demand by type** (FY2025 revenue 11,623, `csi-fy2025` PDF p.5): | Type | Revenue | Share | FX-adjusted organic FY2025 | Character | | --- | ---: | ---: | ---: | --- | | Maintenance and other recurring (support, SaaS, hosting, transactions) | 8,700 | 75% | +5% | Replacement-like: the installed base renews; price increases and module adds drive growth | | Professional services | 2,126 | 18% | −4% | Implementations and projects: growth- and discretionary | | Licenses | 415 | 4% | −8% | New customers and new modules: growth, discretionary | | Hardware and other | 382 | 3% | +4% | Pass-through, discretionary | Regulation drives part of recurring demand (tax and reporting rule changes in public-sector and health software must be implemented each year) but is not separately disclosed. **The cycle, measured.** The 2015 letter's table shows organic net revenue growth of −3% in 2009 and −2% in 2010, against ROIC that stayed at 24-26% (`csi-2015-letter` p.2). In the Covid quarter (Q2 2020) FX-adjusted organic revenue fell 7%, but licenses fell 28% and maintenance and recurring only 1%; recurring was back to +2% by Q4 2020 (`csi-fy2021-mda`, quarterly organic table). In 2024-2025 licenses swung between −22% and +17% by quarter while recurring stayed at +4% to +6% every quarter (FY2025 report p.5). The recurring stream is close to acyclical; the cyclical part is small and low-margin (stage 3: services and licenses about 13% of EBITA). **The market misperception being tested.** Since late 2025 the market has priced the fear that AI-assisted development lets cheaper competitors replace incumbent vertical software or forces price cuts (the stock fell from about C$5,000 in mid-2025 to C$2,196 at its 52-week low, `csu-price-history-2026-11-03`). The series that tests it is FX-adjusted recurring organic growth, especially outside Altera: 5% in FY2025, 2% in Q2 2026 for the group and 4% excluding Altera, whose recurring revenue fell 19% (`csi-q2-2026`, MD&A pp.5 and 16). So far the data show deceleration concentrated in one acquired business with a known contract issue, not broad erosion. That is not evidence the fear is wrong; AI substitution, if it comes, would appear first in new-customer wins and then in renewals, and four quarters of data cannot rule it out. ## Competition Competition happens in two separate markets. **For customers, inside each micro-vertical.** Each business typically faces a handful of specialist rivals in its niche and country plus horizontal platforms (ERP, CRM, low-code) at the edges. In US public-sector software the largest specialist is Tyler Technologies; in several niches Roper's software businesses compete; in Europe the rivals are mostly local private firms. Share by niche is not disclosed and was not estimated. The number of competitors per niche has historically been stable because niches are too small to attract large entrants. The advantage, in one sentence: the incumbent's software holds the customer's data and procedures, so a rival must offer enough to justify a risky migration that the buyer's decision-maker gains little from. Evidence it is durable: recurring revenue grew organically in every quarter of 2024-2025 and fell only 1% in the Covid trough, while CSU kept R&D at 14-15% of revenue to maintain the products. The competitor that could hurt CSU is not a named firm but an AI-native vendor that cuts the cost of building and of **migrating** to a replacement; it hasn't yet because the binding constraint is the customer's risk and data conversion, not the cost of writing code. **For acquisitions.** Sellers choose among private-equity buyers, strategic acquirers (Roper, Tyler, other serial VMS buyers) and CSU's own groups, including Topicus and Lumine. Leonard's description (`csi-2021-capital`): 40-70 "large" VMS businesses are sold each year, mostly through fewer than a dozen brokers; CSU knew of about 80% of them over five years but was invited into only 16% of processes, and most competitors maximize leverage and resell within three to seven years. CSU's advantage in small deals is a proprietary, relationship-sourced pipeline built over 30 years and a promise of permanent ownership; in large deals it has none, which is why it lowered hurdles to be invited. The competitor that could hurt is private equity with cheap debt bidding up prices; when rates were near zero (2021-2022) CSU's deployment still rose (1,337 → 1,782), so the measured 2021-2025 yield of about 15% was earned against that competition. ## Counterparty incentives | Counterparty | What they want | What CSU gives | Their outside option | What would change the arrangement | | --- | --- | --- | --- | --- | | Customers (municipalities, hospitals, SMEs) | Software that keeps working, compliance updates, predictable cost, no failed projects | Continuity, maintained products (R&D 13.9% of revenue), modest annual price increases | Replace via a tender or project: migration, retraining, risk to the decision-maker | Cheaper migration (AI-assisted conversion), a modern product the users demand, or aggressive price increases that make a project worth the risk. Altera shows hospitals do leave or cut scope when a product falls behind | | Founders and corporate sellers | Founders: a permanent home that keeps staff and brand; corporate parents (Allscripts for Altera): speed and certainty for a non-core unit | Buy-and-hold, autonomy for the business-unit manager, quick close | Private equity at a higher price with a later resale; a strategic buyer | If CSU's permanent-owner reputation weakened (forced resales, centralization) it would have to pay auction prices | | M&A brokers | Competitive auctions and fees | Lower hurdles on large deals so CSU bids | Exclude CSU from processes (they did, per the 2021 letter) | More CSU wins at auction prices would raise invitations but lower returns | | Business-unit and operating-group managers | Bonus and wealth | Bonus on ROIC (net of a 5% rate) and net revenue growth; 75% of after-tax bonus in escrowed shares (`csi-mic-2026` pp.12-14) | Leave for PE-backed rivals (one-year non-competes) | A change to the bonus plan; Leonard noted it took years to rebuild trust after the 2011 sale attempt (`csi-2015-letter` p.7) | | Joday Group (Topicus founders' holding) | Liquidity and value for its Topicus Coop units | IRGA/membership agreement: CSU must buy the units at a formula tied to Topicus recurring revenue, net assets and investment prices (`csi-2021-ownership`) | Hold; the agreement sets exercise windows | The claim (1,186 at June 2026) grows with Topicus; the model carries it separately | | Topicus and Lumine minority holders | Per-share value of their listed stock | Access to CSU's method; CSU keeps 50.1% votes through a super-voting share | Sell their shares | They cannot change control; their claim is in NCI (243 of FY2025 pre-NCI FCFA2S) | | Lenders (non-recourse subsidiary debt, CSI senior notes) | Covenants met, interest paid | Recurring cash, low leverage (debt 4,131 against EBITA 3,077) | Refuse refinancing | A covenant breach at a subsidiary could trap cash (1,172 at year-end is already restricted) | Dependence runs both ways with customers: CSU needs their renewals, but each customer needs the software more than CSU needs any one customer. With sellers, the dependence runs toward CSU only in small, relationship-sourced deals; in large deals the brokers hold the power, which is why returns there are lower. # Management and incentives Sources: 2026 management information circular (`csi-mic-2026`; printed pages), the 2015 and February 2021 letters, issuer announcements (`csi-leadership-2026`, `csi-ai-2026`), and market data (`csu-price-history-2026-11-03`). Money in USD unless marked C$. Who decides where the cash goes: business-unit and operating-group managers propose and close most acquisitions within hurdle rates set by head office and the board (FY2025 note 23); head office (now Mark Miller as President, Jamal Baksh as CFO) allocates between groups, pursues large deals and non-VMS investments, and sets hurdles. Mark Leonard, founder and President since 1995, resigned as President on 25 September 2025, did not stand for re-election in May 2026, and stays as an adviser on "permanent engaged minority" investments (`csi-leadership-2026`). ## Ownership and incentives Holdings at 28 March 2026 (MIC pp.21-22); value at C$3,397.23 (3 November 2026); pay is 2025 total compensation (MIC pp.15-16). | Person | Role | CSU shares | Value (C$m) | Other holdings | 2025 pay (USD) | Stake ÷ pay | | --- | --- | ---: | ---: | --- | ---: | ---: | | Mark Miller | President and COO (since Sep 2025); Exec. Chair Volaris; director since 2025 | 254,533 (1.20%) | 865 | 521,237 Topicus; 806,261 Lumine | 1,036,219 (salary and bonus waived from 1 Jan 2026) | ~600x | | John Billowits | Chair; CSI 2003-2020 (CFO, CEO Vela) | 35,961 | 122 | 74,086 Topicus; 113,106 Lumine | 60,000 fees | — | | Jamal Baksh | CFO and director; at CSI since 2003 | 1,811 | 6.2 | 4,888 Topicus; 7,033 Lumine | 866,296 | ~5x | | Lawrence Cunningham | Vice-Chair, independent | 1,324 | 4.5 | small | 120,000 fees | — | | Robert Kittel | Independent, since 2013 | 1,203 | 4.1 | small | 100,000 fees | — | | Other five independent nominees | Kennedy, Pastor, Parr, Schultz | 108-369 each | <1.3 each | small | 60,000-80,000 fees | — | | Jeff Bender, Robin van Poelje, Damian McKay | CEOs of Harris, Topicus, Vela (not directors) in the 2026 MIC; van Poelje handed the Topicus CEO role to Ramon Zanders (previously TSS Blue CEO) on 1 October 2026 and remains Topicus Chair (`topicus-ceo-2026-09`) | not disclosed in MIC | — | Van Poelje must buy Topicus shares | 1.54m, 2.12m, 2.57m | — | | Mark Leonard | Founder; adviser since 2026 | not disclosed (not a nominee) | — | — | waived salary and bonus 2023-2025 | — | No one owns more than 10% of the common shares (MIC p.11). The share count has been 21,191,530 since at least 2020; there are no options, restricted units or other share compensation. Directors must spend after-tax fees on shares held in escrow for an average of four years (MIC p.17). **What compounding does to these stakes.** If the stock compounds at 10% a year for five years, Miller's stake gains about C$528m, roughly 500 times his last salary and bonus, which he has in any case waived. For the President, ownership dominates pay entirely. For the CFO the stake is about five years of pay, and for operating-group CEOs (holdings not disclosed) the bonus is the larger lever, so the bonus design matters most for the people who actually close deals. **The bonus design** (MIC pp.12-14). Bonus = base salary × company performance factor × individual factor. The company factor rests on ROIC and net revenue growth. ROIC is "net income for bonus purposes" (excluding amortization, deferred tax, unrealized FX, bargain gains and minorities' share) divided by "Average Invested Equity Capital", a running tally of shareholders' money that adds bonus-basis income and subtracts dividends. A 5% "risk-free" rate is netted from ROIC in all but one operating group; below it, no bonus. Groups under a "keep your capital" policy treat dividends to the parent as notional loans, so capital can't be shrunk to inflate ROIC. Executives generally must buy shares with 75% of the after-tax bonus, escrowed for an average of four years (one all-cash year in five). The board did not exercise discretion for 2025. Operating-group managers are paid on their own group's ROIC, not the group's. What the targets imply. Because amortization is excluded and capital is a running tally, ROIC here is close to cash return on all money ever invested, measured without writing it down; it can't be manufactured by amortizing old deals. The weaknesses: the 5% threshold is below any reasonable cost of equity (about 9%), and the growth term rewards size, so a manager earns more bonus from a deal at 7% than from no deal. ROIC still falls when that happens, which limits it. The weighting between the two terms is not disclosed, so the implied hurdle can't be backed out. ## Track record and references **Total shareholder return.** C$100 at December 2020 became C$206 by December 2025, against C$211 for the S&P/TSX Composite (MIC p.14), so the last five years matched the index after a 2024 peak of C$277. Price alone (unadjusted for the Topicus and Lumine distributions or dividends) went from C$345 (December 2014) to C$3,301 (December 2025), about 23% a year, and has since moved between C$2,196 and C$3,605 over 52 weeks (`csu-price-history-2026-11-03`). **Returns on capital before and after.** Bonus-basis ROIC rose from 21% (2006) to 38% (2015) while average invested capital grew from 123 to 965 (`csi-2015-letter`, table 1). On today's evidence, which uses different definitions: return on all acquisition cost about 19% after tax (FY2025, stage 2), and the measured 2021-2025 deal yield 15.1% (stage 6). Returns on new capital have fallen by about half since the large-deal practice began in 2021, as Leonard said they would; the absolute sum deployed rose from 582 (2020) to 2,501 (twelve months to June 2026). **Decisions that failed or disappointed.** - Altera: bought at 0.86x revenue in 2022; FCFA2S of 92 in FY2025 is 12.7% on price, but revenue is 16% below the purchase run rate and recurring revenue fell 19% in Q2 2026 (stage 3). - Hurdle-rate cuts: CSU's own review of 1995-2015 vintages found each cut dragged the expected IRR of all later deals down toward the new hurdle ("hurdle rates are magnetic", 2015 letter p.6). The 2021 decision to cut hurdles for large deals is the same experiment, now running. - 2011: when major investors put CSU up for sale, managers maximized short-term profit and cut acquisitions and initiatives (2015 letter p.7), which shows how fragile the incentive culture is to an ownership change. - Public-equity investments before 2015 earned high IRRs that Leonard called "largely unimportant" because they were small and short (2015 letter p.3). Asseco (580 in 2025) is the first large test of the non-VMS circle of competence; no return can be measured yet. **References.** No independent references were collected (no interviews with former employees, sellers or competitors); this is a gap, recorded as zero of zero. Published evidence: Leonard's letters (candid about mistakes and incentive problems), annual meeting Q&A, and the CFO's 2021 ownership Q&A. Secondary commentary since September 2025 attributes part of the share-price fall to Leonard's departure on health grounds; that is evidence of a market reaction, not of a change in decisions. **Bench.** The operating-group heads have long tenures: Miller co-founded Trapeze, CSU's first acquisition (1995), and has been with the group over 30 years; Baksh joined in 2003; Billowits (Chair) ran Vela and was CFO; Bender (Harris), McKay (Vela) and van Poelje (Topicus) run groups producing their own deal flow. Daan Dijkhuizen became Chief AI Officer on 21 October 2026 (`csi-ai-2026`). The succession was internal and the method is codified in the groups, which reduces key-person risk; whether head office keeps the hurdle discipline without Leonard is the open question. **Insider dealings.** Insider purchases are mostly the mandatory bonus and fee purchases; SEDI filings were not reviewed, so open-market sales and their timing are unknown. ## Capital allocation principles Governing objective. Leonard's 2021 letter recasts head office as a "good steward of investors' capital" that would rather accept a lower return on a larger sum than return cash; the bonus rewards return on all invested equity plus growth. Management behaves as if long-term value per share is the goal (no issuance, no options), with a stated preference for deploying everything over distributing. 1. **Zero-based allocation: partly.** Capital does move: groups with very high ROIC are told to keep and redeploy their cash, groups with poor returns send it to head office (2015 letter p.8), and head office added large-deal and non-VMS practices in 2021. But acquired businesses are almost never sold ("buy and hold forever"), so capital in a weak unit stays there. 2. **Funding strategies, not projects: mixed.** Deals are approved against hurdles one by one, but within a strategy (small VMS at traditional hurdles, large VMS at lower ones, a separate non-VMS circle). Organic "initiatives" are tracked and killed if their hypotheses fail (2015 letter pp.5-6). 3. **Capital has a cost: yes.** No share compensation at all, so there is no disguised financing; debt is mostly non-recourse at subsidiaries; the IRGA and Topicus/Lumine structures are explicit claims. The hurdle-magnetism finding shows management understands that a lower hurdle costs more than the marginal deal. 4. **Zero tolerance for bad growth: weak.** There are almost no divestitures; the response to Altera's decline is restructuring, not exit. Shrinking businesses are kept if they produce cash, which is consistent with the IRR focus (2015 letter: the same IRR from a declining high-margin business is as good as from a grower), but it means capital is not actively recycled out of declining units. 5. **Price versus value: limited.** CSU has never bought back shares, including in early 2026 at C$2,196-2,500, about 15-17x today's estimate of normalized common owner cash (C$145 a share, stage 7), and has never issued CSU shares. It paid three special dividends when it couldn't deploy cash, stopped them in 2021, and Topicus paid a special dividend in 2025. Price discipline is shown in acquisitions (walking away from auctions), not in its own stock. ## Integrity and governance Concrete evidence: - Leonard waived all salary and bonus from 2023 to 2025; Miller waived his from 1 January 2026 (MIC p.14). The board used no discretion on 2025 bonuses. - Seven of nine nominees are independent; the audit and compensation committees are independent; a majority-voting policy and an advisory vote on pay apply (MIC pp.20-30). - Single class of common shares, no controlling holder. CSU itself controls Topicus and Lumine with super-voting shares (50.1% of votes against about 31% and 61% economic stakes); that is an agency question for their minorities, not for CSU's holders. - There is no formal anti-hedging policy for officers' and directors' shares (MIC p.13), a gap given the reliance on share ownership. - Related parties: the IRGA with the Joday Group is the main one, set by formula and disclosed (`csi-2021-ownership`). Agency costs (Mauboussin and Callahan pp.47-48): - **Control of resources: moderate.** Pay grows with size: the growth term of the bonus, plus base salaries that rise as teams manage more businesses (2015 letter p.8). The 5% threshold is low. The offsets are the ROIC term on unamortized capital and the 75% share purchase. Measured outcome: the 2021-2025 aggregate deal yield (15.1%) is well above the cost of capital, so empire-building has not yet shown up in aggregate; the large-deal returns (Altera 12.7%) are where it would. - **Risk tolerance: mild conservatism.** Heavily invested managers keep parent leverage low (no CSI facility drawn at December 2025) and hold large cash balances (3,089), which lowers returns slightly but protects against permanent loss. - **Time horizon: well aligned.** Bonus shares are escrowed for about four years; ROIC is measured on a running tally of capital, so early-year gains from under-investing are not rewarded for long. Buybacks don't offset share compensation because there is neither. **How the valuation treats agency cost:** no explicit charge. The model uses zero dilution, the measured aggregate deal yield in the base, and an 8% yield (12.5x) in the downside for a world where size-driven deployment pushes returns down. Head-office pay is immaterial against owner cash (about 2,200). # Capital allocation and ownership USD millions. Cash flow statements: FY2021 and FY2022 columns from the FY2022 report (`csi-fy2022`, PDF p.42), FY2023 (`csi-fy2023`, PDF p.41), FY2024 and FY2025 (`csi-fy2025`, PDF p.44). Facts are `csu*` in `financials/facts.json`; arithmetic is in `valuation/calculations.json` (prefixes `rates`, `reinv`, `uoc`, `agg`). ## Sources and uses of cash | | FY21 | FY22 | FY23 | FY24 | FY25 | Total | | --- | ---: | ---: | ---: | ---: | ---: | ---: | | Operating cash before working capital (after cash tax) | 1,255 | 1,357 | 1,815 | 2,241 | 2,738 | 9,406 | | Working capital | 45 | (60) | (36) | (45) | (6) | (102) | | **Cash from operations** | **1,300** | **1,297** | **1,779** | **2,196** | **2,732** | **9,304** | | Leases (interest and principal) | (92) | (105) | (120) | (132) | (147) | (596) | | Capex | (29) | (41) | (42) | (67) | (68) | (247) | | **Unlevered owner cash** (`uoc*`) | **1,179** | **1,151** | **1,617** | **1,997** | **2,517** | **8,461** | | Interest paid on debt | (40) | (74) | (133) | (178) | (196) | (621) | | Interest and dividends received | 5 | 5 | 4 | 33 | 56 | 103 | | Acquisitions: closing cash | (1,224) | (1,633) | (1,609) | (1,347) | (1,227) | (7,040) | | Acquisitions: holdbacks and settlements | (145) | (149) | (238) | (336) | (286) | (1,154) | | Cash acquired with businesses | 153 | 216 | 152 | 164 | 173 | 858 | | Financial investments, net (Asseco in FY25) | 2 | (91) | 94 | (15) | (529) | (539) | | Common dividends | (85) | (85) | (85) | (85) | (85) | (425) | | Minorities, preferred and Joday distributions | (22) | (30) | — | (175) | (3) | (230) | | Net borrowing | 200 | 787 | 657 | 727 | 570 | 2,941 | | Other financing (fees, other) | (3) | (10) | (1) | (41) | 15 | (40) | | Currency on cash | (16) | (39) | 17 | (48) | 102 | 16 | | **Change in cash** | **5** | **48** | **473** | **696** | **1,109** | **2,331** | Cash went from 758 (December 2020) to 3,089 (December 2025), a change of 2,331; the lines sum to within rounding (4, 48, 475, 696 and 1,107 by year). Cash taxes inside operating cash were 2,010 over the window. Non-cash acquisition consideration not in the table: Topicus.com B.V. preference units (136, 2021) and WideOrbit consideration beyond the parent's cash (323, 2023). What the table shows about priorities. Acquisitions took 7,336 net of acquired cash, 87% of the 8,461 of unlevered owner cash (92% including the non-cash consideration). Dividends took 5%, flat at USD4 a share in every year including FY2020. There were no buybacks and no share issuance. Net borrowing of 2,941, mostly non-recourse at subsidiaries plus the 2024 senior notes, roughly funded the 2,331 cash build, so CSU's own cash funded its deals and the debt mostly sits as cash (1,172 of it in subsidiaries whose distributions to the parent are restricted). The 2025 Asseco purchase (580) is the first large non-VMS use. ## Investment rates per sales change Window FY2021-FY2025: revenue rose from 3,969 (FY2020) to 11,623, a change of 7,654 (`ratesSalesChange`). | Rate per $1 of revenue change | FY21 | FY22 | FY23 | FY24 | FY25 | FY21-25 | | --- | ---: | ---: | ---: | ---: | ---: | ---: | | Revenue change | 1,137 | 1,516 | 1,785 | 1,659 | 1,557 | 7,654 | | Incremental M&A (net cash; window incl. non-cash) | 1.07 | 1.03 | 0.95 | 0.92 | 0.86 | **1.02** (`ratesMaRate`) | | Internally generated intangibles (R&D increase) | | | | | | **0.14** upper bound (`ratesRdRate`) | | Fixed capital (capex − est. PP&E depreciation) | | | | | | **(0.004)** (`ratesFixedRate`) | | Working capital | (0.04) | 0.04 | 0.02 | 0.03 | 0.004 | **0.013** (`ratesWcRate`) | Definitions and limits: - **M&A**: closing cash plus holdbacks paid less cash acquired, plus documented non-cash consideration in the window total (7,795, `ratesAcqAllIn`). Including the non-cash items, FY21 is 1.19 and FY23 1.13. There were no material divestitures. - **Intangibles**: CSU capitalizes no development. R&D staff rose from 579 to 1,615 (1,036), but R&D stayed between 13.9% and 14.6% of revenue every year, so the increase mostly follows acquired revenue rather than growth spending. 0.14 is an upper bound on growth R&D. - **Fixed capital**: capex 247 against an estimated PP&E depreciation of 274 (total depreciation 809 less lease principal 535, which approximates right-of-use depreciation). Leases are excluded from this rate. - **Working capital**: the cash-flow-statement change (102 over five years). Acquired businesses bring negative working capital (deferred revenue), which is inside the purchase price. Where CSU really invests: 99% of growth investment (7,795 of 7,870, `reinvGrowthCapital`) went into acquisitions. Organic growth needs no capital at all; customers prepay. The M&A rate per dollar of sales change drifted down from 1.07 to 0.86 (net cash basis) across the window because the Altera-type large deals were cheap per dollar of revenue and because acquired revenue grew after purchase. Organic growth (cumulative factor 1.18, `aggOrganicFactor`) explains about 714 of the 7,654 revenue increase, so acquisitions cost about 1.1x each dollar of revenue they brought. No peer rates were computed; Roper's and Tyler's statements were not read for this stage. Inflection points and who was in charge. Leonard was President and Baksh CFO throughout the window. The February 2021 letter lowered hurdles for large deals: Altera (May 2022, 727) and the 2022 borrowing surge (787) followed. 2023 added WideOrbit and Optimal Blue (1,209 together) and a C$213m debenture tap. 2024 replaced the drawn bank facility with 1,000 of senior notes. Deployment then stalled (1,847 → 1,683 → 1,513 from 2023 to 2025) while cash grew, and 580 went to Asseco in 2025. After Miller took over (September 2025), twelve-month deployment reached a record 2,501 to June 2026 (`deploymentTtm`), including DerbySoft and Synchronoss. ## Reinvestment and returns Measured, consolidated, unlevered (operating cash after tax, leases and capex, before debt interest, IRGA and minorities), which matches consolidated acquisition spending however it was financed: | | FY2020 | FY2025 | Change | CAGR | | --- | ---: | ---: | ---: | ---: | | Unlevered owner cash | 1,087 | 2,517 | 1,430 | 18.3% | | FCFA2S (common, after interest, IRGA and NCI) | 989 | 1,683 | 694 | 11.2% | | FCFA2S per share (USD) | 46.68 | 79.40 | | 11.2% | | Revenue | 3,969 | 11,623 | 7,654 | 24.0% | - **Reinvestment rate FY21-25**: growth investment 7,870 ÷ matching unlevered owner cash 8,461 = **93%** (`reinvRate`). - **Lagged incremental return**: the 1,430 cash increase against acquisition capital deployed 2020-2025 with the edge years half-weighted (8,156, `aggMatchedCapital`) = 17.5% (`reinvGrossIncrementalReturn`). Removing what the 2020 business would earn after its own organic growth (1,087 × 1.18 = 1,283) leaves 1,234 earned by acquisitions: **15.1%** after tax, a 6.61x cash multiple (`aggAchievedYield`). It is a five-year rolling ROIIC, already lagged by the half-year weighting. - **Identity check**: organic growth on the base (about 3.4% a year) plus reinvestment × deal return (0.93 × 15.1% = 14.0%) gives about 17.4%, against 18.3% measured. The identity holds within a point, so the definitions are consistent. - **Operating leverage versus capital efficiency**: the EBITA margin fell from 28.4% to 26.5% and the incremental margin over the window was 25.5%. None of the growth came from operating leverage; all of it came from deploying capital at about 15%, plus a modest organic contribution. - **The mix and each use's return**: acquisitions about 99% of growth investment at 15.1%; organic growth required no net capital (working capital 102, net capex −27) and added about 196 of cash, an effectively unlimited return but a small amount; payouts were dividends (5%) and minority distributions (3%). **Where the consolidated growth went.** Common owners received 694 of the 1,430 increase. The rest: the IRGA revaluation charge grew by 375 (65 → 440; 252 of FY2025's charge was the mark on the Asseco and Sygnity stakes, MD&A p.13), minorities' share by 243 (0 → 243, after the Topicus and Lumine spin-outs), and debt interest by 164, partly offset by 55 more interest received. So per-share owner cash compounded at 11%, not 18%. The model's baseline is already common-owner cash after these claims; but it means the consolidated 15.1% deal yield overstates what common shareholders keep from deals that minorities co-fund or that are debt-funded, and that the Topicus claim grows with Topicus's success. **Biases in the 15.1%.** Upward: acquired businesses' post-purchase organic growth and margin maturation are counted (the model holds new cohorts flat), and the Topicus founders' ordinary units are not valued. Downward: capital is gross of the 858 of cash acquired with the businesses; if that cash were all excess the yield would be 16.8% (`reinvYieldNetOfAcquiredCash`). CSU itself compares operating cash with "net capital deployed". ## Acquisition record | Year | Deal(s) | All-in price | Revenue at purchase (annual) | Profit at purchase | Later evidence | | --- | --- | ---: | ---: | --- | --- | | 2021 | Topicus.com B.V. (merged with TSS into Topicus) and others | 1,337 cash used + 136 preference units | n/d | n/d | Topicus FY2025 revenue 1,746, recurring organic +7% | | 2022 | Altera (Allscripts hospitals and large practices) | 727 | 846 | net income 0 (May-Dec 2022) | FCFA2S 71 (8 months), 94 (FY23), 92 (FY25); revenue 712 in FY25 | | 2022 | Other deals | 1,148 consideration | not material individually | n/d | — | | 2023 | WideOrbit (via Lumine) | 504 (273 cash, 222 special shares, holdbacks) | 172 | net income 1 (10 months) | in Lumine; not separately disclosed | | 2023 | Optimal Blue | 705 | 187 | net income 0 (3.5 months) | not separately disclosed | | 2023 | Other deals | 1,400 consideration | n/d | n/d | — | | 2024 | Many small and mid deals | 1,683 cash used (incl. holdbacks for earlier deals) | n/d | n/d | 2,069 of intangibles recognized | | 2025 | Cipal Schaubroeck (Belgian local government) | 258 | 122 | net loss 6 (7 months) | — | | 2025 | Other deals | 1,321 consideration | not material | n/d | — | | H1 2026 | Synchronoss (via Lumine), DerbySoft and others | 309 cash + 999 other consideration | Synchronoss ~169 | Synchronoss post-close losses | `lumine-q2-2026-fs`, `csi-q2-2026` | | 9M 2026 (final) | Synchronoss 309, DerbySoft 391 (91.8%), additional deals 1,605 total consideration (1,241 cash, 313 holdbacks, 51 contingent) | 2,305 | Synchronoss 172 run rate, DerbySoft about 133 (pro forma) | DerbySoft −6 and Synchronoss −2 net loss since purchase; bargain gains 17 | `csi-q3-2026-fs` pp.10–15 | Aggregate FY2021-FY2025: about 7,795 of all-in net capital bought businesses that, by 2025, earned about 1,234 of unlevered owner cash (15.1% on the matched 8,156; 6.61x). Revenue acquired was roughly 6,940, so the price was about 1.1x revenue and the acquired businesses' cash margin about 18%, below the 2025 group margin of 21.7%. The large carve-outs were bought at near-zero profit, so the measured yield is mostly the result of CSU's post-purchase cost discipline, not of buying profits cheaply. Observed deployment pace (cash used, including holdbacks): 582 (2020), 1,337, 1,782, 1,847, 1,683, 1,513 (2025), 2,501 in the twelve months to June 2026 and 2,733 in the twelve months to September 2026 (2,471 net of acquired cash; 1,861 without Synchronoss and DerbySoft). Deployment grew about 15% a year from 2021 to mid-2026 but stalled from 2023 to 2025. Net of acquired cash the 2021-2025 average is 1,461 (best 2023 +16%, worst 2021 −19%, 2020 −67%; longest quiet run two years), and since 4 November 2026 the model states a yearly deal budget anchored on that average rather than on the record twelve months (`deploy*` steps; register entry of that date). ## Model funding and ownership bridge The model begins with cash available to common shareholders after financing, taxes, ongoing software spending, leases and capex. It pays the USD4 annual dividend and older acquisition claims before funding new deals. Full deployment is the base and upside assumption; the downside deploys half the surplus and retains the balance. No new bank leverage finances future purchases. Annual purchases follow a stated yearly budget (base USD1,651m net in year 1; the upside starts at the September TTM, 2,471 net), deals earn half a year of cash in the purchase year, and cash the cap can't place is distributed that year rather than hoarded, following CSU's stated alternatives for FCFA2S and the 2025 Topicus special dividend. Topicus accounting ownership is 60.65%, including the deemed future purchase of Joday's units; that is not present direct economic ownership (`csi-2021-ownership`, pp.2-3). The model reverses the non-cash IRGA remeasurement from annual cash and separately deducts the full 1,295 economic claim (30 September 2026), grown at an estimated 10% a year. Removing both charge and liability would create value common shareholders do not own. The [9 November review](../updates/2026-11-09-daily-review.md) distinguishes settlement form from economic cost; signed revised terms remain unresolved. The current engine deducts this claim at exit, not as a cash payment in annual acquisition funding. September 2026 group cash was 2,693 (debt 3,999, of which 2,527 without recourse to CSI). Minority cash, an operating reserve, the post-September closing commitments (526) and the October dividend reduce modeled deployable cash to about 1,634. 1,025 of group cash sat in subsidiaries with restricted upstream distributions, so a pooled look-through forecast doesn't prove every dollar can move. Asseco and other financial investments stay outside the acquisition budget: their dividends are removed from operating cash and their values enter terminal equity once. Bank principal is assumed refinanced; accrued interest stays charged. # Thesis and valuation Current assessment: 9 November 2026, using the September trailing bridge and 30 September balances. The financial update was completed on 6 November. The results call is now reviewed with stated source limits; financial paths and numerical scenario weights remain unchanged. Mission-critical vertical software earns recurring maintenance cash, while decentralized purchases add businesses across many specialist markets. Customer replacement costs support durability, but Altera and acquisition competition show that neither recurring revenue nor purchase returns are guaranteed. CSU common owners also share consolidated value with subsidiary minorities, redemption claims and creditors. ## Current valuation Saved quote CAD 3392.29, market time 2026-11-09T21:00:00.000Z; valuation date 9 November 2026 and original model exits. These are forward estimates. | Horizon | Bear | Base | Bull | Weighted cash-flow IRR | | --- | ---: | ---: | ---: | ---: | | 5 years | -1.78% | 13.02% | 22.65% | 13.05% | | 10 years | 0.82% | 12.55% | 18.94% | 11.96% | The parent now reports through September. Synchronoss, DerbySoft, other H1 purchases and July Lumine businesses add only their missing consolidation months; the late-FY25 cohort is wholly inside the new trailing window and has been removed. Other Q3 target sales and post-September commitments remain separately labeled estimates. September cash and claims already include July closings, so those payments are not deducted again. Gross consolidated sales are paired with common cash at 61.4% Lumine ownership where known; unsegmented pools retain the explicit allocation proxy. The base’s gain from the filing roll depends mainly on normalizing the USD179m TTM working-capital outflow. The split between payment timing and acquired liabilities is undisclosed. Capping the add-back at the previous USD70m level gives about 11.45% five-year base IRR at C$3,511.69 on 6 November; this is a sensitivity, not the current quote or an observed recurring cost. [Filing roll and attribution](../updates/2026-11-06-q3-roll.md). ## Pre-mortem Synchronoss and July sales annualization are flat-quarter estimates; mature target cash margins are not reported owner cash. Undisclosed pool sales use matched net or gross capital/revenue prices with explicit timing assumptions. Open commitments are not labelled completed. Restricted subsidiary cash, the IRGA claim, customer attrition and target returns remain genuine risks. The central estimate can fail even when a useful product or good acquisition history survives. Customer demand, supplier pricing, the capital needed for delivery and ownership/financing claims must all work together. The adverse case is a representative failure mechanism, not the largest possible loss. No new report or positive share-price move alone resolves these risks. ## Exit multiple The existing source-supported scenario capitalization and year-by-year deal budgets are retained. This bridge correction does not make more acquisition supply or justify a larger multiple. The independent debate’s finite-duration and ownership/funding attacks still constrain conviction; a claimed long runway is an estimate rather than a measured future duration. The revenue bridge does not supply independent proof of terminal value or remove the survival risks identified in the written debate. Shorter duration, poorer cash conversion and higher required returns remain relevant stresses. Historical source reproduction and current underwriting remain distinct. ## Bear case Synchronoss and July sales annualization are flat-quarter estimates; mature target cash margins are not reported owner cash. Undisclosed pool sales use matched net or gross capital/revenue prices with explicit timing assumptions. Open commitments are not labelled completed. Restricted subsidiary cash, the IRGA claim, customer attrition and target returns remain genuine risks. The existing adverse operating and capital-allocation paths remain the coherent modeled stress. Matching sales to cash prevents a missing denominator from hiding those assumptions; it does not turn a proxy margin or assumed refinancing into a reported fact. Debt, deferred claims, operating liquidity and distributions remain funded in the same cash ledger. ## Variant view Recurring maintenance, decentralized acquisition execution and the measured purchase cash-yield proxy support the central mechanism. September TTM deployment is USD2,471m net; excluding Synchronoss and DerbySoft gives USD1,861m against the base year-1 budget of USD1,651m. That supports current executable capacity but does not measure ten years of seller availability. The explicit outside-view exception retains the modest deal budget, distributes unplaced cash and records missing target counts and win rates. Its acceptance is a judgment, not statistical proof. At C$3,392.29 the five-year weighted estimate is 13.05%; the increase comes from a lower entry quote, not a better operating forecast. The five-year base is 13.02%, 12.29% if the working-capital add-back is capped at USD70m, and 11.81% without that add-back. These are conditional sensitivities, not reassessed scenario distributions; the apparent margin of safety depends on a normalization the call does not quantify. [9 November decision](../updates/2026-11-09-daily-review.md). Acquisition price, mature target margins, customer retention and the 20x base exit remain material judgments. Q3 recurring organic growth outside Altera held at 4%, while Altera recurring fell 8% and newly acquired high-attrition healthcare businesses limit any group-level reassurance. Future evidence needs to establish cash collection and acquired economics, not merely recurring-revenue labels or a larger deal budget. The [independent claim judgment](12-debate.md) and [Q3 probability review](11-scenario-probabilities.md) remain the current risk record. Earlier investor comparisons remain in the historical reference; no new source-author forecast or market-consensus survey is inferred. ## Why the opportunity exists The following observations retain the measurement dates in their cited sources; no new holder, liquidity or short-interest survey is claimed. Evidence (`csu-price-history-2026-11-03`; MIC p.11): - **Size and liquidity.** Market capitalization C$72.8bn (21.19m shares × C$3,433.89). October 2026 volume was 1.29m shares, about 59,000 a day or C$200m of daily trading value. No holder has more than 10%; directors and officers listed in the circular own about 0.3m shares, so the float is about 20.9m and turns over in roughly 360 trading days. - **Coverage, short interest, retail ownership, index membership.** Not retrieved; CSU is a large Canadian technology stock widely discussed in investor media, so neglect is unlikely to explain any mispricing. - **What drove sellers.** The price peaked near C$5,300 in May 2025. Leonard's resignation on 25 September 2025 coincided with a September close of C$3,778 (from C$4,550) on higher volume. The steepest fall came in January 2026 (close C$2,513 from C$3,301; record volume of 2.16m shares), when investor commentary attributed the selling of vertical-software stocks to fears of AI substitution (secondary sources, not verified against fund flows). Q2 2026 showed 1% organic growth and Altera's 19% recurring decline. The price has since recovered to about C$3,400-3,450. So the discount is not neglect: it is a widely shared fear about AI and succession, partly justified by Altera and by the halving of returns on new capital since the large-deal practice began. ## What would prove us wrong Q3 2026 (6 November) was checked against this list: recurring growth outside Altera held at 4% and net deal cash rose, so neither customer erosion nor a deployment stall is yet visible (`updates/2026-11-06-q3-roll.md`). Still open: recurring growth outside Altera below 2% for two quarters, FY2026 net deployment below the base budget, or acquired cohorts that stay loss-making. Compare future observations with the same ownership, cost and reporting periods used in this model. Preserve earlier frozen checks and score them as their reports arrive; do not edit their forecasts to match the result. Unavailable deal disclosure needs a bounded estimate and a revision trigger, not a claim of observed target profitability. [Original revenue-bridge closure](../updates/2026-11-05-revenue-bridge-closure.md) and [September filing roll](../updates/2026-11-06-q3-roll.md). The earlier initial completion audit’s three revenue gaps are now closed with reported observations and explicit estimates; the remaining limits above are ordinary underwriting uncertainty, not missing forecast outputs. # Risks and disconfirming evidence USD millions. ## Paths to permanent loss 1. **Installed-base erosion.** Recurring maintenance carries about 85% of EBITA (stage 3). If AI-assisted development and data migration lower switching costs across many niches at once, renewals and price increases fall together and the decline is permanent: migrated customers don't come back. Altera (revenue −16% since purchase; recurring −19% in Q2 2026 and −8% in Q3; total organic −18% over nine months) shows what this looks like in one business. Persistent erosion outside Altera, falling retention or rising development needs would break the cash-growth assumption. [Q2 MD&A, pp.5 and 16](https://www.csisoftware.com/wp-content/uploads/2026/08/Q2-2026-Shareholder-Report.pdf); [Q3 MD&A, pp.5 and 17](https://www.csisoftware.com/wp-content/uploads/2026/11/CSI-MDA-Q3-2026-Final.pdf). 2. **Overpaying at scale.** Returns on new capital have roughly halved since 2015 (38% bonus-basis ROIC then; 15.1% measured on 2021-2025 deals now). Lower hurdles for large deals are "magnetic" by CSU's own evidence (2015 letter p.6). If the bonus's growth term and the new President's mandate push deployment to auction prices, new capital can earn less than its cost while the reported deal count keeps rising. Historical corporate ROIC, a deal IRR hurdle and a cash purchase yield measure different things. 3. **Claims ahead of common.** The IRGA claim (1,295 in September 2026, 741 of it classed current) grows with Topicus's value and investment marks; minorities took 243 of FY2025's pre-NCI FCFA2S and 208 in 9M 2026; 1,025 of September cash sat in subsidiaries with restricted distributions. Puts, contingent payouts and refinancing can require cash sooner than modeled. The model assumes bank refinancing and does not simulate a covenant crisis or recapitalization. Legacy acquisition tax benefits can also expire. 4. **Paying a high multiple for a slower compounder.** At 23x normalized cash, a fall to the trough multiple (15x) with flat earnings would cost about a third of the price. The downside reaches −2.5% a year (C$3,511.69) only because five years of retained cash and dividends offset most of that. ## Unverified inputs Shared-engine sensitivities at C$3,392.29 on9November2026; base13.02%, fixed original exits. Each switch changes only the named assumption, without assigning new scenario probabilities. | Input | Current model | Uncertainty | Isolated five-year base sensitivity | | --- | --- | --- | --- | | Exit multiple | 20x | Durability and post-exit reinvestment judgment | 15x: 7.07%; 25x: 17.92% | | Yearly deal budget | USD1,651m in year1, then stated paths | No measured future seller/win-rate funnel | 0.75x: 11.71%; 1.25x: 14.25% | | Acquired owner-cash margin | 15% then16.5%, reaching17.8% | Mature purchase-yield proxy, not reported cohort owner cash | 12% throughout: 10.34% | | Price per dollar of acquired revenue | 1.17x | Matched historical net purchase price; future competition remains | 1.5x: 11.22% | | Organic/acquired revenue growth | 1.5%, then2%, then3% | Customer durability, product mix and eventual Altera stabilization | 1 point lower each year: 12.19% | | Owner-cash margin on organic revenue | 17.48% flat | Common-owner cash on consolidated sales; mix and recurring investment estimates | 1 point lower: 12.03% | | Normalized owner cash | USD2,309m | USD179m working-capital add-back remains unverified quantitatively | Add-back capped70m: 12.29%; no add-back: 11.81% | | Missing funded/announced contribution | About USD243m mature cash | Missing periods, purchase-yield and integration estimates | No new reported margin inferred from the call | | IRGA economic claim growth | 10% annually | Mixed underlying claims and unresolved revised settlement allocation | 15% growth: 12.89% | | Deployable cash | About USD1,634m | Restricted subsidiary cash and operating-liquidity estimates | Not all consolidated cash is fungible | ## Disconfirming evidence to watch **Customer economics.** FX-adjusted group organic growth was 2% in Q3 2026 and 1% over nine months; recurring 4% in Q3 (2% in Q2). Outside Altera recurring grew 4% in both quarters; Altera recurring fell 8% in Q3 after −19% in Q2, and management expects sizable client departures to run until the end of 2027. The US healthcare group outside Altera fell 10% in Q3, −3% without two newly bought businesses whose high attrition CSU says it priced in. Licenses fell 17% and services 5%. The steady +4% outside Altera supports the base; buying businesses that are known to shrink is a new test of the purchase yield. **AI and competition.** Integration, domain rules and migration costs support incumbents, but cheaper development can lower entry barriers and change pricing expectations. The new Chief AI Officer is a response, not proof the risk is resolved. More required product spending would reduce the cash margin of existing and acquired businesses. Pricing AI substitution into acquisition screens raises the effective hurdle on AI-exposed targets, which was offered as one explanation for slower deployment alongside fewer qualifying sellers. Q3 2026 showed no group-level slowdown to explain: nine-month net deal cash was USD1,986m against 855 a year earlier (`08-underwriting-plan.md`, deployment test). **Deployment at scale.** A larger capital base can exhaust small attractive targets, move CSU into auctions or accept lower returns. The base purchase yield is the group's measured 2020-2025 aggregate (restated net of acquired cash and at entry), earned at USD1.3-2.5bn a year of deployment; faster scaling at the same yield is the upside, not the base. Failed margin convergence, higher integration costs and idle cash reduce compounding. **Industry check, updated8 November 2026.** Roper reported 7% organic growth and continuing software bolt-ons on 22 October 2026, which supports ongoing demand and acquisition competition in some specialist markets, but its mix differs from CSU's and it does not override CSU's weaker growth. [Roper Q3 release](https://investors.ropertech.com/news-releases/news-release-details/roper-technologies-announces-third-quarter-financial-results-2). Recovered Tyler's original28 October [SEC earnings exhibit](https://www.sec.gov/Archives/edgar/data/860731/000086073126000056/a991earningsrelease-9302026.htm), including cash and non-GAAP reconciliations. It reports11% recurring growth but also a Texas payments-contract wind-down; SaaS migration and transaction growth are different drivers. Quarterly FCF includes a large working-capital benefit and adds back share compensation in CFO. It is not CSU owner cash, a measured acquisition yield or evidence resolving CSU's working-capital normalization. No Tyler input enters the forecast. Earlier source-access limits remain in their dated records. Coverage: issuer filings, the 2026 circular, management letters and that peer release, not every vertical or customer. No interviews or fieldwork. Investment marks are at 30 September 2026; post-September deals are a disclosed pool, not completed facts. The observations that would refute the thesis, with dates, are listed in `research/05-thesis.md` under "What would prove us wrong". ## Surviving debate attacks — 5 November 2026 Low to moderate; recurring product economics supported, current ownership/funding and full revenue bridge incomplete. Reconstructed the independent isolated stress at the same quote/date/horizons; do not treat stress size as a probability or a measured current fact. Discriminating evidence named then (Q3 2026) has arrived; see 12-debate.md for the outcome. The [complete debate](12-debate.md) sizes these risks using the shared forecast; a sensitivity is neither a measured fact nor a calibrated probability. ## Revenue bridge closure and remaining uncertainty — 5 November 2026 The previously missing sales/cash perimeter is now modeled. Synchronoss and July sales annualization are flat-quarter estimates; mature target cash margins are not reported owner cash. Undisclosed pool sales use matched net or gross capital/revenue prices with explicit timing assumptions. Open commitments are not labelled completed. Restricted subsidiary cash, the IRGA claim, customer attrition and target returns remain genuine risks. Q3 2026 has since replaced the June bridge (6 November; `07-assumption-register.md`). This closes the missing forecast output rather than claiming that the target economics or refinancing outcomes are certain. Earlier source-gap statements in the dated debate describe its initial information set; current inputs are in 07-assumption-register.md. # Current assumption register Evidence assessment: 9 November 2026; financial bridge: 6 November Q3 roll. Reported facts, issuer pro-forma estimates, normalization and forward judgments remain separately classified. The [9 November call review](../updates/2026-11-09-daily-review.md) retains the quantitative bridge and weights. IRGA is an economic exit claim, not assumed all-cash funding; working-capital neutrality remains an unverified normalization. Source limits and fixed-price impact checks are recorded there. ## Reported starting bridge The bridge now starts from the twelve months to 30 September 2026 (FY2025 + 9M 2026 − 9M 2025) and 30 September balances. Reported TTM FCFA2S is USD1,946m; adding back the USD309m noncash IRGA charge, neutralizing the USD179m working-capital outflow, and normalizing interest and investment income gives USD2,309m of common-owner operating cash (June: 2,197). Missing-month windows shrink: Synchronoss, DerbySoft and the July Lumine businesses now have owned Q3 sales in the base; other H1 deals lose half their gap, and the late-FY25 cohort is removed because it now lies wholly inside the TTM. Gross consolidated sales are paired with common cash at the specific 61.4% Lumine ownership where known and the allocation estimate only for unsegmented pools. The largest judgment in the roll is the working-capital add-back: capping the add-back at the June TTM outflow (70) would lower the five-year base by about 0.7 points. | Input | Value | Classification | Definition/evidence | | --- | ---: | --- | --- | | equityCashEarnings | 3237.859 | underwriting estimate | Reported TTM FCFA2S plus the noncash IRGA revaluation; normalize working capital, remove cash/investment income, charge accrued economic debt cost and adjust the identified transaction cost. The full IRGA stock is separately valued at exit. Reported cash already excludes NCI; the allocation fraction is applied only to otherwise unallocated adjustments. | | shares | 21.192 | reported | 21,191,530 basic and diluted common shares at 30 September and 6 November 2026, unchanged from December. No second preferred or super-voting economic class is added. | | openingCash | 2292.339 | underwriting estimate | 30 September cash less external minority cash, a maximum Adapt minority reserve, the post-September closing commitments, the October dividend and ten days of operating liquidity. This is a look-through funding estimate; cash inside subsidiaries is not freely transferable to parent. | | revenue | 18518.610 | reported | Trailing twelve months to 30 September 2026 consolidated revenue (FY2025 + 9M 2026 − 9M 2025 = USD13,204m) in CAD at 1.4025; calculation revenueTtmCad. | ## Known and committed cohort bridge Revenue is consolidated gross; cash is on the model’s common-owner or cash-NOPAT perimeter. A reported sales observation is not a reported cash margin. Missing months exclude sales already in the starting baseline. Capital already paid or reserved is not purchased again. | Cohort | Missing annual sales | Missing/mature cash | Timing and basis | | --- | ---: | ---: | --- | | Synchronoss missing owned contribution | 91.306 | 11.818 | Latest complete owned quarter (Q3, 43.1) times four, less the 107.4 already in the September TTM. Cash still ramps because applying a group margin is an integration estimate. | | DerbySoft missing contribution | 121.082 | 25.524 | Nine-month pro-forma sales (53 missing before 1 June plus 47 owned) annualized flat, less the 47 already in the September TTM. Seasonality and profit normalization remain estimates, so both sales and cash ramp. | | Other H1 funded acquisitions, missing months | 542.878 | 86.103 | Matched net capital/revenue price; an average April closing leaves half a year missing from the September TTM (was 75% of a year against June). Sales and cash are explicit pooled estimates, not new reported facts. | | Quortex and Imagine: completed July acquisitions | 240.341 | 40.858 | Q3 is the first full owned quarter; it is already in the September TTM, so only three quarters are added. Current purchase cash/deferred claims replace old estimates; mature cash margin ramps separately. No new purchase or facility draw. | | Post-September completed/open commitments | 746.168 | 118.345 | Post-September completed/open pool only. Paid closing amounts are reserved from September cash; deferred claims reserved separately. Mixed closing status requires a timing ramp. | | Other Q3 acquisitions, provisional missing months | 362.790 | 57.540 | Q3 additional acquisitions: 9M additional consideration (1,605) less the H1 cohort (999) and the separately modeled July Lumine businesses (231.5) = 374.5, which includes measurement-period revisions. Sales use the matched gross capital/sales price; an average mid-quarter close leaves 10.5 months missing. Provisional estimate, not a reported cohort. | ## Forward scenario mechanisms Mission-critical vertical software earns recurring maintenance cash, while decentralized purchases add businesses across many specialist markets. Customer replacement costs support durability, but Altera and acquisition competition show that neither recurring revenue nor purchase returns are guaranteed. CSU common owners also share consolidated value with subsidiary minorities, redemption claims and creditors. The existing source-supported scenario capitalization and year-by-year deal budgets are retained after the Q3 test: nine-month net deal cash (USD1,986m) already exceeds the base full-year budget, so the base is not too high on current evidence. The budget waits for FY2026 to enter the full-year average. The upside year 1 follows its binding to the September TTM (USD2,471m). This roll does not make more acquisition supply or justify a larger multiple. The independent debate’s finite-duration and ownership/funding attacks still constrain conviction; a claimed long runway is an estimate rather than a measured future duration. ### base Latest group organic growth is 1% after FX. Continue current ongoing spending and flat dividends; deploy surplus into cash-funded acquisitions at the group's achieved 2020–2025 aggregate acquisition cash yield. Exit at a chosen 20x operating equity cash multiple, with investments and all retained claims added/deducted separately. | Driver | Value/path | Reason and sources | | --- | --- | --- | | exitMultiple | 20.000 | Chosen terminal operating equity cash yields are 5%, 6.67% and 4%, giving 20x, 15x and 25x. They reflect alternative cash durability and reinvestment runway; they are independent judgments, not historical deal returns or current-price-reset formulas. Retained assets and IRGA/other claims are valued separately. | | dilutionRate | 0.000 | Zero further issuance follows the unchanged reported share count. Future DRIP, compensation or financing issuance would require a revision; no speculative buyback gains are included. | | reinvestmentRate | 1.000 | Base/upside full deployment follows the stated acquisition objective after dividends, operating reserve and old claims. The 50% stress retains cash if qualifying opportunities are scarce. Full deployment is not proof that the same purchase yield remains available at scale. | | organicGrowth, years 1–1 | 0.015 | Year 1 (Nov 2026-Oct 2027): revenue-weighted mix of the latest FX-adjusted organic rates (Q3 2026). CSU excluding Altera (96% of nine-month revenue) grew 3% in Q3 and 2% over nine months, with recurring revenue +4% in both; Altera (445 of 10,027, 4%) fell 17% in Q3 and 18% over nine months (recurring −8% and −11%; licenses −52%, services −30%). Mix: about 2.4 points from the rest less about 0.8 from Altera, roughly 1.6%. Group FX-adjusted organic growth over the last four quarters ran 2%, 1%, 1%, 2%. Q3 confirms 1.5%; no change. Sources: csi-q2-2026, csi-fy2025, csi-q3-2026-fs, csi-q3-2026-mda | | organicGrowth, years 2–3 | 0.020 | Years 2-3: Altera laps its upfront-contract comparison and declines about 5% a year (its FY2025 recurring was −3%); licenses and services shrink about 2% as their 2024-2025 declines slow. Recurring outside Altera stays at +4%. Mix gives about 2%. Q3 2026 check: management expects the sizable Altera client departures to run their course by the end of 2027 (inside year 2), so Altera may fall faster than 5% in year 2; at 4% of revenue a 10% fall would cost only about 0.2 points. Kept at 2%. Sources: csi-q2-2026, csi-fy2025, csi-q3-2026-mda | | organicGrowth, years 4–5 | 0.030 | Years 4-5: Altera stabilizes and licenses/services stop shrinking, so group organic growth returns to the recurring-led rate seen in FY2025 (3% FX-adjusted) and the 2021-2025 average (3.4% reported, about 3% FX-adjusted). Sources: csi-fy2021-mda, csi-fy2022, csi-fy2023, csi-fy2025 | | organicGrowth, years 6–10 | 0.030 | Years 6-10: the long-run rate of an installed base that renews and takes modest price increases: FY2025 3% FX-adjusted, 2021-2025 average about 3%. No AI disruption and no AI-driven acceleration is assumed; both are scenario questions. Sources: csi-fy2021-mda, csi-fy2022, csi-fy2023, csi-fy2025 | | organicMargin, years 1–10 | 0.175 | Flat at the starting owner-cash margin (17.48%), justified for all ten years: (1) the mix shift nets to about zero, because shrinking licenses carry a 63% EBITA margin and shrinking services 6%, while growing recurring revenue carries 30% (stage 3 unit economics; year-1 EBITA +1.7% against revenue +1.5%); (2) R&D has stayed at 13.9-14.6% of revenue every year 2020-2025; (3) Topicus grows faster with about 40% leaking to minorities and the IRGA, a drift of only about −0.02 points a year; (4) no AI productivity gain is evidenced yet. Sources: csi-fy2021-mda, csi-fy2022, csi-fy2023, csi-fy2025 | | priceToSales, years 1–10 | 1.169 | Measured price per dollar of acquired revenue, 1.17x (net capital at entry, 2021-2025 cohorts). Flat because the evidence shows no rising trend: acquisition capital per dollar of sales change fell from 1.07 (FY2021) to 0.86 (FY2025) while deal size grew, and the aggregate already includes the large deals (Altera 0.86x, Synchronoss 1.8x, Cipal Schaubroeck 2.1x, WideOrbit 2.9x, Optimal Blue 3.8x). Sources: csi-fy2022, csi-fy2023, csi-fy2025, lumine-q2-2026-fs, csi-q2-2026 | | acquiredMargin, years 1–1 | 0.150 | Year 1: only newly bought businesses contribute (half a year), and new cohorts start below the blended 17.8%: Altera earned 12.8% FCFA2S on revenue in its first eight months (71 / 556), Cipal Schaubroeck and Synchronoss reported post-close losses. 15% is the midpoint between Altera's first-year level and the blended aggregate. Sources: csi-fy2022, csi-fy2023, csi-fy2025, lumine-q2-2026-fs, csi-q2-2026 | | acquiredMargin, years 2–3 | 0.165 | Years 2-3: the model's acquired pool is still young (average age about 1-1.5 years against 2.3 years for the measured 2021-2025 pool), so margins are between first-year levels (Altera 12.8%, 11.6% in its first full year) and the blended 17.8%. Sources: csi-fy2022, csi-fy2023, csi-fy2025, lumine-q2-2026-fs, csi-q2-2026 | | acquiredMargin, years 4–10 | 0.178 | Years 4-10: the pool reaches the age mix of the measured 2021-2025 cohorts (about 2.3 years), so it earns their measured entry-basis margin, 17.8% (acquiredMarginMeasured). Sources: csi-fy2022, csi-fy2023, csi-fy2025, lumine-q2-2026-fs, csi-q2-2026 | | acquiredGrowth, years 1–1 | 0.015 | Year 1: acquired businesses grow like the group (CSU's organic measure includes them from purchase), with the same Altera and license drag. Sources: csi-q2-2026, csi-fy2025 | | acquiredGrowth, years 2–3 | 0.020 | Years 2-3: as organic growth; recent large carve-outs (Altera, Synchronoss) shrink while small VMS deals grow like the base. Sources: csi-q2-2026, csi-fy2025 | | acquiredGrowth, years 4–10 | 0.030 | Years 4-10: the group long-run organic rate, about 3% FX-adjusted 2021-2025, which includes acquired businesses after purchase. Sources: csi-fy2021-mda, csi-fy2022, csi-fy2023, csi-fy2025 | | deploymentCapacity, years 1–1 | 2315.112 | Year 1: the expected budget, with no lump placed (no deal beyond the reserved post-September pool has an evidenced date): the 2021-2025 through-cycle net average (USD1,461m) moved to 2027 at the measured ~3% trend = USD1,651m. Q3 2026 test: nine-month net deal cash was USD1,986m (1.20 times this full-year budget), Q3 alone 550 against 307 a year earlier; the twelve months to September excluding Synchronoss and DerbySoft were USD1,861m (1.13 times). Group deployment did not slow in 2026, so neither fewer sellers nor a higher AI hurdle is visible in the aggregate. Kept: the stage rule uses full-year through-cycle averages, not a peak, and a partial year with two large deals is not a full year. Revision trigger: when FY2026 is reported, the 2022-2026 window replaces 2021-2025; at about USD2.5bn net for 2026 the average would rise about 18%. Sources: csi-fy2021-mda, csi-fy2022, csi-fy2023, csi-fy2025, csi-q2-2026, csi-2021-capital, csi-q3-2026-fs, csi-q3-2026-mda | | deploymentCapacity, years 2–2 | 2384.600 | Year 2: the expected budget, the year-1 anchor grown at the measured ~3% trend (USD1,700m net). The 2021-2025 average already contains the record's weak years (2021 at −19%), and no deal with an evidenced date falls in this year, so no dip or lump is placed. The ~3% is the measured growth in qualifying deal supply: net deployment rose 3% a year from 2021 to 2025 while owner cash more than doubled (deployNetToOwnerCash* fell from 100% to 53%). Sources: csi-fy2021-mda, csi-fy2022, csi-fy2023, csi-fy2025, csi-q2-2026, csi-2021-capital | | deploymentCapacity, years 3–3 | 2456.100 | Year 3: the expected budget, the year-1 anchor grown at the measured ~3% trend (USD1,751m net). The 2021-2025 average already contains the record's weak years (2021 at −19%), and no deal with an evidenced date falls in this year, so no dip or lump is placed. The ~3% is the measured growth in qualifying deal supply: net deployment rose 3% a year from 2021 to 2025 while owner cash more than doubled (deployNetToOwnerCash* fell from 100% to 53%). Sources: csi-fy2021-mda, csi-fy2022, csi-fy2023, csi-fy2025, csi-q2-2026, csi-2021-capital | | deploymentCapacity, years 4–4 | 2529.800 | Year 4: the expected budget, the year-1 anchor grown at the measured ~3% trend (USD1,804m net). The 2021-2025 average already contains the record's weak years (2021 at −19%), and no deal with an evidenced date falls in this year, so no dip or lump is placed. The ~3% is the measured growth in qualifying deal supply: net deployment rose 3% a year from 2021 to 2025 while owner cash more than doubled (deployNetToOwnerCash* fell from 100% to 53%). Sources: csi-fy2021-mda, csi-fy2022, csi-fy2023, csi-fy2025, csi-q2-2026, csi-2021-capital | | deploymentCapacity, years 5–5 | 2605.700 | Year 5: the expected budget, the year-1 anchor grown at the measured ~3% trend (USD1,858m net). The 2021-2025 average already contains the record's weak years (2021 at −19%), and no deal with an evidenced date falls in this year, so no dip or lump is placed. The ~3% is the measured growth in qualifying deal supply: net deployment rose 3% a year from 2021 to 2025 while owner cash more than doubled (deployNetToOwnerCash* fell from 100% to 53%). Sources: csi-fy2021-mda, csi-fy2022, csi-fy2023, csi-fy2025, csi-q2-2026, csi-2021-capital | | deploymentCapacity, years 6–6 | 2683.800 | Year 6: the expected budget, the year-1 anchor grown at the measured ~3% trend (USD1,914m net). The 2021-2025 average already contains the record's weak years (2021 at −19%), and no deal with an evidenced date falls in this year, so no dip or lump is placed. The ~3% is the measured growth in qualifying deal supply: net deployment rose 3% a year from 2021 to 2025 while owner cash more than doubled (deployNetToOwnerCash* fell from 100% to 53%). Sources: csi-fy2021-mda, csi-fy2022, csi-fy2023, csi-fy2025, csi-q2-2026, csi-2021-capital | | deploymentCapacity, years 7–7 | 2764.400 | Year 7: the expected budget, the year-1 anchor grown at the measured ~3% trend (USD1,971m net). The 2021-2025 average already contains the record's weak years (2021 at −19%), and no deal with an evidenced date falls in this year, so no dip or lump is placed. The ~3% is the measured growth in qualifying deal supply: net deployment rose 3% a year from 2021 to 2025 while owner cash more than doubled (deployNetToOwnerCash* fell from 100% to 53%). Sources: csi-fy2021-mda, csi-fy2022, csi-fy2023, csi-fy2025, csi-q2-2026, csi-2021-capital | | deploymentCapacity, years 8–8 | 2847.300 | Year 8: the expected budget, the year-1 anchor grown at the measured ~3% trend (USD2,030m net). The 2021-2025 average already contains the record's weak years (2021 at −19%), and no deal with an evidenced date falls in this year, so no dip or lump is placed. The ~3% is the measured growth in qualifying deal supply: net deployment rose 3% a year from 2021 to 2025 while owner cash more than doubled (deployNetToOwnerCash* fell from 100% to 53%). Sources: csi-fy2021-mda, csi-fy2022, csi-fy2023, csi-fy2025, csi-q2-2026, csi-2021-capital | | deploymentCapacity, years 9–9 | 2932.700 | Year 9: the expected budget, the year-1 anchor grown at the measured ~3% trend (USD2,091m net). The 2021-2025 average already contains the record's weak years (2021 at −19%), and no deal with an evidenced date falls in this year, so no dip or lump is placed. The ~3% is the measured growth in qualifying deal supply: net deployment rose 3% a year from 2021 to 2025 while owner cash more than doubled (deployNetToOwnerCash* fell from 100% to 53%). Sources: csi-fy2021-mda, csi-fy2022, csi-fy2023, csi-fy2025, csi-q2-2026, csi-2021-capital | | deploymentCapacity, years 10–10 | 3020.700 | Year 10: the expected budget, the year-1 anchor grown at the measured ~3% trend (USD2,154m net). The 2021-2025 average already contains the record's weak years (2021 at −19%), and no deal with an evidenced date falls in this year, so no dip or lump is placed. The ~3% is the measured growth in qualifying deal supply: net deployment rose 3% a year from 2021 to 2025 while owner cash more than doubled (deployNetToOwnerCash* fell from 100% to 53%). Sources: csi-fy2021-mda, csi-fy2022, csi-fy2023, csi-fy2025, csi-q2-2026, csi-2021-capital | ### downside Stress 2% annual erosion of the existing cash base, deploy only half the available surplus, earn an 8% mature purchase cash yield and exit at 15x. This tests weaker retention, Altera pressure, poorer deal economics and a shorter runway; retained cash remains an asset. It is not a default scenario. | Driver | Value/path | Reason and sources | | --- | --- | --- | | exitMultiple | 15.000 | Chosen terminal operating equity cash yields are 5%, 6.67% and 4%, giving 20x, 15x and 25x. They reflect alternative cash durability and reinvestment runway; they are independent judgments, not historical deal returns or current-price-reset formulas. Retained assets and IRGA/other claims are valued separately. | | dilutionRate | 0.000 | Zero further issuance follows the unchanged reported share count. Future DRIP, compensation or financing issuance would require a revision; no speculative buyback gains are included. | | reinvestmentRate | 0.500 | Base/upside full deployment follows the stated acquisition objective after dividends, operating reserve and old claims. The 50% stress retains cash if qualifying opportunities are scarce. Full deployment is not proof that the same purchase yield remains available at scale. | | organicGrowth, years 1–1 | 0.000 | Year 1: AI-assisted substitutes and budget pressure stall renewals: recurring growth outside Altera falls from 4% to about 1%, Altera keeps falling 15%, licenses and services −5%. Net about 0%. Sources: csi-q2-2026, csi-fy2025 | | organicGrowth, years 2–10 | -0.020 | Years 2-10: migrations under way; recurring revenue declines about 1-2% a year as customers leave and price increases stop, licenses and services keep shrinking. Explicit erosion stress. Sources: csi-q2-2026, csi-fy2025 | | organicMargin, years 1–1 | 0.175 | Year 1: starting margin; costs have not yet been cut. Sources: csi-fy2021-mda, csi-fy2022, csi-fy2023, csi-fy2025 | | organicMargin, years 2–5 | 0.165 | Years 2-5: operating deleverage. R&D (13.9% of revenue) is held in absolute terms to defend products while revenue falls 2% a year, cutting the margin about 0.2-0.3 points a year. Sources: csi-fy2021-mda, csi-fy2022, csi-fy2023, csi-fy2025 | | organicMargin, years 6–10 | 0.155 | Years 6-10: further deleverage and price pressure; margin settles about 2 points below today, still above Altera's 12.9%. Sources: csi-fy2021-mda, csi-fy2022, csi-fy2023, csi-fy2025 | | priceToSales, years 1–10 | 1.500 | Auction pricing: private capital and CSU's own scale push the price per dollar of revenue to 1.5x, between the measured 1.17x and Synchronoss's 1.8x. Sources: csi-fy2022, csi-fy2023, csi-fy2025, lumine-q2-2026-fs, csi-q2-2026 | | acquiredMargin, years 1–10 | 0.120 | Acquired businesses reach only Altera-like margins (12.9% FCFA2S margin in FY2025), so the deal yield is 12% / 1.5 = 8%, the same 8% stress as before the rebuild. Sources: csi-fy2022, csi-fy2023, csi-fy2025, lumine-q2-2026-fs, csi-q2-2026 | | acquiredGrowth, years 1–10 | -0.020 | Acquired businesses erode like the rest of the portfolio. Sources: csi-q2-2026, csi-fy2025 | | deploymentCapacity, years 1–1 | 1879.350 | Year 1: the latest full year, FY2025 net USD1,340m (8% below the five-year average), as the 2023-2025 decline continues. Sources: csi-fy2021-mda, csi-fy2022, csi-fy2023, csi-fy2025, csi-q2-2026, csi-2021-capital | | deploymentCapacity, years 2–3 | 680.200 | Years 2-3: a two-year dry spell at the 2020 level (USD485m net, 33% of the average), at least as long as the record's longest quiet run (2020-2021, both below 90% of the average): sellers wait out lower valuations or private capital outbids CSU. Sources: csi-fy2021-mda, csi-fy2022, csi-fy2023, csi-fy2025, csi-q2-2026, csi-2021-capital | | deploymentCapacity, years 4–10 | 1879.400 | Years 4-10: back to the FY2025 level with no growth, as in 2022-2025 when deployment was flat to falling. Sources: csi-fy2021-mda, csi-fy2022, csi-fy2023, csi-fy2025, csi-q2-2026, csi-2021-capital | ### upside Use 3% existing cash growth, full cash deployment, the achieved 15.1% aggregate purchase cash yield (no better-than-record deal prices) and a 25x exit. Requires resilient customer economics and continued sourcing discipline. No margin expansion, new leverage or rerating of retained investments is credited. | Driver | Value/path | Reason and sources | | --- | --- | --- | | exitMultiple | 25.000 | Chosen terminal operating equity cash yields are 5%, 6.67% and 4%, giving 20x, 15x and 25x. They reflect alternative cash durability and reinvestment runway; they are independent judgments, not historical deal returns or current-price-reset formulas. Retained assets and IRGA/other claims are valued separately. | | dilutionRate | 0.000 | Zero further issuance follows the unchanged reported share count. Future DRIP, compensation or financing issuance would require a revision; no speculative buyback gains are included. | | reinvestmentRate | 1.000 | Base/upside full deployment follows the stated acquisition objective after dividends, operating reserve and old claims. The 50% stress retains cash if qualifying opportunities are scarce. Full deployment is not proof that the same purchase yield remains available at scale. | | organicGrowth, years 1–1 | 0.025 | Year 1: recurring growth outside Altera returns to 5% (FY2025 FX-adjusted recurring), Altera stabilizes at −5%. Sources: csi-q2-2026, csi-fy2025 | | organicGrowth, years 2–10 | 0.040 | Years 2-10: recurring at its 2024-2025 range (4-6% FX-adjusted by quarter) and licenses and services flat; reported organic growth was 4% in FY2025. Sources: csi-fy2021-mda, csi-fy2022, csi-fy2023, csi-fy2025 | | organicMargin, years 1–3 | 0.175 | Years 1-3: starting margin; mix shift nets to about zero. Sources: csi-fy2021-mda, csi-fy2022, csi-fy2023, csi-fy2025 | | organicMargin, years 4–10 | 0.180 | Years 4-10: AI tooling lowers development, support and services cost (staff in those three functions is about 35% of revenue); a 2% efficiency adds about 0.6-0.7 points of margin. Not evidenced yet, hence upside only, and only after year 3. Sources: csi-fy2021-mda, csi-fy2022, csi-fy2023, csi-fy2025, csi-ai-2026 | | priceToSales, years 1–10 | 1.169 | Measured price per dollar of acquired revenue, 1.17x (net capital at entry, 2021-2025 cohorts). Flat because the evidence shows no rising trend: acquisition capital per dollar of sales change fell from 1.07 (FY2021) to 0.86 (FY2025) while deal size grew, and the aggregate already includes the large deals (Altera 0.86x, Synchronoss 1.8x, Cipal Schaubroeck 2.1x, WideOrbit 2.9x, Optimal Blue 3.8x). Sources: csi-fy2022, csi-fy2023, csi-fy2025, lumine-q2-2026-fs, csi-q2-2026 | | acquiredMargin, years 1–1 | 0.160 | Year 1: new cohorts start below the blended margin, as in the base, but integrate faster. Sources: csi-fy2022, csi-fy2023, csi-fy2025, lumine-q2-2026-fs, csi-q2-2026 | | acquiredMargin, years 2–10 | 0.178 | Years 2-10: measured blended acquired margin, 17.8%. Sources: csi-fy2022, csi-fy2023, csi-fy2025, lumine-q2-2026-fs, csi-q2-2026 | | acquiredGrowth, years 1–1 | 0.025 | Year 1: recurring growth outside Altera returns to 5% (FY2025 FX-adjusted recurring), Altera stabilizes at −5%. Sources: csi-q2-2026, csi-fy2025 | | acquiredGrowth, years 2–10 | 0.040 | Years 2-10: recurring at its 2024-2025 range (4-6% FX-adjusted by quarter) and licenses and services flat; reported organic growth was 4% in FY2025. Sources: csi-fy2021-mda, csi-fy2022, csi-fy2023, csi-fy2025 | | deploymentCapacity, years 1–1 | 3465.578 | Year 1: the record twelve months to September 2026 continue (USD2,471m net, bound to the observed TTM; USD2,308m at June): the new President's first year set a record and the large-deal practice keeps closing deals like Synchronoss and DerbySoft. Sources: csi-fy2021-mda, csi-fy2022, csi-fy2023, csi-fy2025, csi-q2-2026, csi-2021-capital, csi-q3-2026-fs, csi-q3-2026-mda | | deploymentCapacity, years 2–3 | 3716.600 | Years 2-3: USD2,650m average, about 7% a year above the September record (set when the June record was USD2,308m; not rebased, since a higher TTM does not show more sellers), below the 15% a year from 2021 to mid-2026, as more operating groups and the large-deal team source in parallel. Sources: csi-fy2021-mda, csi-fy2022, csi-fy2023, csi-fy2025, csi-q2-2026, csi-2021-capital | | deploymentCapacity, years 4–5 | 4516.100 | Years 4-5: continued ~10% growth (USD3,220m average). Sources: csi-fy2021-mda, csi-fy2022, csi-fy2023, csi-fy2025, csi-q2-2026, csi-2021-capital | | deploymentCapacity, years 6–10 | 6367.400 | Years 6-10: continued ~10% growth (USD4,540m average); needs sourcing to scale with cash, which 2022-2025 did not show. Sources: csi-fy2021-mda, csi-fy2022, csi-fy2023, csi-fy2025, csi-q2-2026, csi-2021-capital | ## Estimation limits and revision triggers Synchronoss and July sales annualization are flat-quarter estimates; the Q3 additional-acquisition cohort (USD374.5m of consideration) is a residual that includes purchase-accounting revisions; mature target cash margins are not reported owner cash. Undisclosed pool sales use matched net or gross capital/revenue prices with explicit timing assumptions. Open commitments are not labelled completed. Restricted subsidiary cash, the IRGA claim, customer attrition and target returns remain genuine risks. Next triggers: the 9 November Q3 call (deal-flow and AI-hurdle commentary, treated as management claims); FY2026 results (full-year net deployment re-bases the deal budget; Q4 working capital tests whether the 9M outflow reverses); Altera recurring growth through the end-2027 client departures. Named arithmetic, stable fact IDs and source links are retained in valuation/calculations.json. Inputs follow the filing; the IRR effect was computed afterwards and no input was tuned to it. ## Q3 2026 roll, 6 November 2026 The Q3 statements and MD&A were read from the primary PDFs and the September bridge replaces the June one. Before/after inputs and returns at the same quote and date are in [the dated update](../updates/2026-11-06-q3-roll.md) and its [model-change record](../updates/2026-11-06-q3-model-change.json). # Underwriting plan Written on 3 November 2026 for the model anchored 27 October 2026; rolled to the Q3 2026 filing on 6 November 2026 (exits unchanged). Money in USD millions unless marked CAD; the model runs in CAD at 1.4025. ## Opportunity ledger | Opportunity | Starting point | Customer need and why CSU wins | Reachable in five years | Investment and cash cost | Timing and limits | Evidence against | | --- | --- | --- | --- | --- | --- | --- | | Installed-base price and retention (recurring) | 8,700 recurring revenue (FY2025), 75% of revenue, about 85% of EBITA | Mission-critical workflow software with costly migration; CSU maintains it (R&D 13.9% of revenue) | +3% to +5% a year if 2024-2025 FX-adjusted recurring growth (4-6% by quarter) holds; roughly flat if Q2 2026 (+2%) persists. Q3 2026: +4% group, +4% outside Altera, +5% outside US healthcare and Altera | Already in the cost base: R&D, maintenance staff. No capital | Annual renewals; public-sector budget cycles | Altera recurring −19% in Q2 2026, −8% in Q3; AI-assisted replacement could raise churn and cap price increases | | New licenses, implementation, hardware | 2,923 (FY2025), declining (licenses −8%, services −4% FX-adjusted) | New modules and new customers | Flat to declining | Services staff 1,202 | Cyclical; Covid showed licenses −28% in one quarter | Low margin; matters little to cash (stage 3) | | Altera stabilization | 712 revenue, FCFA2S 92 (FY2025); 445 and 72 in 9M 2026, 4% of group revenue | Hospital EHR and practice software | Range: further 10-20% revenue decline to stabilization | Restructuring cost already expensed | Hospital IT replacement cycles of several years | Revenue down 16% since purchase; 9M 2026 total organic −18% (recurring −11%); management expects sizable client departures to continue to the end of 2027 | | Small and mid VMS acquisitions at traditional hurdles | Most of the 2,733 deployed in the year to September 2026 (2,471 net of acquired cash; 1,861 net excluding Synchronoss and DerbySoft) | Founders who want a permanent owner; proprietary pipeline across six operating groups | Base budget USD1,651m net in year 1 growing about 3% (2021-2025 average 1,461; record TTM 2,471 not used as the anchor) | Purchase price about 6.6x entry owner cash (16.4% measured net, 15.2% at entry); acquired working capital negative | Limited by sourcing and by managers' capacity to integrate; no new leverage | PE competition when rates fall; two recent US healthcare purchases carry very high known attrition (healthcare recurring −10% in Q3, −3% without them); acquired cohorts earned 18% cash margins against 21.7% for the group | | Large VMS businesses and carve-outs (lower hurdle) | Altera 727, Optimal Blue 705, WideOrbit 504, Synchronoss 309, DerbySoft 391 (91.8%, June 2026) | Corporate sellers want speed; CSU now bids in auctions | Lumpy: 0-2 deals a year (40-70 large VMS businesses are sold annually) | Lower yields: Altera 12.7% levered; DerbySoft cost about 2.5x net closing cash per dollar of annualized sales, twice the 1.17x the base pays | Brokers invite CSU into a minority of processes | Shrinking revenue at the large carve-outs | | Non-VMS investments (Asseco, PEMS) | Asseco group fair value 1,092 (September 2026); Sabre 13% | Leonard's new circle of competence | Not forecast | Carried at fair value, outside deployment | Leonard now an adviser for this | No measurable record | | AI-driven internal productivity | R&D, maintenance and services staff cost about 4,040 | Cheaper development and support | Upside only, from year 4: +0.6 points of owner-cash margin (a 2% saving on that cost base); base and downside hold the margin flat because savings are first spent defending the installed base | Chief AI Officer appointed October 2026 (`csi-ai-2026`) | No margin gain in years 1-3 in any case | Savings may be competed away through lower prices or reinvested in product without lifting margin | Overlap: acquired cohorts' later organic growth is not in the organic rate (the engine holds new cohorts flat), and the deal yield is the measured aggregate that includes such growth, so each effect appears once. Organic cash growth applies only to the existing portfolio and previously funded cohorts. Absolute scale check (base, five years). Revenue rises from USD13,204m (trailing to September 2026) to USD23,245m in year 5, including committed and missing-month cohorts. Owner cash rises from CAD3,238m to CAD5,702m, 1.8x. Five-year deployment is CAD12,291m (USD8,764m net), about 1.2x the 7,304 of net acquisition cash in 2021-2025, matching the record's roughly 3% annual growth in deal supply; about CAD11,277m beyond regular dividends is distributed because the budget, not cash, binds (as in 2025, when CSU deployed 53% of its unlevered owner cash). At the measured price of about 1.1-1.4x revenue, that buys roughly 10,000-13,000 of revenue, close to doubling the group, which would need roughly as many additional employees as CSU has today and twice the recent number of deals or larger deals. That is the base case's largest demand on the organization. The downside deploys CAD6,999m at 1.5x revenue (with a two-year drought) and the upside CAD19,931m. Management targets as bounds. CSU publishes no growth or margin targets. The stated intent (2021 letter) is to deploy all FCFA2S rather than distribute it; the bonus rewards ROIC above 5% and revenue growth. The base states a yearly deal budget anchored on the 2021-2025 through-cycle average (about 3% growth, no placed dips) rather than assuming all cash is placed, and prices deals at the measured aggregate rather than at any hurdle. Following the overprecision lesson, the scenario range is wider than any historical pattern: the downside has deployment and returns well below anything in 2021-2025. ## Scenario mechanisms | | Downside | Base | Upside | | --- | --- | --- | --- | | Organic revenue growth (years) | 0% (1), −2% (2-10) | 1.5% (1), 2% (2-3), 3% (4-10) | 2.5% (1), 4% (2-10) | | Owner-cash margin on organic revenue | 17.5% (1), 16.5% (2-5), 15.5% (6-10) | 17.48% flat | 17.5% (1-3), 18.0% (4-10) | | Acquired businesses' margin | 12% | 15% (1), 16.5% (2-3), 17.8% (4-10) | 16% (1), 17.8% (2-10) | | Acquired revenue growth | −2% | as organic | as organic | | Deal cash in the purchase year | half | half | half | | Price per dollar of acquired revenue | 1.5x (8% yield at 12% margin) | 1.17x (15.2% yield once mature) | 1.17x | | Share of surplus cash offered to deals | 50% | 100% | 100% | | Yearly deal budget (USD m net) | 1,340; 485 dry spell (y2-3); 1,340 (y4-10) | 1,651 growing about 3% a year to 2,154 (y10); no placed dips | 2,471 (September TTM); 2,650 (y2-3); 3,220 (y4-5); 4,540 (y6-10) | | Unplaced cash | retained (half the surplus) and paid if the cap binds | distributed in the year | distributed in the year | | Financing and dilution | no new debt, no shares | same | same | | IRGA claim | 1,295 (September 2026) growing 10% a year | same | same | | Exit multiple of operating equity cash | 15x | 20x | 25x | | Year-5 revenue (USD m) | 16,488 | 23,245 | 28,652 | | Year-5 cash earnings (CAD m) | 3,631 | 5,702 | 7,161 | Mechanisms behind each assumption: - **Downside.** AI-enabled competitors and customers' budget pressure erode the installed base: renewal pricing stalls and churn rises enough that existing cash declines 2% a year (recurring revenue falling about 1-2% after cost cuts). At the same time private capital bids up VMS prices or CSU accepts lower returns at scale, so new deals earn 8%. Fewer deals clear the hurdle, so half the surplus stays in cash. The 15x exit follows from slower growth and lower returns on new capital (Koller: at 9% cost of equity, 4% growth and 15% RONIC give about 15x), not from the lower earnings themselves. This is a lasting impairment of part of the franchise, not a temporary under-use: the software doesn't come back once customers migrate. - **Base.** The installed base grows about 3% a year: recurring revenue (about 85% of EBITA) grows 4-5% outside Altera while licenses, services and Altera shrink, which weights to about 3% of profit, the same as the 2021-2025 average organic rate (3.4%). Acquired businesses grow at the same rate, because CSU's organic measure already includes them; their purchase yield is stated at entry so that growth is not counted twice. Deals continue at the measured 2021-2025 yield within a yearly budget anchored on the 2021-2025 through-cycle average (USD1,461m net), moved to 2027 at the measured ~3% trend and growing at that trend. No dips are placed: the average already contains the record's weak years, and a lump goes in a specific year only when its timing is evidenced by an announced, signed or financed deal (none beyond the reserved post-September pool of USD673m). The downside's dry spell is a scenario mechanism, not a timing forecast. Each deal is funded from cash before it contributes and earns half a year of cash in its purchase year, the same convention used to measure the yield. The 20x exit corresponds to about 6% perpetual growth at 15.2% RONIC and a 9% cost of equity. - **Upside.** Recurring growth returns to the 2024-2025 range (5% cash growth) and sourcing scales with cash (the record pace continues and grows about 10% a year) at the same measured yield, with no better prices than the record. The 25x exit needs its own mechanism: a longer runway at 15% RONIC (about 6.8% perpetual growth gives 25x). It is not granted by higher growth alone. Temporary versus permanent. A deployment stall (2023-2025) is temporary under-use: cash accumulates and the model distributes or retains it at par. Customer migration away from a product is permanent and is what the downside's −2% represents. Integration takes time: a new cohort earns half a year in its purchase year and its entry run rate from the next year; previously funded cohorts valued on mature estimates ramp over one year. The model uses the shared `equity-acquirer` adapter (after-financing common cash with funded acquisition cohorts, deployment capacity, retained claims and assets); input notes in `valuation/model.json` record each input's classification and sources. ## Q3 2026 deployment test — 6 November 2026 Question: did 2026 deployment slow, and if so, because fewer qualifying sellers came to market or because CSU raised its hurdle on AI-exposed targets (`updates/2026-11-06-topicus-ai-briefing.md`)? The premise fails at group level. Net deal cash was USD1,986m in the first nine months of 2026 against 855 a year earlier; Q3 alone was 550 against 307. Nine months already exceed the base full-year budget (1.20x), and the twelve months to September (2,471) are a record. Without the two named large deals (Synchronoss 275 and DerbySoft 335 net) small and mid-sized deals still absorbed 1,861, 1.13x the base year-1 budget. The decline that is real is deployment as a share of cash, 2021-2025. What each explanation predicts, and what Q3 shows: | Test | Fewer qualifying sellers | Higher hurdle on AI-exposed targets | Q3 2026 evidence | | --- | --- | --- | --- | | Deal cash | Falls | Falls, or shifts to less-exposed verticals | Rose: 9M net 1,986 vs 855 | | Price per dollar of sales | Rises (more bidders per seller) | Falls on exposed targets | Not measurable for small deals (no target sales disclosed); DerbySoft about 2.5x net closing cash per dollar of sales, high | | Verticals bought | Unchanged | Fewer content and knowledge-work niches | Still bought content management, marketing, legal, education and accounting software | | Declining targets | Avoided | Bought only at prices that model the decline | Two US healthcare purchases with very high, known attrition were "modeled as part of the respective investment theses" (MD&A p.5) | | Bargain gains | No change | More | 17 vs 6 (9M), small | Verdict: neither explanation is visible in the aggregate, because deployment did not slow. The healthcare purchases fit a hurdle that prices decline rather than one that avoids it. The filings cannot separate the two explanations at the level of individual deals: no deal count, no target sales and no rejected-deal data are disclosed. No deployment path changes. The base stays at the full-year through-cycle average and moves when FY2026 enters that average. Next evidence: the 9 November call (management claims, not data), Topicus's own deal pace, and FY2026 net deployment by segment. ## Revenue bridge — rolled to September 2026 The 5 November closure built the cohort bridge on the June parent window. Q3 now replaces it: Synchronoss, DerbySoft and the July Lumine businesses have owned Q3 sales inside the September TTM, so only their missing quarters remain. Other H1 deals keep half a year, the late-FY25 cohort is removed (wholly inside the TTM), and a provisional Q3 cohort (USD374.5m consideration) is added. The post-September pool (526 closing, 148 deferred) replaces the post-June pool. Base revenue is CAD32,601m in year 5 and CAD50,002m in year 10; operating owner cash is CAD5,702m and CAD8,787m. Sales annualizations are flat-quarter estimates and mature target cash margins are not reported owner cash. Restricted subsidiary cash, the IRGA claim, customer attrition and target returns remain genuine risks. [Inputs before and after, and returns at the same quote and date](../updates/2026-11-06-q3-model-change.json). The 5 November closure record stays in `updates/2026-11-05-revenue-bridge-closure.json`. # Capital productivity and reinvestment Assessment:5November2026. These are shared-engine diagnostics from the same owner-cash projection shown in the dashboard, not a new historical measurement. Money is in CAD millions. | Horizon | Scenario | Return/proxy definition | Rate | Period-weighted reinvestment | New growth capital | Matching earnings | Funding residual | | --- | --- | --- | ---: | ---: | ---: | ---: | ---: | | 5 years | downside | Acquisition ROIIC · mature cash proxy | 8.00% | 40.69% | 6998.550 | 17200.963 | 5.46e-12 | | 5 years | base | Acquisition ROIIC · mature cash proxy | 14.33% | 54.35% | 12291.312 | 22614.334 | 1.82e-12 | | 5 years | upside | Acquisition ROIIC · mature cash proxy | 14.96% | 74.85% | 19607.287 | 26193.828 | 3.64e-12 | | 10 years | downside | Acquisition ROIIC · mature cash proxy | 8.00% | 46.21% | 16395.550 | 35477.826 | 3.64e-12 | | 10 years | base | Acquisition ROIIC · mature cash proxy | 14.81% | 44.75% | 26540.212 | 59312.204 | 3.64e-12 | | 10 years | upside | Acquisition ROIIC · mature cash proxy | 15.12% | 64.68% | 51444.287 | 79539.772 | 7.28e-12 | The June parent reporting window remains unchanged. The bridge now separates Synchronoss, DerbySoft, other H1 purchases, late-FY25 purchases, completed July Lumine businesses and other completed/open commitments. Latest Lumine Q3 owned sales update the forward missing-period estimates; the parent has not yet reported Q3. Gross consolidated sales are paired with common cash at specific 61.4% Lumine ownership where known, using the existing allocation estimate only for unsegmented pools. The same ownership now applies to known July cash/deferred reservations. Reinvestment is total new growth investment divided by total matching earnings, not an average of annual percentages. Previously funded cohort contributions are not new investment returns. Acquisition cash-yield proxies and forecast-implied productivity are not measured historical ROIIC or project IRRs. All dividends, existing claims, liquidity and financing are reconciled before new purchase capital. Synchronoss and July sales annualization are flat-quarter estimates; mature target cash margins are not reported owner cash. Undisclosed pool sales use matched net or gross capital/revenue prices with explicit timing assumptions. Open commitments are not labelled completed. Restricted subsidiary cash, the IRGA claim, customer attrition and target returns remain genuine risks. Historical measurements and definitions remain in 04-capital-allocation.md and valuation/calculations.json. [Dated source bridge and funding ledger](../updates/2026-11-05-revenue-bridge-closure.json). # Scenario probabilities Assessment date: 2026-10-27; last reassessed 2026-11-06. These are initial subjective judgments, with low confidence. Financial sources support the mechanisms, not measured numerical odds. The original authors did not provide or endorse these weights. | Horizon | Downside | Base | Upside | | --- | ---: | ---: | ---: | | 5 years | 20% | 60% | 20% | | 10 years | 30% | 50% | 20% | 5-year reasoning: A diverse installed base and demonstrated acquisition process support the central cash-generation mechanism. The forecast still depends on normalized common-owner cash, restricted subsidiary funding and sustainable purchase returns. 10-year reasoning: The growing absolute acquisition budget and accumulated ownership/refinancing claims make full deployment less certain over a decade. Shift weight from central to downside instead of assuming historical deal returns persist indefinitely. Downside: Retention, Altera performance or higher purchase prices can impair cash while slower deployment preserves liquidity. Financing, IRGA and subsidiary claims justify a material downside weight despite diversification. Base: Low organic growth, recurring customer cash and disciplined purchases are consistent with the observed model. Full deployment at the inferred yield remains a judgment rather than an observed forward pipeline. Upside: Stronger existing cash growth and continued attractive sourcing can outperform. The larger transaction scale and assumed 25x exit prevent treating the favorable path as equally established as base. Review triggers: - Organic cash growth and customer retention. - Acquisition spend, all-in cash yield and integration capacity. - IRGA/Joday obligations, subsidiary liquidity and refinancing. Each path represents a broad outcome region, not the probability of hitting an exact share price or IRR. Paths are a coarse approximation of a continuous joint distribution; the displayed downside and upside are not bounds. Joint driver dependencies must be considered, rather than multiplying marginal driver guesses. Scenario weights sum to 100% within this approximation. Severe outcomes remain possible outside the modeled paths. Price movement alone does not change these weights. A changed causal premise, acquisition capacity, financial report or scenario definition does. Local assumption sensitivities suspend the researched weights until the full distribution is reassessed. No weighted-average IRR is presented as an expected shareholder return. Method: [Damodaran, chapter 33](https://pages.stern.nyu.edu/~adamodar/pdfiles/val3ed/c33.pdf), printed pp.894-897. See `valuation/model.json` for machine-readable weights and `docs/method/cash-flow-and-reinvestment.md` for the capital definitions. ## 3 November 2026 review The scenario definitions changed. The base purchase yield is now the measured CSU-group aggregate, deal capacity is capped at the observed pace with capacity growth of 0%/10%/15% by scenario, and unplaced cash is distributed. Weights are reassessed and kept: the downside adds a deployment stall that 2022–2025 history makes plausible; the base becomes more constrained; the upside loses better-than-record deal prices but requires capacity to scale with cash. None of this justifies a finer numerical revision. Neither the label comparison nor price movement changes odds. ## 3 November 2026 pipeline review The full pipeline review changed no scenario definition or input. It found evidence on both sides of the existing weights: the profit pool (recurring maintenance, about 85% of EBITA) still grows outside Altera and deployment reached a record after the leadership change, supporting the base; returns on new capital have about halved since 2015 and common owners kept only about half of the consolidated cash growth, supporting a material downside weight. Neither shifts the coarse weights; 20/60/20 (5 years) and 30/50/20 (10 years) are kept. Q3 2026 results on 6 November are the next reassessment point. ## 3 November 2026 reference reconciliation Scenario definitions changed (base organic growth 3%, upside 5%; acquired cohorts grow at each scenario's rate, so the downside's acquired businesses also erode at −2%; half-year deal cash; net-of-acquired-cash entry yield and capacity). The weights were reassessed and kept at 20/60/20 (5 years) and 30/50/20 (10 years). The base is now the evidence-weighted central path rather than a conservative one, which is what a 60% weight should describe; the downside still captures the AI-erosion and overpayment mechanisms with unchanged severity. The changes came from evidence on growth and timing, not from the investor labels' returns, and no weight was fitted to them. ## 3 November 2026 revenue rebuild The scenarios are now stated as yearly revenue, margin and deal-price paths. The base path is slightly less favorable early (year-1 organic revenue 1.5%, young-cohort acquired margins) and the downside slightly more severe (margin deleverage). The outcome regions the weights describe are unchanged, so 20/60/20 and 30/50/20 are kept. ## 4 November 2026 yearly deal budget Deployment is now a stated yearly budget: the base is the expected budget, anchored on the 2021-2025 through-cycle average with about 3% growth and no placed dips, the downside carries a two-year drought, and the upside keeps the record pace growing about 10%. These redefine where each path sits but not which business outcomes they represent; the weights stay 20/60/20 and 30/50/20. The weighted 5-year return (12.1%) now sits at the 12% hurdle. ## Source bridge review — 5 November 2026 The revenue-to-cash coverage gap is closed. The June parent reporting window remains unchanged. The bridge now separates Synchronoss, DerbySoft, other H1 purchases, late-FY25 purchases, completed July Lumine businesses and other completed/open commitments. Latest Lumine Q3 owned sales update the forward missing-period estimates; the parent has not yet reported Q3. Gross consolidated sales are paired with common cash at specific 61.4% Lumine ownership where known, using the existing allocation estimate only for unsegmented pools. The same ownership now applies to known July cash/deferred reservations. Existing horizon-specific weights were reassessed and retained: the core joint demand/capital scenarios have not been replaced and exact numerical frequency evidence has not appeared. Synchronoss and July sales annualization are flat-quarter estimates; mature target cash margins are not reported owner cash. Undisclosed pool sales use matched net or gross capital/revenue prices with explicit timing assumptions. Open commitments are not labelled completed. Restricted subsidiary cash, the IRGA claim, customer attrition and target returns remain genuine risks. These are subjective estimates, not measured odds. ## Q3 2026 review — 6 November 2026 Scenario definitions are unchanged; the September bridge changes inputs only. Weights are reassessed and kept at 20/60/20 (five years) and 30/50/20 (ten years). For the base: recurring growth outside Altera held at 4% after FX, Altera's recurring decline narrowed to −8%, and nine-month net deal cash (USD1,986m) already exceeds the base full-year budget. For the downside: two US healthcare purchases with very high known attrition, DerbySoft at about 2.5x net closing cash per dollar of sales, a USD157m Q3 working-capital outflow that may or may not reverse, and IRGA up to USD1,295m. One quarter of steady recurring growth does not reduce the erosion risk the downside represents, and one strong deal year does not show the upside's sourcing scaling with cash. No weight was chosen with the return in view. Sources: csi-q3-2026-mda pp.4–5, 13, 17; csi-q3-2026-fs pp.10–15, 32. # Bull–bear review — Constellation Software, 5 November 2026 Subsequent source closure: the missing revenue bridge identified in this debate has now been completed. [Current source-derived estimates, corrections and same-price/date effects](../updates/2026-11-05-revenue-bridge-closure.md). The original independent argument record below is retained with its information set; it is not a claim that all estimated target margins have become reported facts. Independent advocates prepared opening cases without seeing the other draft, then exchanged point-by-point cross-examinations. A separate judge who did not author the prior model assessed the competing claims. Source reconstructions retain their stated provenance; inaccessible private reports were not newly read. Quoted sensitivities test arguments and are not automatically adopted forecasts. Fixed quote CAD 3465.98, market time 2026-11-04T21:00:00.000Z; valuation date 2026-11-05. Original forecast exits retained. | Horizon | Before bear/base/bull | After bear/base/bull | Weighted before → after | | --- | --- | --- | --- | | 5 years | -2.75% / 11.83% / 21.36% | -2.75% / 11.83% / 21.36% | 11.86% → 11.86% | | 10 years | 0.34% / 11.95% / 18.36% | 0.34% / 11.95% / 18.36% | 11.40% → 11.40% | ## Bull case As of 5 November 2026. No opposing advocate draft was read. Existing company research, fact/calculation ledgers and exact registered source locators were examined; original issuer sources additionally checked live: Lumine Q3, CSU Q2, Diploma Q3 transcript, NVIDIA CFO commentary, TerraVest Highland/KBK, WOSG FY26. Harvia Q3 PDF retrieval failed; its registered extraction remains the evidence limit. Private source interpretations below are the repository's documented reconstruction, not a new claim to have read an unavailable private original. The strongest bear case is slow portfolio-wide erosion while reported renewals temporarily mask migration, together with poorer large-deal economics and minority/redemption claims. Altera's recurring decline shows that switching costs are an economic question, not a guarantee. **CSU-B1 — Recurring cash deserves a less depressed organic path.** Recurring maintenance supplies about 85% of EBITA; outside Altera it grew 4% in Q2. Retaining products that integrate workflows can sustain pricing and renewals without the cost of winning a new customer. Test the existing upside organic-growth path (2.5% year 1, 4% thereafter) against base1.5%→2%→3%. Sources `csi-q2-2026`, PDF pp.5–6 and supplemental operations pp.15–17; `csi-fy2025`, PDF pp.15–17. This is a conditional stronger path, not evidence that Altera has stabilised. **CSU-B2 — Supply, rather than cash, can recover faster.** A decentralised sourcing network and the record TTM USD2.308bn net deployment make the upside budget plausible without new borrowing. Test only `yearly.deploymentCapacity` from the upside, retaining base purchase economics. Sources `csi-q2-2026`, acquisition note PDF pp.35–40; `csi-fy2025`, acquisition cash discussion PDF p.18; `research/04-capital-allocation.md`, through-cycle deployment calculations. The outside view is USD1.461bn 2021–25 average: a record window cannot establish the base. I defend upside capacity, not automatic full reinvestment. **CSU-B3 — A terminal premium can represent continued purchases.** Test25× versus 20×, holding all operating paths. Specialised products, distributed capital allocation and a reconstructed 15.1% purchase yield can support value creation beyond exit (`csi-2015-letter`, pp.3–9; `csi-2021-capital`, pp.1–2; stage 6 measured cohort). But20× already requires approximately 19 post-exit years at 10% earnings growth, 15.1% reinvestment return and 9% required return before mature no-growth value. That is a duration sensitivity, not a forecast; 25× needs still longer. I cannot promote 25× from historical market multiples alone. Missing funded-cohort revenue and uncertain IRGA/fungibility remain surviving attacks. ### Isolated shared-engine effects Saved quote 3465.98; market timestamp 2026-11-04T21:00:00.000Z. Fixed valuation date 2026-11-05; model f681130ec4c1; exits remain the original model vintage. These are sensitivities, not joint forecasts or probability revisions. Baseline base IRR 11.83% / 11.95% (5/10 years). | Claim | Exact input/path | 5-year base | 10-year base | Change, pp (5/10) | | --- | --- | ---: | ---: | --- | | CSU-B1 | `cases.base.inputs.yearly.organicGrowth` | 12.86% | 12.69% | +1.03 / +0.74 | | CSU-B2 | `cases.base.inputs.yearly.deploymentCapacity` | 14.59% | 14.90% | +2.77 / +2.94 | | CSU-B3 | `cases.base.inputs.exitMultiple` | 16.68% | 14.17% | +4.85 / +2.22 | ## Bear case ### Strongest opposing case Thousands of specialist systems make replacement risky, recurring maintenance carries most of the profit, and recurring growth outside Altera was 4% in Q2. Decentralized acquisition and modest product spending have produced measurable cash returns across many deals, so AI headlines alone are insufficient to infer that this installed base will disappear. ### Bear case The current price already capitalizes much of that durability. At the saved quote the five-year base is approximately the 12% hurdle, before personal taxes or trading costs. The ownership bridge matters: common owners retain only part of consolidated growth after financing, minorities and the Topicus IRGA claim. USD1,307m of June cash sat in subsidiaries with restrictions, and IRGA wasUSD1,186m. A common-equity cash multiple must preserve those claims; the existence of accounting cash is not proof it can fund the next acquisition. [csi-q2-2026 PDF pp.13,15,17,54–59; csi-2021-ownership pp.2–3; valuation/calculations.json ownership/claims bridge.] The outside view has deteriorated at the margin. The2021–2025 aggregate entry yield near 15% is materially below earlier bonus-basis corporate returns; those are different measures, yet both warn against carrying the small-deal era's exceptional economics to a larger capital base. CSU's2015 letter itself describes lower large-deal hurdles drawing behavior towards them. Altera recurring revenue fell 19% in Q2. It does not prove AI caused the decline, but it defeats the claim that all maintenance is immune. [csi-2015-letter pp.3–7; csi-fy2025 acquisition and Altera tables; csi-q2-2026 organic/Altera supplemental tables.] The base assumes purchased cash margins rise 15→17.8%, organic growth returns to 3%, purchased revenue grows with it, and a20x exit. The latter requires roughly 6% growth beyond 2031 at the assumed 15.2% new-capital return and 9% required return. That is a perpetual value-driver assumption, not a specified ten- or twenty-year moat. Mature niche relationships can last a long time while the supply of attractive new targets still binds. The documented fall in net deployment from about 100% to 53% of cash by 2025 is direct evidence of this constraint. [research/05-thesis.md Exit multiple; research/07-assumption-register.md 4 November deal budget; csi-fy2021-mda p.15 and csi-fy2025 acquisition cash.] ### Contested claims - CSU-B1 —At15x rather than 20x, cash generation does not provide the claimed margin of safety. The15x is the model's existing downside/no-expansion cross-check, not a prediction of market pricing. - CSU-B2 —New deals may earn Altera-like12% cash margins rather than converge to 17.8%. Keep current acquisitions and funding fixed to isolate execution risk. - CSU-B3 —One percentage point less annual organic and purchased-cohort growth tests replacement/customization spending and customer bargaining. Growth shortfall is plausible from the latest 1% group organic rate; it is not an asserted AI forecast. No verified matched peer multiple or finite post-exit excess-return period exists in the current research. The issuer news index was reopened and still lists Q3 publication after close 6 November; there is no new CSU quarter to infer from Lumine's results. The September Lumine contribution does not update every subsidiary or the parent's full ownership bridge. ### Shared-engine effects Fixed saved quote CAD3465.98; market timestamp 2026-11-04T21:00:00.000Z; valuation date 5 November 2026; fixed original exit calendar. Base before: 11.83% / 11.95% (5/10years). These are one-change sensitivities to the base, except named bundles; existing scenario probabilities are not revised or claimed as validated. | Claim | Change | 5-year base IRR | 10-year base IRR | 5-year effect | 10-year effect | | --- | --- | ---: | ---: | ---: | ---: | | CSU-B1 | Exit15x | 5.93% | 9.22% | -5.90pp | -2.73pp | | CSU-B2 | Acquired margin 12% throughout | 9.11% | 9.68% | -2.72pp | -2.27pp | | CSU-B3 | Organic and acquired revenue growth 1point lower | 11.02% | 11.17% | -0.81pp | -0.78pp | ## Rebuttals ### Bull response 5 November 2026. Written after independently frozen openings were exchanged. Every numbered bear claim is addressed below; original bull openings and sensitivities remain unchanged. 1. **Bear exit 15× versus 20×: concede duration risk; reject an automatic 15× central case.** Source `csi-q2-2026`, PDFp17 confirms 4% recurring outside Altera and−19% Altera. It does not measure nineteen years of 10% future earnings growth. My20× duration diagnostic exposes, rather than verifies, the assumption. The bear's15× is also a chosen stress, not a transaction-backed present value. Keep15× adverse; demand a finite opportunity-backed terminal derivation before promoting a new central multiple. Neither own-history40× nor today's quote decides this. 2. **Bear acquired margin 12%: concede a surviving execution risk.** Altera is an observed adverse example; the aggregate reconstructed 15.1% entry yield includes more than that one unit (`csi-fy2022/2023/2025`, acquisition notes; stage 6 cash/capital reconciliation). The base17.8% mature acquired margin is an estimated group convergence, not acquired-company guidance. I defend separate base/downside paths; do not assert 12% across every purchase from one troubled carve-out. My larger-budget bull survives only conditionally at maintained prices and margins. 3. **Bear organic/cohort growth−1point: partly rebut.** Multiple years of positive recurring revenue and 4% outside Altera are evidence against generic negative portfolio growth today (`csi-q2-2026`, p17; `csi-fy2025`, recurring tables). They cannot guarantee future pricing or retention. The one-point stress is useful, not independently established as the most likely path. My4% upside remains upside, not upgraded base. 4. **Claims, restricted cash and record-deal-window attacks: concede.** No claim is removed, no announced pool is relabelled closed, no restricted cash is released in my sensitivities. The USD2.308bn record cannot establish repeatable seller supply at 15% yield. The larger-budget claim is a conditional option, not a central upgrade. Missing known-cohort sales, IRGA growth and funding fungibility remain unresolved; conviction is moderate, not high. ### Bear response 5 November 2026, after independent openings were exchanged. 1. **Bull recurring 4% path: partly rebutted.** Q2 outside-Altera recurring 4% and the multi-year installed-base record support durability (`csi-q2-2026`, supplemental operations pp.15–17; `csi-fy2025`, revenue tables). I concede they do not establish a central negative portfolio path. They also do not establish 4% for every future year or permanent margin. Altera's observed decline proves differentiated customer/unit analysis is necessary, not that every product is being displaced by AI. Keep4% as upside and 1point lower growth as sensitivity. 2. **Bull larger sourcing budget: conditional, not a central revision.** RecordTTM2.308bn is real acquisition spending, but 2021–2025 net average 1.461bn and declining cash deployment are the outside view. New supply at maintained returns must be demonstrated separately (`csi-fy2021-mda` p.15; `csi-fy2025` acquisition cash; `csi-q2-2026` acquisition note). The bull concedes this. Likewise I concede one troubled carve-out cannot make 12% acquired margins the new central level; the aggregate 15.1% entry yield is evidence for the present conditional base. 3. **Bull20/25x continuation: central uncertainty survives.** A19year10%-growth diagnostic is a burden, not a measured advantage life. The bull concedes no matched peer cash table and no sourced finite duration. A15x stress exposes dependence but is not uniquely established as central. No mechanical cut to 15x is warranted solely because it lowers shareholder returns. Current20x needs explicit duration/risk disclosure; 25x remains upside sensitivity. 4. **Bear claims defended or conceded.** CSU-B1 retained as stress, not central verdict. CSU-B2 retained as execution downside; central 12% conceded unproven. CSU-B3 retained as one-point growth stress; evidence does not select it centrally. IRGA, restricted cash and ownership claims remain actual underwriting constraints, not removed in either case (`csi-q2-2026` pp.13,15,17,54–59). The next 6 November group quarter must be reconciled before an operating revision; Lumine alone does not establish a new whole-group bridge. Conviction is moderate in product retention, lower in exact deployment, acquired margins and terminal duration; current base at about 12% offers little margin of safety. No supported current numerical change was established by these exchanged arguments. ## Judgment Existing estimates receive no presumption merely because they are already in the model. Positive company evidence, the relevant outside view and the capital/funding bridge determine each verdict; neither arbitrary conservatism nor an unsupported favorable sensitivity replaces it. The evidence and reconstruction below explain numerical changes or explicit no-change conclusions claim by claim. | Contested claim | Verdict | Evidence/locator | Reason and model action | | --- | --- | --- | --- | | bull advocate: CSU-B1: cases.base.inputs.yearly.organicGrowth | reject | csi-q2-2026 PDF pp.5–6,13,15–17,35–40,54–59; csi-fy2025 acquisition/Altera supplemental tables; csi-2015-letter pp.3–9. Recurring ex-Altera Q2 growth 4% and FY2025 maintenance support a positive but modest base; no evidence Altera stabilizes enough for upside 4% portfolio growth. | Reconstructed the isolated proposed input/path from independent opening and shared engine, at frozen saved quote/date/horizons; source mechanism separated from unsupported numerical extrapolation. | | bull advocate: CSU-B2: cases.base.inputs.yearly.deploymentCapacity | reject | csi-q2-2026 PDF pp.5–6,13,15–17,35–40,54–59; csi-fy2025 acquisition/Altera supplemental tables; csi-2015-letter pp.3–9. Long-run average deployment and observed decline in cash deployment support capped central budgets; latest record spending supports an upside, not permanent central sourcing. | Reconstructed the isolated proposed input/path from independent opening and shared engine, at frozen saved quote/date/horizons; source mechanism separated from unsupported numerical extrapolation. | | bull advocate: CSU-B3: cases.base.inputs.exitMultiple | reject | csi-q2-2026 PDF pp.5–6,13,15–17,35–40,54–59; csi-fy2025 acquisition/Altera supplemental tables; csi-2015-letter pp.3–9. Decentralized purchases and measured 15.1% cash yield support a conditional premium, but 20x entails roughly 19 post-exit years at 10%growth/15.1%return/9%requiredreturn; 25x is not substantiated. | Reconstructed the isolated proposed input/path from independent opening and shared engine, at frozen saved quote/date/horizons; source mechanism separated from unsupported numerical extrapolation. | | bear advocate: CSU-B1: Exit15x | reject | csi-q2-2026 PDF pp.5–6,13,15–17,35–40,54–59; csi-fy2025 acquisition/Altera supplemental tables; csi-2015-letter pp.3–9. 15x sizes shorter runway; the niche renewal base and measured acquisition value creation support retaining a premium conditionally, not a no-runway central case. | Reconstructed the independent isolated stress at the same quote/date/horizons; do not treat stress size as a probability or a measured current fact. | | bear advocate: CSU-B2: Acquired margin 12% throughout | reject | csi-q2-2026 PDF pp.5–6,13,15–17,35–40,54–59; csi-fy2025 acquisition/Altera supplemental tables; csi-2015-letter pp.3–9. Altera deterioration is a real adverse observation, but its 12% scenario cannot represent every future target when matched multi-deal evidence supports higher acquired margins. | Reconstructed the independent isolated stress at the same quote/date/horizons; do not treat stress size as a probability or a measured current fact. | | bear advocate: CSU-B3: Organic and acquired revenue growth 1point lower | reject | csi-q2-2026 PDF pp.5–6,13,15–17,35–40,54–59; csi-fy2025 acquisition/Altera supplemental tables; csi-2015-letter pp.3–9. Latest weak organic growth warrants the modest recovery already central; one-point erosion remains downside because ex-Altera maintenance still grew 4%. | Reconstructed the independent isolated stress at the same quote/date/horizons; do not treat stress size as a probability or a measured current fact. | Retain existing coarse horizon-specific weights after reassessment: scenarios already represent these adverse/favorable operating mechanisms. Numerical odds are subjective, not verified frequencies; unresolved gaps prevent high conviction. No tuning to return. ## Conviction Low to moderate; recurring product economics supported, current ownership/funding and full revenue bridge incomplete. These are forward estimates with surviving attacks, not confirmed outcomes. Neither numerical calibration, a completed checklist nor passing software tests establishes investment skill. Reported facts stay separate from forward judgments, and adverse paths are not worst-case bounds. **What would change the view:** 6 November Q3: Altera maintenance, ex-Altera recurring growth, gross acquisition consideration, restricted subsidiary cash and IRGA claims. [Exact inputs, isolated sensitivities, model hashes and same-price/date before/after returns](../updates/2026-11-05-debate-review.json). ## Q3 2026 outcome of the named evidence — 6 November 2026 The debate named five Q3 observations. Results, from the Q3 MD&A and statements: | Named evidence | Q3 2026 | Bearing on the debate | | --- | --- | --- | | Altera maintenance | Recurring −8% (Q2 −19%), nine months −11%; total organic −18%; departures expected through the end of 2027 | Decline continuing but narrower; supports the base's Altera drag, not a recovery | | Recurring growth outside Altera | +4% in Q3 and over nine months | The bull premise held for another quarter; CSU-B3's one-point erosion stays a stress | | Gross acquisition consideration | 9M total 2,305; post-September pool 673 | Supply did not dry up in 2026; DerbySoft's high price supports CSU-B2's caution on large deals | | Restricted subsidiary cash | 1,025 (June 1,307) | Less trapped cash, still material | | IRGA claim | 1,295 (June 1,186); 741 current | Grew 9% in a quarter: €63m of the €102m rise came from the Asseco share price, the rest from Topicus maintenance revenue and net assets | No contested claim changes verdict. At C$3,511.69 on 6 November the five-year base is 12.18% (before the roll: 11.56%), still about the 12% hurdle; the gain comes almost entirely from adding back the 9M working-capital outflow (`07-assumption-register.md`). ## Public Primary & Secondary Sources - `csi-q2-2026` (primary): CSU Q2 2026 shareholder report and interim financial statements — https://www.csisoftware.com/wp-content/uploads/2026/08/Q2-2026-Shareholder-Report.pdf (published 2026-08-11) - `csi-fy2025` (primary): CSU FY2025 shareholder report and audited financial statements — https://www.csisoftware.com/wp-content/uploads/2026/04/Q4-2025-Shareholder-Report.pdf (published 2026-03-09) - `csi-2021-ownership` (primary): CFO explanation of Topicus deemed ownership and IRGA — https://www.csisoftware.com/wp-content/uploads/2026/04/qa-september-2021-final.pdf (published 2021-09-17) - `csi-2021-capital` (primary): Mark Leonard: 2021 capital allocation letter — https://www.csisoftware.com/wp-content/uploads/2026/04/letter-to-shareholders-february-15-2021.pdf (published 2021-02-15) - `csi-2015-letter` (primary): Mark Leonard: 2015 president’s letter — https://www.csisoftware.com/wp-content/uploads/2026/04/pl_2015.pdf (published 2016-04-26) - `csi-leadership-2026` (primary): Mark Leonard board departure and advisory role — https://www.csisoftware.com/constellation-software-inc-announces-mark-leonards-decision-to-not-stand-for-re-election-to-board-of-directors/ (published 2026-03-27) - `csi-ai-2026` (primary): Appointment of Daan Dijkhuizen as Chief AI Officer — https://www.csisoftware.com/constellation-software-inc-announces-appointment-of-daan-dijkhuizen-as-chief-ai-officer/ (published 2026-10-21) - `csi-q3-calendar` (primary): CSU Q3 2026 results and call timetable — https://www.csisoftware.com/constellation-software-inc-announces-conference-call-to-discuss-third-quarter-results-5/ (published 2026-10-16) - `csi-news` (primary): CSU issuer news index — https://www.csisoftware.com/latest-news/ - `roper-q3-2026` (primary): Roper Q3 2026 results — https://investors.ropertech.com/news-releases/news-release-details/roper-technologies-announces-third-quarter-financial-results-2 (published 2026-10-22) - `lumine-q2-2026-fs` (primary): Lumine Q2 2026 financial statements: Synchronoss acquisition — https://www.luminegroup.com/wp-content/uploads/2026/08/Lumine-Group-FS-2026-Q2.pdf (published 2026-08-04) - `boc-usdcad-2026-10-26` (primary): Bank of Canada daily USD/CAD observation — https://www.bankofcanada.ca/valet/observations/FXUSDCAD/json?start_date=2026-10-26&end_date=2026-10-26 (published 2026-10-26) - `csi-price-2026-10-26` (market data): CSU.TO delayed closing quote — https://finance.yahoo.com/quote/CSU.TO/ (published 2026-10-26) - `capital-method-roiic` (methodology): Mauboussin and Callahan: Return on Invested Capital — https://www.morganstanley.com/im/publication/insights/articles/article_returnoninvestedcapital.pdf (published 2022-10-06) - `capital-method-cash` (methodology): Dan Shuart, Eagle Point Capital: A Framework For Spotting Value Traps — https://eaglepointcapital.substack.com/p/spotting-value-traps-the-anti-compounders (published 2023-08-26) - `scenario-method` (methodology): Aswath Damodaran: Probabilistic Approaches in Valuation, chapter 33 — https://pages.stern.nyu.edu/~adamodar/pdfiles/val3ed/c33.pdf - `csi-fy2021-mda` (primary): CSU FY2021 management's discussion and analysis — https://www.csisoftware.com/wp-content/uploads/2026/04/CSI-MDA-Q4-2021-Final.pdf - `csi-fy2022` (primary): CSU FY2022 shareholder report — https://www.csisoftware.com/wp-content/uploads/2026/04/Q4-2022-Shareholder-Report.pdf - `csi-fy2023` (primary): CSU FY2023 shareholder report — https://www.csisoftware.com/wp-content/uploads/2026/04/Q4-2023-Shareholder-Report.pdf - `lumine-fy2023-mda` (primary): Lumine FY2023 management's discussion and analysis — https://www.luminegroup.com/wp-content/uploads/2025/02/Lumine-MDA-2023-Q4.pdf - `csi-mic-2026` (primary): CSU 2026 management information circular — https://www.csisoftware.com/wp-content/uploads/2026/04/MIC-EN-2026.pdf (published 2026-04) - `csu-price-history-2026-11-03` (market data): CSU.TO and USD/CAD monthly price and volume history — https://query1.finance.yahoo.com/v8/finance/chart/CSU.TO?range=12y&interval=1mo (published 2026-11-03) - `lumine-q3-2026-fs` (primary): Lumine Q3 2026 condensed consolidated financial statements — https://www.luminegroup.com/wp-content/uploads/2026/11/Lumine-Group-FS-2026-Q3.pdf (published 2026-11-04) - `topicus-ceo-2026-09` (news): Topicus.com CEO change: Ramon Zanders succeeds Robin van Poelje — https://www.hstong.com/news/hk/detail/26091505300178806 (published 2026-09-15) - `csi-q3-2026-fs` (primary): CSU Q3 2026 interim financial statements (interim financial report) — https://www.csisoftware.com/wp-content/uploads/2026/11/CSI-Financial-Statement-Q326-Final.pdf (published 2026-11-06) - `csi-q3-2026-mda` (primary): CSU Q3 2026 management discussion and analysis (Q3 shareholder report) — https://www.csisoftware.com/wp-content/uploads/2026/11/CSI-MDA-Q3-2026-Final.pdf (published 2026-11-06) - `csi-q3-2026-call-event` (primary): CSU Q3 results conference-call notice — https://www.globenewswire.com/news-release/2026/10/16/3382808/0/en/constellation-software-inc-announces-conference-call-to-discuss-third-quarter-results.html (published 2026-10-16) - `tyler-q3-2026` (competitor primary): Tyler Technologies Q3 2026 earnings release, SEC Exhibit99.1 — https://www.sec.gov/Archives/edgar/data/860731/000086073126000056/a991earningsrelease-9302026.htm (published 2026-10-28) - `csi-q3-2026-call-transcript` (secondary): CSU 9 November Q3 call: public transcription of management remarks — https://www.investing.com/news/transcripts/earnings-call-transcript-constellation-software-beats-on-eps-in-q3-2026-93CH-5012262 (published 2026-11-09) - `csi-q3-2026-call-transcript-crosscheck` (secondary): CSU Q3 2026 call: second public transcription, IRGA section — https://stockanalysis.com/quote/tsx/CSU/transcripts/731164-q3-2026/ (published 2026-11-09) --- # Diploma PLC (DPLM.L) - Legal Name: Diploma PLC - Exchange: London Stock Exchange (Main Market, FTSE 100) (United Kingdom) - Sector: Value-add specialist distribution (controls, seals, life sciences) - Currency: GBP - Reviewed: 2026-11-15 - Model Framework: Organic value-add distribution plus capped bolt-on acquisitions - Financial Period: FY2026 (year to 30 September 2026) provisional earning power from H1 actuals plus nine-month guidance; March 2026 balance sheet rolled to September with estimated H2 cash; price 2 November 2026 (anchored 2026-11-03) - Research Status: 8 of 11 research checks complete. Still to complete: Security and evidence base; Opportunities and funded scenarios; Variant view and catalysts - Static Report URL: https://longview-investing.vercel.app/research/diploma.html - Interactive Model URL: https://longview-investing.vercel.app/#/company/diploma - JSON API: https://longview-investing.vercel.app/api/companies/diploma ## Executive Summary A decentralised group of niche value-add distributors. Its owner cash comes from small, high-margin businesses that grow organically on very little capital, plus bolt-on purchases at about 8-9x EBIT. Today's earnings include an aerospace-fastener windfall at Peerless that this model does not capitalise at full value. ## Key Operational Watch Items - Peerless revenue and margin: spot aerospace fastener pricing versus contract volume - Organic growth excluding Peerless versus the 5% financial model - Acquisition spend, entry EBIT multiples and the annualised profit of the 15 deals from the last twelve months - Working-capital build (inventory) and free-cash-flow conversion ## Valuation Scenario Assumptions ### Downside Case Aerospace fastener supply catches up: Peerless revenue falls about 20% over FY27-FY28 and its margin returns to about 30% (owner-cash margin 17.3% in FY27, 16.3% after); UK and European industrial markets stay weak, so organic revenue is flat for two years, then grows 2.5%. Deals stop to FY25's 56 for two years (a dry spell longer than the record's), then return to 250, at 9.5x EBIT (2.0x revenue), growing 2.5% after purchase. Exit at 16x, near Diploma's trough trailing multiples (18-21x) less a further step for a slower business. - Key Drivers: Owner cash / new working capital and capex: 50.0% | Exit / after-financing owner cash: 16.00× | Annual share dilution: 0.0% | Surplus offered to deals: 100.0% - Organic revenue growth: 0.0%, 0.0%, 2.5%, 2.5%, 2.5% - Owner-cash margin on organic revenue: 17.3%, 16.3%, 16.3%, 16.3%, 16.3% - Price paid per unit of acquired revenue: 1.99×, 1.99×, 1.99×, 1.99×, 1.99× - Owner-cash margin of acquired businesses: 15.8%, 15.8%, 15.8%, 15.8%, 15.8% - Acquired revenue growth after purchase: 2.5%, 2.5%, 2.5%, 2.5%, 2.5% - Deal budget (capital available for acquisitions): 56.00, 56.00, 250.00, 250.00, 250.00 ### Base Case Revenue build from FY26E revenue (1,807.7) at an 18.9% owner-cash margin. Organic revenue grows 5% in FY27 (management's guide back towards its model), then 5.65% (sector rates weighted by revenue). Peerless's spot-price margin fades from 54% to 47% in FY27 and 40% in FY28 on 9% volume growth, taking the owner-cash margin to 17.6%; mix then lifts it towards 18.0%. Deal budget 250 a year in FY27-FY28 (the FY21-FY25 average), 280 in FY29-FY31 and 320 after; bought at 1.63x revenue (a 21% operating margin at the last 15 deals' 12.9% entry yield) and growing at the non-Peerless rate (4.4%, then 5.1%); about half a year's profit arrives in the purchase year; retained cash saves facility interest; new capital at an 0.6 owner-cash return. Exit at 22x after-financing owner cash. - Key Drivers: Owner cash / new working capital and capex: 60.0% | Exit / after-financing owner cash: 22.00× | Annual share dilution: 0.0% | Surplus offered to deals: 100.0% - Organic revenue growth: 5.0%, 5.6%, 5.6%, 5.6%, 5.6% - Owner-cash margin on organic revenue: 18.3%, 17.6%, 17.8%, 17.8%, 17.8% - Price paid per unit of acquired revenue: 1.63×, 1.63×, 1.63×, 1.63×, 1.63× - Owner-cash margin of acquired businesses: 15.8%, 15.8%, 15.8%, 15.8%, 15.8% - Acquired revenue growth after purchase: 4.4%, 5.1%, 5.1%, 5.1%, 5.1% - Deal budget (capital available for acquisitions): 250.14, 250.14, 280.00, 280.00, 280.00 ### Upside Case Structural end markets (aerospace build backlog, defence, datacentres, energy) keep organic revenue growth at 7.5%, near the seven-year record; Peerless eases only to 47-50%, so the owner-cash margin stays about 18.4-18.7%. - Key Drivers: Owner cash / new working capital and capex: 90.0% | Exit / after-financing owner cash: 27.00× | Annual share dilution: 0.0% | Surplus offered to deals: 100.0% - Organic revenue growth: 7.5%, 7.5%, 7.5%, 7.5%, 7.5% - Owner-cash margin on organic revenue: 18.5%, 18.4%, 18.4%, 18.4%, 18.4% - Price paid per unit of acquired revenue: 1.58×, 1.58×, 1.58×, 1.58×, 1.58× - Owner-cash margin of acquired businesses: 15.8%, 15.8%, 15.8%, 15.8%, 15.8% - Acquired revenue growth after purchase: 7.1%, 7.1%, 7.1%, 7.1%, 7.1% - Deal budget (capital available for acquisitions): 310.00, 310.00, 380.00, 380.00, 380.00 ## Research Notebook # Business model Researched 3 November 2026 from Diploma's filings and free market data, as a blind test of the pipeline. Money is GBP millions unless stated. Source IDs refer to `sources/registry.json`; fact IDs to `financials/facts.json`. ## Evidence plan What we are buying: one class of 5p ordinary shares listed in London (FTSE 100), 134.3m in issue, no controlling holder (largest notified holder Capital Group 12.96%). The fiscal year ends 30 September. FY26 results are due on 17 November 2026, two weeks after this study, so FY26 figures here are estimates from H1 actuals and the 16 July guidance. | Question | Evidence used | What it cannot settle | | --- | --- | --- | | Group economics, capital, cash | FY21-FY25 results and annual reports, five-year records back to FY16, H1 FY26 interim | Business-level P&Ls; Diploma reports three sectors only | | Acquisition economics | Acquisition notes (price, net assets, pro forma revenue and profit) FY21-H1 FY26; Peerless announcement | Post-acquisition results by deal, except Peerless indirectly | | Peerless (largest profit driver) | Peerless announcement (pre-acquisition P&L), FY24 note 11, AR2025 sector mix, H1 FY26 sector data, Q3 call | Its current revenue and margin, which are bounded estimates here | | Demand and cycle | Sector reviews, organic growth by year, issuer statements on aerospace backlog, datacentres, defence | Independent industry volume data; no paid aerospace or construction databases were used | | Competition and counterparties | Issuer risk disclosures, sector reviews, peer trailing multiples | Supplier contracts, customer interviews, competitor private accounts | | Management | Remuneration report, PDMR notices, AR governance sections | References from former employees or counterparties | No fieldwork was done: no interviews with customers, suppliers, former employees or competitors, and no site visits. Statements by management are evidence of what management says, not independent confirmation. The parent session holds an investor's result for Diploma; it was not opened. ## Value-creation mechanism Diploma owns about 40 small specialist distributors (with some light manufacturing, kitting and machining) that sell critical, low-cost-to-the-customer parts: aerospace fasteners, seals, cable, connectors, diagnostic consumables. The customer pays for availability, traceability, technical help and speed, not for the part, so gross margins are about 47% and operating margins 20-25% on very little capital. The repeatable mechanism has two steps. First, each business grows organically while needing only about 20p of working capital per £1 of new revenue, so organic growth earns roughly 80p of owner cash per £1 of new capital. Second, the group buys more such niches at about 8-9x EBIT, leaves them decentralised, and lets them compound inside the group. **Most value is created by the first step**: the measured FY21-FY25 achieved deal yield is about 13% pre-tax (about 10% after tax), close to Diploma's own post-tax discount rate (9.3-9.8%, FY25 note 10), while the organic step earns far above any cost of capital. Acquisitions matter because they add more growing niches to compound, not because they are bought cheaply. In FY24-FY26 one purchase, Peerless, earned a windfall from an aerospace fastener shortage that management did not underwrite. ## Business-model profile Comparable firms are chosen by these traits, not by the industry label. **Growth vehicle: Mixed: organic and acquisitions.** Revenue rose from 538.4 in FY2020 to 1,524.5 in FY2025 (2.83×); organic growth alone compounded to 1.64× over the same years, so acquisitions supplied the other 1.73×, roughly half of the growth each. Acquisition cash was 14.6% of revenue over FY2023–FY2025. Both vehicles earn their own return: organic growth needs about 20p of working capital per £1 of new revenue, while deals achieved about 13% pre-tax. The model forecasts organic and acquired revenue separately, each with its own capital cost. **Capital intensity: 1.4% of sales, asset-light.** Capital spending on PP&E and software is about 1.4% of revenue over FY2023–FY2025 (14.4 in FY25, 0.96× depreciation): warehouses and IT for value-added distribution. Growth consumes working capital instead (18.4% of revenue, 134 inventory days) and acquisition cash (14.6% of revenue over FY2023–FY2025), so capex understates what growth costs. **Comparables.** A comparable firm distributes specialized, low-cost but critical consumables and components to fragmented industrial and healthcare customers, holds inventory for availability, grows both organically and by regular bolt-on acquisitions, and needs little fixed capital. Matching traits: mixed growth vehicle, asset-light, working-capital-funded growth. Not the label: "Industrial distributor" suggests thin-margin, cyclical volume wholesale; Diploma earns a 22.5% operating margin from service and availability on small-ticket parts. Declared in `valuation/model.json` (`businessProfile`). `npm run check` measures deal and capital intensity from the reported facts, and the outside view grades the forecast only against firms that share them. ## Economic snapshot FY25 (year to 30 September 2025), adjusted figures, statutory balance sheet (fact IDs in brackets). | Measure | Value | Definition | | --- | ---: | --- | | Revenue (`revenue2025`) | 1,524.5 | | | Gross margin | 46.7% | 1 - cost of inventories sold 812.7 / revenue | | Adjusted operating margin | 22.5% | before acquisition amortisation and deal costs | | Incremental margin | 35.8% (FY25), 29.4% (FY24) | change in adjusted operating profit / change in revenue | | Trade working capital | 281.1 (18.4% of revenue) | inventories 297.4 + receivables 229.0 - payables 245.3 | | PP&E / right-of-use assets | 66.7 / 76.1 | | | Tangible capital incl. leases | 423.9 | trade working capital + PP&E + ROU | | Return on tangible capital, after tax | 60.6% | 342.7 x 0.75 / 423.9 (leases capitalised) | | Return on capital employed, after tax | 73.1% | (342.7 - 3.9 lease interest) x 0.75 / 347.8 (leases expensed) | | Return incl. goodwill and acquisition intangibles | 17.5% after tax | 342.7 x 0.75 / (423.9 + 563.5 + 479.1) | | Issuer ROATCE (pre-tax) | 20.9% | pro forma profit / capital incl. goodwill and past amortisation (note 15.6) | | Inventory / receivable / payable days | 134 / 55 / 110 | on cost of sales, revenue, cost of sales | | Capex / depreciation | 0.96x | 14.4 capex vs 15.0 PP&E and software depreciation; c.1% of revenue | | Net debt / covenant EBITDA | 299.4 / 360.6 = 0.8x | excludes leases (83.8) | | Excess cash | none | 81.7 cash sits inside a revolving facility | Normalisation before returns: capex was close to depreciation over FY23-FY25 (net capex 21.6, 14.0, 13.1 against depreciation of a similar size), so no capex adjustment; minorities are tiny (0.6); acquisitions are consolidated at their own margins with pro forma annualisation in ROATCE. The material normalisation is Peerless (below): about £17m of FY25 profit and about £34m of FY26E profit above a 40% through-cycle margin. Trailing to March 2026: revenue 1,647.1, adjusted operating profit 394.7 (24.0%), trade working capital 353.5 (21.5% of trailing revenue, up because of inventory bought for growth and Middle East supply risk). ## Economic segmentation Reported sectors (FY25): Controls 55% of revenue / 65% of sector profit (30.0% margin), Seals 30% / 23% (19.3%), Life Sciences 15% / 12% (19.5%); corporate costs 41.3. Revenue share understates where the profit sits. Re-cut by how value is created (FY25, pre-corporate sector profit 384.0; estimates bounded from sector data and the AR2025 business mix): | Economic unit | Est. revenue | Est. profit | Profit share | Demand driver | Character | | --- | ---: | ---: | ---: | --- | --- | | Peerless (US/EU aerospace fasteners) | c.196 | c.96 (49%) | c.25% | aircraft build rates, MRO, fastener supply shortage | structural demand, **cyclical windfall margin** | | Other aerospace and defence interconnect and fasteners (Clarendon, IS Group, Swift, Spring; CDM from June 2026) | c.350 | c.85 | c.22% | aerospace, European defence, energy | structural growth, some cyclicality | | Windy City Wire (US low-voltage cable) | c.307 | c.70 | c.18% | commercial construction, datacentres (c.15% of WCW) | cyclical construction with share gains | | Seals aftermarket and OEM (Hercules, R&G, DICSA, M Seals, DAS) | 456 | 88 | 23% | repair of installed equipment (replacement), OEM builds | mostly replacement; UK/Australia weak | | Life Sciences (Canada, ANZ, UK/Ireland, Nordics) | 232 | 45 | 12% | procedure and test volumes, supplier agency agreements | defensive, supplier-dependence risk | Method: Controls excluding Peerless earned 24.2% in FY24 (169.9 - 25.0 Peerless post-close profit, over 652.4 - 54.1). Holding that margin in FY25 leaves Peerless with about 95.6 of the 250.6 Controls profit on about 196 revenue (23% of pro forma Controls revenue, AR2025 p.23): a 49% margin, against c.30% when Diploma bought it (FY to Sep 2023 EBIT USD33.5m on USD107.7m). The split of the remaining Controls profit between WCW and the rest uses the same margin and is illustrative only. Two conclusions. About 25% of FY25 sector profit (and more in FY26E, about 27% of group profit) comes from one business on 13% of revenue, earning a margin the business never earned before 2024. And half the portfolio's profit is tied to aerospace, defence and datacentres, which explains both the recent acceleration and the elevated multiple. ## Unit economics The unit through which Diploma adds capacity is the acquired niche business. Three disclosed units, in absolute money (acquisition notes): | | Alpha Labs (UK IVD, Jul 2025) | Swift Aerospace (Dec 2025) | Peerless (May 2024) | | --- | ---: | ---: | ---: | | Price (net of cash) | 27.5 | 19.1 | 243.3 | | Revenue (pro forma annual) | 22.9 | c.26.6 (13.3 for six months) | 129.9 (FY24); USD107.7m before purchase | | Operating profit | 4.7 (20.5%) | c.4.6 (2.3 for six months, 17%) | 59.9 (46%) in FY24; USD33.5m (31%) before purchase | | Net tangible assets acquired | 10.2 (inventory 3.0, receivables 3.3, payables -3.1, PP&E 5.0, investment 1.2) | 8.3 (inventory 8.8) | 63.5 (inventory 65.6) | | Pre-tax return on tangible capital | 46% | c.55% | 94% (FY24) | | Pre-tax yield on price | 17.1% (5.9x) | c.24% on the six-month run-rate, doubled (4.2x) | 14% on the purchase underwriting (c.7x); 39% on FY25E profit | A representative bolt-on therefore looks like this (illustrative, built from the disclosed units and group ratios): revenue 25, gross profit 11.5 (46%), staff 4.0, occupancy and other costs 2.0, an allocated share of central cost 0.5 (central cost is about 2.7% of revenue and largely fixed), operating profit 5.0 (20%). Tangible capital is about 5.5 of working capital and 1.0 of PP&E, so tax at 25% leaves 3.75 of NOPAT on 6.5 of tangible capital (58%). Bought at 8x EBIT (40), it yields 9.4% after tax at entry. If it then grows 6% a year at constant margin, needing about 21p of working capital per £1 of new revenue, profit is about 6.7 after five years and the cumulative return on price plus added working capital rises to about 12% after tax. The comparison that matters is before and after Diploma. The evidence of an operating uplift is mixed. Diploma claims sales excellence, scaling investment (new facilities in the Czech Republic and Germany), cross-selling and leadership development. Measurable cases: DICSA, R&G and T.I.E. needed new leadership after purchase; Peerless's margin rose from 31% to about 49% within a year, but management attributes that to "positive tailwinds and share gains in the aerospace market" (FY24 results), not to group levers. Diploma does not centralise purchasing, so there is no TerraVest-style buying-power lever. The honest reading is that Diploma's lever is capital and patient ownership plus selective investment, which keeps good niches growing; it is not a restructuring machine. Failure case: T.I.E. (US industrial automation repair), bought March 2023 for 75.1 at a pro forma 6.9 of profit (10.9x EBIT). The FY25 Controls review says it had "more challenging performance since its acquisition in 2023, reflecting softness across automation markets", and a new leadership team was formed in late FY25. R&G (UK fluid power, 99.1 in 2022, plus four bolt-ons in FY24) has been weak since, hit by UK construction, oil and gas and agriculture. Both show the limit of the model: a niche distributor follows its end market, and decentralised ownership cannot create demand. Growth capacity inside a unit: Diploma's organic investment is mainly inventory and receivables, not capacity. Stage 6 measures the group-level organic return (about 0.8 of owner cash per £1 of new working capital and capex). ## Independent checks 1. **Unit model against group accounts.** The representative unit's ratios (46% gross, 20-24% operating margin, about 20% working capital) match the group (46.7%, 22.5%, 18.4%; 21.5% at March 2026). The group's after-tax return on tangible capital (61%) is consistent with the unit's 58%. Goodwill-inclusive after-tax returns (17.5%) sit well above the deal-level after-tax entry yields (8-10%), as expected when the group includes decades-old, fully grown businesses bought cheaply or started in-house. 2. **Acquisition record against group profit growth.** FY20-FY25 adjusted operating profit rose from 87.1 to 342.7. Organic revenue growth compounded to 1.637x. Even with all margin expansion credited to the base business, acquisitions explain at least 127.6 of FY25 profit (after removing the Peerless windfall) on 1,209.4 of matched deal capital, a 10.6% pre-tax floor. The high bound (base held at its FY20 margin) is 15.1%. Entry yields disclosed at the time (FY24 deals "6x", FY25 "8x", LTM "8x", latest seven "9x") sit inside this range. See `valuation/calculations.json` (dealYieldLow/Mid/High). 3. **Peerless allocation, adverse test.** If Controls excluding Peerless had in fact risen to 25% in FY25, Peerless's FY25 profit falls to about 91 (46%) instead of 95.6; at 24% it is about 97 (49%). Every plausible allocation gives Peerless 46-49% in FY25 and above 50% in H1 FY26, against 30% before purchase. The conclusion that a large windfall sits in current earnings survives the adverse allocation. 4. **Peer cross-check.** Large broadline distributors earn much lower margins (Bunzl, RS Group and Wesco trade at 19-23x statutory trailing earnings). The high-margin specialist peers (Fastenal 43x, Grainger 35x) earn mid-teens to 20% margins on far larger, more commoditised ranges. Diploma's 22-27% margins put it at the specialist end, consistent with the value-add service model rather than scale purchasing. Where the consensus may disagree: the market and management treat FY26's 26.5% margin and 20%+ ROATCE as evidence of a better business; this study treats about 2 points of that margin (34 of profit on about 1,800 of revenue) as a supply-shortage windfall at one business. The observation that would settle it is Peerless's margin as aerospace fastener supply catches up with build rates; management has already said it is "easing off the margin" (Q3 call). # History, market & cycle Consistent history: fiscal years to 30 September; issuer organic growth is at constant currency and includes acquired businesses' growth after their first year of ownership. FY20-FY25 revenue rose 2.83x: organic 1.637x, the rest acquisitions net of disposals and currency (FY24 -3%, FY25 -2% from currency). ## Demand drivers and cycle **Who decides.** The end user is an engineer or technician (an aerospace subcontractor's buyer, an MRO shop, a hydraulic repair shop, a hospital lab). The payer is that business. The decision-maker is often the engineer who specifies the part or the buyer who needs it today; price matters less than availability and certification because the part is a small share of the customer's cost (fasteners are up to 5% of aircraft build value, but each fastener is trivial; Peerless announcement p.1). Diploma's pitch is therefore service: kitting, traceability, vendor-managed inventory, same-day delivery. **Demand split (judgment from sector disclosures).** | Part | Share of profit (est.) | Mechanism | Cyclicality | | --- | ---: | --- | --- | | Replacement / MRO / consumables | c.45% | seals and fluid-power repair (Hercules Aftermarket two-thirds online), lab and diagnostic consumables, aerospace MRO | low | | Growth capex and new builds | c.40% | aircraft builds, defence programmes, datacentre and commercial construction cabling, OEM seals | medium to high | | Regulation-driven | c.5% | aerospace certification and traceability, medical device rules (raise barriers rather than volume) | low | | Spot-shortage premium | c.10% | Peerless spot pricing while fastener manufacturers cannot meet demand | high, reverses | **The cycle, measured.** Organic revenue growth: FY20 -7% (underlying, Covid), FY21 +12%, FY22 +15% (inflation and reopening), FY23 +8%, FY24 +6%, FY25 +11%, H1 FY26 +15%, nine months FY26 +15%. Operating margin: FY16-FY19 17.2-17.8%, FY20 16.2% at the trough, then 18.9%, 18.9%, 19.7%, 20.9%, 22.5% and 24.5% in H1 FY26 (26.5% guided for FY26). The portfolio has not had a revenue decline outside Covid in the ten years reviewed; the FY20 trough cost 10% of operating profit (97.2 to 87.1). Seals and Life Sciences have been flat to low growth since FY24, so recent group growth is Controls (FY25 +20% organic, H1 FY26 +26%). **Market misperception to test.** The market reads FY25-FY26 as a structural step-up ("sustainable quality compounding", management's phrase). The competing explanation is a supply shortage in aerospace fasteners: Peerless's margin went from c.30% to about 49% within a year of purchase and above 50% in H1 FY26, while the rest of Controls stayed around 24%. Management's own March 2026 language ("aerospace demand and supply characteristics are favourable") and July comments ("taking a little bit of the heat out of some of the spot pricing", growth moderating "back towards that 10%ish") support the shortage reading. The series that would distinguish them is Peerless's margin and Controls organic growth as aircraft build rates rise and fastener manufacturers add capacity; neither is disclosed directly, so stage 11 freezes a check on the Controls margin. ## Competition Competition happens locally and by niche, not across Diploma as a whole. - **Aerospace fasteners (Peerless, Clarendon, Swift).** Peerless says it is "one of the largest independent value-added distributors" in a "ca. $7 billion, highly fragmented" market (announcement p.1). Competitors are the large aerospace distributors and the manufacturers' own direct channels. The advantage is approved-supplier status with all major primes and about 80 subcontractors, inventory breadth (Peerless inventory 65.6 on 130 of revenue at purchase) and traceability. The competitor who could hurt it is a fastener manufacturer that chooses to sell direct to subcontractors once capacity is ample, or a larger distributor bidding for contract volume as spot premiums fade. - **US low-voltage cable (WCW).** Competes against national electrical distributors on specification (being written into datacentre and building designs), service and inventory. WCW has grown 17% a year on average since its 2020 acquisition (H1 FY26 p.7) and sources copper 100% in the US. Construction cycles and copper prices pass through. - **Seals and fluid power.** Fragmented regional markets. Hercules competes on breadth and e-commerce in US repair; R&G and DAS face weak UK and Australian markets. - **Life Sciences.** Diploma distributes other companies' devices and diagnostics under agency agreements in Canada, Australia/New Zealand, the UK, Ireland and the Nordics. Competitors are other local distributors and the manufacturers' own subsidiaries. The number of competitors in Diploma's niches is not disclosed; private-equity buyers of distributors are the main competitors for acquisitions (H1 FY26 deals cost about 6x annualised profit, Q3 deals about 10x). One-sentence advantage: Diploma's niches win on availability and technical service for parts whose failure is costly and whose price is minor, and the evidence that this is durable is ten years of 16-26% operating margins across varied end markets, including the Covid trough (five-year records FY16-FY25 and FY26 guidance), not any single scale advantage. ## Counterparty incentives | Counterparty | What they want | What Diploma gives them | Their outside option | What would make them change | | --- | --- | --- | --- | --- | | Aerospace fastener manufacturers (supplying Peerless) | full factories, broad reach to subcontractors without carrying small orders | a channel to about 80 subcontractors, kitting and inventory risk on small lots | sell direct or through other distributors | ample capacity after the shortage: they may take spot volume direct or squeeze distributor margins | | Aerospace primes and subcontractors | on-time, certified parts; no line stoppages | inventory, traceability, VMI | direct supply or other distributors | a normalised supply market lets them re-tender and push price | | Life Sciences principals (device and diagnostics makers) | market access in small countries without building a subsidiary | local sales, service and regulatory work | set up their own subsidiary or switch distributor | once a market is large enough, disintermediation; the Nordics business lost a key supplier in FY25 | | Datacentre and building contractors (WCW) | specified cable delivered to site on time | stock, cut-to-length, specification support | national electrical distributors | weaker construction or a cheaper rival written into the specification | | Vendors of acquired businesses | a home that keeps their name, staff and autonomy; a fair price | decentralised ownership; some founders stay (DAS founder led six years under Diploma) | private equity, trade buyers | a better price elsewhere; Diploma walks away from many deals (FY25) | | Lenders (USPP, RCF) | covenant safety (3.5x) | leverage 0.8x, fixed-rate notes to 2036 | — | not a binding constraint at current leverage | Dependence runs both ways. Distributors exist because manufacturers and principals do not want small, scattered customers; the risk is concentrated where a market becomes big enough to serve directly, which the AR2025 risk register names ("a key supplier revokes a supply agreement and access the market through a competitor or directly"). The Peerless windfall is the clearest counterparty-driven exposure: its margins depend on fastener manufacturers being unable to serve demand. Headlines and management statements about these counterparties are evidence of actions and of management's view, not of the counterparties' motives; no supplier or customer was interviewed. # Management & incentives The capital allocators are the CEO, Johnny Thomson (since 2019), the CFO, Wilson Ng (board member from August 2025, formerly Group Financial Controller), and the board chaired by David Lowden. Acquisitions are sourced and run largely by sector and business leaders under group approval. Sources: AR2025 remuneration report pp.81, 89, 92-93; directors' report p.98; PDMR notice 16 January 2026 (`dplm-pdmr-jan26`). ## Ownership and incentives | Person | Role | Shares (latest) | Options / awards | FY25 pay | Dealings | | --- | --- | --- | --- | --- | --- | | Johnny Thomson | CEO since 2019 | 196,483 at 30 Sep 2025; 208,500 after the January 2026 sale | 79,424 vested nil-cost options (exercised Nov 2025); unvested PSP 2023 73,608, PSP 2024 61,091, PSP 2025 55,213 (granted Nov 2025 at 5,390p) | £7.15m single figure: salary £918k, bonus £1.84m (100% of max), PSP £4.33m | Exercised 57,007 options Nov 2024 at 4,536p (26,794 withheld for tax); sold 30,000 at about 5,668p on 16 Jan 2026 | | Wilson Ng | CFO (board from 14 Aug 2025) | none at 30 Sep 2025 | 9,433 vested, 12,856 unvested | £0.66m (part year) | 50% of bonus deferred into shares until 250% of salary is held | | Chris Davies | former CFO, resigned 13 Aug 2025 "with immediate effect" | 7,741 held under post-cessation rules | 2022-2024 awards lapsed | no bonus, no payment in lieu of notice | — | | Non-executives | seven, incl. Chair David Lowden (£365k fee) | not material | none | £49k-£84k each | — | The CEO's stake is about £15.1m at 7,265p, plus about £13.8m of unvested performance shares. If the shares compound at 8% a year for five years, his owned stake gains about £7.1m, roughly one year of his FY25 pay. His interests are therefore broadly but not overwhelmingly aligned with shareholders': pay is of the same order as the gain on his holding. He meets his 300% shareholding requirement (1,389% of salary at September 2025). He took a non-executive advisory role at TDR Capital, a private-equity firm (AR2025 nomination report). Incentive metrics: - **Annual bonus** (max 200% of salary): 50% adjusted operating profit, 20% revenue, 25% free cash flow, 5% engagement. FY25 targets (on-target 308.1 profit, 1,436.8 revenue, 201.5 FCF) were beaten (342.2, 1,524.5, 247.2), paying 100%. No return-on-capital metric in the bonus. - **PSP** (300% of salary for the CEO): 75% adjusted EPS growth (25% vesting at 5% a year, 100% at 13% a year, over three years) with a ROATCE underpin; 25% total shareholder return against the FTSE 100. What the targets imply: the 2025 PSP vests in full if adjusted EPS reaches 176.0p x 1.13^3 = 254p in FY28. FY26E adjusted EPS is already about 254p (479 of operating profit less 23 of interest, taxed at 25%, over 134.3m shares), largely because of the Peerless windfall. Even on the normalised earnings in this model (about 235p), only about 4% a year growth is needed to FY28. The EPS metric rewards acquisitions funded by debt at 5% (any deal above about a 7% pre-tax yield is accretive) with only the ROATCE underpin as a return guard. ## Track record and references | Measure | Under Thomson (FY19-FY25) | | --- | --- | | Revenue | 544.7 → 1,524.5 (2.8x) | | Adjusted operating profit | 97.2 → 342.7 (3.5x) | | Adjusted EPS | 64.3p → 176.0p (18% a year) | | Shares | about 113m → 134.3m (+19%, equity raises of 189.8 in FY20 and 231.9 in FY23) | | Operating margin | 17.8% → 22.5% (FY26 guided 26.5%) | | ROATCE (issuer) | 19.1% (FY20) → 17.3-18.1% (FY22-FY23) → 20.9% (FY25) | | Deals | 57 since 2019 for £1.6bn (H1 FY26 p.5), "collectively generating returns over 20%" on the issuer's pre-tax ROATCE basis | | TSR | 119% over the three years to September 2025 (91st percentile of the FTSE 250 comparator) | Measured, not quoted: the FY21-FY25 achieved deal yield is 10.6-15.1% pre-tax (stage 6), below the "over 20%" claim, which includes the organic growth of businesses after purchase and is measured on pre-tax ROATCE. ROATCE fell while big deals were absorbed (FY21-FY23) and rose when deal spend paused (FY25: 29.6 net), as the CFO's review itself says. Decisions that failed or under-delivered: T.I.E. (2023, automation downturn, leadership replaced), R&G (2022, UK markets), Life Sciences Nordics (key supplier lost, FY25). Disposals: nine in five years at about 6x, including Kubo, Pennine and Gremtek (FY25 proceeds 46.6, gain 17.3). Price awareness: the CEO sold 30,000 shares at about 5,668p in January 2026, after the November 2025 vesting; the price is now 7,265p. A partial sale after a vesting is normal and not a signal. References and bench: no independent references (former employees, counterparties) were collected here. Bench signals from filings: an internal CFO succession in 2025, new sector HR roles, a first graduate programme, leadership changes at DICSA, T.I.E. and Diploma Australia Seals, and founders retained after purchase (Peerless's Bill Way and Don Russo; DAS's founder for six years). Key-person risk sits with Thomson, who has led the strategy since 2019. ## Capital allocation principles Tested against the five principles (Mauboussin and Callahan pp.52-54). Sources: FY21-FY25 preliminary results (CEO reviews, funds flow, cash flow statements, EPS and disposal notes); AR2025 remuneration policy (PDF p.89). Investment rates are in `04-capital-allocation.md`. | Principle | Verdict | Decision used as evidence | | --- | --- | --- | | 1. Zero-based allocation | Partly | Capital moved to new platforms rather than incumbents: about 96% of FY21-FY25 growth investment was deals (1,224 of 1,273), and the three largest went to US platforms in cable (WCW 347.7) and aerospace fasteners (Peerless 243.3; CDM about 127 in FY26). Nine businesses worth 104.5 were sold, including a1-CBISS to its largest supplier (FY21, 12.0), Kentek, the Russian filters business (November 2021, 10.0), and Hawco as "lower growth, lower margin" (FY23, 24.5). The limit: "We don't divest due to underperformance – we see that as our job to resolve" (FY25 prelim p.10). Disposals test strategic fit, not returns, so T.I.E. and the Nordics keep their capital while being fixed. | | 2. Funding strategies, not projects | Pass | The group funds platform-plus-bolt-on strategies: WCW, DICSA and T.I.E. were bought as "strategic platform acquisitions" (FY23), and R&G added two bolt-ons for 5 straight after purchase (FY22). Group capital follows the sector strategy: FY22 Healthcare field equipment (6.8, "directly supports revenue growth") and the FY23 capex step-up to scale Shoal, Hercules Aftermarket and R&G. The financial model sets group outcomes and states "what is most important is the return each business generates", so businesses holding more inventory must earn higher margins (FY25 prelim p.9). No project-by-project capital budget is disclosed. | | 3. Capital is accessible but has a cost | Mostly | Equity was not treated as free: the two placings (September 2020, 193.7 gross for about 10%; March 2023, 9.35m shares at 2,525p, 7.5%) were each tied to a named deal programme (WCW; DICSA and T.I.E.). Share compensation is settled from the EBT, which buys shares in the market (2.8 FY22, 1.9 FY23, 2.3 FY24, 4.1 FY25), and the issuer's free cash flow deducts those purchases. Net of the placings the share count is flat (weighted 124.47m→124.53m FY21-FY22; 134.02m→134.08m FY24-FY25), so awards (expense 22.0 over FY21-FY25) do not dilute. Weakness: the hurdle is pre-tax ROATCE and incentives pay on adjusted EPS, so debt-funded deals look accretive above about a 7% pre-tax yield, below the 9.3-9.8% post-tax discount rates the company itself uses. | | 4. Zero tolerance for bad growth | Partly | Management did shrink when prices were wrong: FY25 net deal spend fell to 29.6 because "we walked away from a lot of deals" (FY25 prelim p.10), at the cost of a lower M&A contribution that year. It sold lower-margin, lower-growth units, and sold Kentek before the 2022 invasion. But "delivering a Group return on capital of over 20% is not our goal", because high teens "reflects the right level of disciplined investment for growth" (FY25 prelim p.10). That accepts more deals at lower returns, which is sound only while new deals earn their cost of capital. On matched definitions they earned about that, not more: 9.6% after tax. | | 5. Price versus value | Mixed | Equity was issued when it was dear relative to what it bought: about 1,700p (derived; about 30x trailing adjusted EPS of 56.4p) in 2020 and 2,525p (23.5x FY22's 107.5p) in 2023, to buy businesses at about 8x EBIT, roughly 11x after-tax cash. Adjusted EPS then compounded at 18% a year. Measured against later value, though, about 20.7m new shares (about 15% of today's count) were sold at a quarter to a third of today's 7,265p. There have been no buybacks, even at the 4,970p 52-week low: at 28-41x adjusted earnings, a buyback yields less than the roughly 9.6% after-tax yield deals achieved, so not buying is consistent with knowing price against value. Deal prices show the same awareness (bolt-ons under 5x in FY23, 6x in FY24 and FY25, nine disposals at about 6x), but prices are rising: the last seven deals averaged about 9x and CDM about 10x. | **Governing objective.** Stated as "sustainable quality compounding": double-digit adjusted EPS growth, high-teens ROATCE, 90% cash conversion, leverage below 2x and a 5% progressive dividend (FY25 financial model). The targets are per share, and the record mostly matches. Equity was raised only for specific deals at accretive prices, award dilution is neutralised, leverage stayed below policy, and deals were refused when prices rose (FY25). Two features point towards size rather than value per share. The return measure excludes amortisation and deal costs and is pre-tax. And management deliberately trades a higher return for more deployment. The matched deal yield (12.8% pre-tax, 9.6% after tax) is about Diploma's own cost of capital, so per-share value since 2019 has come mainly from capital-light organic growth on acquired niches, not from the deals themselves. Verdict: management behaves largely as if long-term value per share is the goal, with a size bias from EPS-weighted pay that the FY25 pause shows it can resist. ## Integrity and governance - **Adjustments.** Adjusted profit excludes acquisition amortisation (61.7), deal expenses (4.0), deferred remuneration to vendors (4.9), inventory fair-value unwind (5.7) and disposal gains (17.3) in FY25. These are standard for acquirers; deferred vendor remuneration and fees are real deal costs, so this study treats them as part of all-in acquisition capital rather than ignoring them. FY25 statutory operating profit (283.7) was 83% of adjusted. The remuneration committee applied no discretion to the formulaic outcomes. - **CFO departure.** Chris Davies resigned in August 2025. The annual report attributes it to personal conduct at a company event; the chair says it was unrelated to financial performance (AR2025 pp.5, 71, `dplm-ar25`). These are issuer explanations, not an independent clearance of controls. He received no bonus or payment in lieu of notice and his unvested awards lapsed. The previous claim that the annual report supplied no explanation was wrong. No separate cash penalty was included for that premise, so correcting it leaves the financial paths and probabilities unchanged. [Correction and impact audit](../updates/2026-11-11-evening-review.md). - **Board.** Independent chair and a majority of independent non-executives; fees rose 6% and the CEO's salary 8% for FY26. Shareholders are institutional (Capital Group 12.96%, an LLC holder 5.27%, BlackRock 5.02%, Norges Bank 3.01%); no controlling holder or related-party transactions of note. - **Capital discipline signals.** FY25 deal spend fell to 56.1 gross when "we walked away from a lot of deals"; leverage policy <2x (covenant 3.5x) has been kept, and equity was raised rather than over-borrowing for large deals. - **Agency cost: control of resources.** Pay grows with size. The bonus pays 50% on adjusted operating profit and 20% on revenue, and the PSP 75% on adjusted EPS, all of which acquisitions raise. A revenue metric rewards size for its own sake. Guards: the ROATCE underpin, 25% relative TSR, the <2x leverage policy and the FY25 deal pause. Moderate cost. - **Agency cost: risk tolerance.** The CEO owns about £15m of shares and holds about £14m of unvested awards, against about £7m of annual pay, so his wealth is concentrated in Diploma. That may make him more cautious than diversified owners: equity rather than debt for WCW and the 2023 deals, leverage of 0.8x against the 2x policy. The likely cost is mild under-leverage, not excessive risk. - **Agency cost: time horizon.** The PSP measures over three years, shorter than a deal's payback, but vested shares (net of tax) must be held until the fifth anniversary of grant. Half of the shareholding requirement continues for two years after departure, and half of the bonus is deferred into shares until 250% of salary is held (AR2025 remuneration policy, PDF p.89). Low cost. - **Buybacks against share compensation.** There are no buybacks. The EBT buys shares in the market to settle awards (2.8, 1.9, 2.3 and 4.1 in FY22-FY25, against share-based expense of 22.0 over FY21-FY25), and the weighted share count excluding placings has stayed flat. The issuer's free cash flow deducts the EBT purchases, so share compensation is treated as a cash cost rather than hidden in financing. All dilution since 2019 came from the two placings. Agency cost: modest. The main one is incentive design that pays on EPS growth already secured by a windfall and does not weight returns; the valuation treats it through the deal-yield and exit-multiple assumptions (no premium for management quality beyond the measured record) rather than a separate deduction. The CEO's TDR Capital role is a time and conflict question to watch, since TDR is a possible bidder for distribution assets. # Capital allocation GBP millions. Sources: five-year records (AR2025 p.168, AR2024 p.186), funds-flow tables and cash flow statements (FY21-FY25 results), acquisition notes. Calculations in `valuation/calculations.json`. ## Sources and uses of cash | | FY21 | FY22 | FY23 | FY24 | FY25 | Total | | --- | ---: | ---: | ---: | ---: | ---: | ---: | | Adjusted operating profit | 148.7 | 191.2 | 237.0 | 285.0 | 342.7 | 1,204.6 | | Cash from operations before interest and tax | 145.9 | 180.6 | 257.3 | 279.7 | 370.5 | 1,234.0 | | of which working-capital investment | (12.6) | (25.5) | (4.2) | (8.5) | (4.6) | (55.4) | | Interest paid (incl. fees) | (5.6) | (15.0) | (26.7) | (23.2) | (26.3) | (96.8) | | Tax paid | (24.2) | (40.6) | (41.4) | (58.4) | (76.6) | (241.2) | | Net capex (capex less disposals) | (1.4) | (5.5) | (21.6) | (14.0) | (13.1) | (55.6) | | Leases, employee-trust shares, other | (5.9) | 0.9 | (3.8) | 13.8 | (7.3) | (2.3) | | **Free cash flow (issuer APM)** | **108.8** | **120.4** | **163.8** | **197.9** | **247.2** | **838.1** | | Acquisitions net of disposals (all-in) | (450.5) | (177.8) | (255.3) | (311.0) | (29.6) | (1,224.2) | | Dividends | (53.2) | (56.4) | (70.8) | (77.2) | (80.9) | (338.5) | | Equity issued | 0.6 | — | 231.9 | — | — | 232.5 | | Other | (1.2) | 0.0 | 0.0 | 4.7 | 0.0 | 3.5 | | **Change in net debt before FX** | **(395.5)** | **(113.8)** | **69.6** | **(185.6)** | **136.7** | **(488.6)** | The "leases, employee-trust shares, other" row is the residual between issuer free cash flow and the lines above it (leases, trust purchases, acquisition fees paid inside operating cash, which the issuer reclassifies to acquisitions, and timing), so it is a reconciling line, not an observed item. FY20 opened with 206.8 of net cash (after a 189.8 equity raise for WCW); FY25 closed with 299.4 of net debt, a change of 506.2, of which 488.6 is cash and 17.6 currency and non-cash. What the table shows about priorities. Over five years Diploma spent 1,224 on deals, 146% of its free cash flow, and paid 338.5 in dividends (40% of free cash flow). It funded the gap with 232.5 of new equity (plus 189.8 raised just before the window for WCW) and about 506 of net borrowing. Capex was about 1% of revenue, below depreciation in most years. There were no buybacks; employee awards are bought in the market by the trust (about 2-4 a year). Deal spend is lumpy: 450.5 in FY21 (WCW), 311.0 in FY24 (Peerless), 29.6 net in FY25, about 254 in the first three quarters of FY26. ## Investment rates per sales change Window FY21-FY25: revenue rose from 538.4 (FY20) to 1,524.5, a change of 986.1 (`salesChangeWindow`). Capex, disposal proceeds, D&A and working capital come from each year's own cash flow statement and segment or covenant note (facts `ppePurchases*`, `intangiblePurchases*`, `ppeDisposalProceeds*`, `daTangible*`, `workingCapitalOutflow*`). These are audit steps in `valuation/calculations.json`, not bound to the model. | Rate per £1 of revenue change | FY21 | FY22 | FY23 | FY24 | FY25 | FY21-FY25 | | --- | ---: | ---: | ---: | ---: | ---: | ---: | | Revenue change | 249.0 | 225.4 | 187.5 | 163.1 | 161.1 | 986.1 | | Incremental M&A (acquisitions net of disposals) | 1.81 | 0.79 | 1.36 | 1.91 | 0.18 | **1.24** | | Internally generated intangibles | n/d | n/d | n/d | n/d | n/d | **not disclosed** | | Fixed capital ((net capex − D&A)) | (3.3%) | (2.5%) | 4.2% | (1.2%) | (1.2%) | **(1.0%)** | | Working capital | 5.1% | 12.7% | 2.2% | 5.2% | 2.9% | **5.9%** | | Total | | | | | | **1.29** | Definitions and limits: - **M&A**: all-in acquisition spend net of disposals (1,224.2), including fees, deferred consideration and minorities. The working capital and fixed assets that came with acquired businesses are inside this rate, not in the other two. - **Intangibles**: Diploma capitalises no development spending, and its expensed growth investment (sales capability, engineering, the "upweighting investment in capability" planned for FY26) is not reported separately. Software purchases (5.7 over five years) are already in capex. So no separate rate is shown; FY26 capability spend is expensed growth investment inside the operating margin. - **Fixed capital**: net capex 55.6 (reconciles to the sources-and-uses table) less D&A of PP&E and software 65.6 gives (10.0). Before 23.2 of property disposal proceeds it is +13.2, or 1.3% (`fixedCapitalRateGrossWindow`). Leases are excluded: right-of-use depreciation (15.5 in FY25) roughly equals lease principal paid (14.6). - **Working capital**: the cash-flow-statement increase, 58.6, which excludes working capital bought with businesses and currency translation. The sources-and-uses table shows 55.4 because the FY23 funds-flow presentation restated FY22 from 28.7 to 25.5; the statutory statement keeps 28.7. - The window starts at the Covid trough, so FY21's revenue change includes recovery as well as WCW. Where Diploma really invests: about 96% of growth investment (1,224 of 1,273) went into acquisitions. Organic growth needed almost no capital. Fixed and working capital of 48.6 against an estimated organic revenue gain of 342.8 is 14p per £1 of organic revenue (`organicCapitalRate`, an upper bound because part of it served acquired businesses). Acquisitions cost 1.90 per £1 of non-organic revenue (`dealCapitalPerAcquiredSales`; currency and disposals blur the denominator). So a pound of acquired revenue costs about thirteen times as much capital as a pound of organic revenue. Cross-check: 1/1.90 of revenue per pound at a 22-25% margin is a 12-13% pre-tax return, matching the measured 12.8% deal yield below. Working capital per £1 of organic revenue, about 17% (58.6 / 342.8), is in line with the 15.6-21.5% working-capital-to-revenue stock, which supports the split. Fixed capital is not a lasting source of cash. Capex below depreciation in four of five years reflects a capital-light distributor, plus property sales (9.9 in FY22) and acquired assets arriving through M&A. Diploma itself expects capex of about 2% of revenue ("usually accounting for around 2% of revenue", FY24 prelim; FY25 capex "at c.1% of revenue was below the 2% we typically expect", FY25 prelim p.10), against D&A of about 1%. At that level the forward fixed rate turns slightly positive, so the negative historical rate should not be extrapolated. Inflection points and who was in charge (Johnny Thomson CEO throughout; CFO Barbara Gibbes to 30 September 2022, Chris Davies from 1 November 2022 to August 2025, Wilson Ng from August 2025): - **FY21, M&A 1.81 (Gibbes).** WCW, funded by the September 2020 placing and a new term loan. - **FY22, working capital 12.7% (Gibbes).** A deliberate inventory build "to support customer service" during supply-chain shortages; working capital to revenue still improved slightly, to 15.6%. - **FY23, working capital released to 2.2% and fixed capital turned positive at 4.2% (Davies's first year).** Inventory fell 10.8 as supply eased, while capex rose to scale Shoal, Hercules Aftermarket and R&G. M&A was 1.36 (DICSA, T.I.E.), funded by the March 2023 placing. - **FY24, M&A 1.91 (Davies).** Peerless, debt-funded. - **FY25, M&A 0.18 (Davies, then Ng).** A deliberate pause ("walked away from a lot of deals") with 46.6 of disposals; ROATCE rose to 20.9%. Deal spend resumed at about 310 in the twelve months to June 2026, including CDM at about 10x. No peer rates were computed; peers' filings were not read for this stage. ## Reinvestment and returns Owner earning power, measured as unlevered after-tax adjusted operating profit (25% tax, a matched definition for acquisitions that were funded by a mix of debt and equity): | | FY20 | FY21 | FY25 | FY26E (normalised) | | --- | ---: | ---: | ---: | ---: | | Adjusted operating profit | 87.1 | 148.7 | 342.7 | 444.7 | | NOPAT at 25% | 65.3 | 111.5 | 257.0 | 333.5 | - **Reinvestment rate FY21-FY25.** Growth investment = acquisitions 1,224.2 + working capital 55.4 + net capex 55.6 = 1,335.2. Matching NOPAT = 903.5. Period-weighted reinvestment = 148%: Diploma reinvested more than it earned, funding the excess with equity and debt. - **Incremental return, lagged.** NOPAT change FY20→FY25 = 191.7, against capital deployed FY20-FY24 (deals 1,209.5 incl. FY20's 14.9, working capital and net capex about 90) ≈ 1,300: about 14.7% after tax. This blends organic growth (which needed little capital), margin expansion, the Peerless windfall and acquisitions. - **Identity check.** NOPAT grew 31.5% a year FY20-FY25, while reinvestment (148%) × incremental return (14.7%) is about 22%. The identity does not hold because the window starts at the Covid trough (FY20 margin 16.2% against 17.8% in FY19), ends with the Peerless windfall (about 17 of FY25 profit), and organic growth on the base earned profit with very little capital. Splitting the uses, the identity reconciles: - **Organic.** Base revenue grew 1.637x organically. At the FY20 margin that is 41.6 of extra NOPAT (87.1 × 0.637 × 0.75); if all margin expansion belongs to the base it is 83.3. Working capital and net capex over the window were about 111 (part of it for acquired businesses), so the organic return on new capital was roughly 37-75%. The forward model uses 0.8 (base), from the current working-capital intensity (21.5% of revenue) and a 24% normalised margin. - **Acquisitions.** Measured separately below: 10.6-15.1% pre-tax, 12.8% midpoint, 9.6% after tax. - **Payouts.** Dividends took 40% of free cash flow and grow 5% a year by policy; there were no buybacks. Diploma's capital-light organic growth is the high-return use; acquisitions earn about their cost of capital on a matched basis (Diploma's own post-tax discount rates are 9.3-9.8%, FY25 note 10), and are worth doing mainly because the acquired niches keep growing on little capital afterwards and because the market capitalises their profit at far more than 10x. ## Acquisition record | Year | Deal(s) | All-in price | Pro forma revenue | Pro forma profit at entry | Entry EBIT multiple | Later evidence | | --- | --- | ---: | ---: | ---: | ---: | --- | | FY21 | Windy City Wire (US cable, Oct 2020) | 347.7 | 175.6 | 41.5 | 8.4x | about 17% a year organic growth since; margin "strong" | | FY21 | S&W (Denmark medtech) | 34.9 incl. deferred | 28.4 | 7.6 | 4.6x | Nordics lost a key supplier in FY25 | | FY21 | Eight others incl. Techsil, FITT, AHW | 85.9 incl. deferred | 60.1 | 10.5 | 8.2x | — | | FY22 | R&G (UK fluid power) | 99.1 | 68.6 | 9.7 | 10.2x | weak UK markets since 2023 | | FY22 | Accuscience (Ireland) and five others | 84.1 | c.60 | c.9.1 | c.9x | — | | FY23 | DICSA (Spain) | 159.7 | 79.5 | 19.2 | 8.3x | new leadership FY25, H2 FY25 growth | | FY23 | T.I.E. (US automation repair) | 75.1 | 28.3 | 6.9 | 10.9x | the failure case: automation downturn, leadership replaced | | FY23 | Ten others | 31.0 | n/d | n/d | n/d | — | | FY24 | Peerless (US aerospace fasteners) | 243.3 | 129.9 (USD107.7m before purchase) | 59.9 FY24; underwritten at c.7x FY24 EBIT (c.34) | c.7x underwritten (4.1x on FY24 pro forma) | profit about 96 in FY25 and about 133 in FY26E (estimates) | | FY24 | PAR and five others | 50.0 | c.25 | c.8.5 | c.6x | — | | FY25 | Alpha Labs, Haagensen, Electramed, Astro, Viking | 56.1 | 46.6 | 8.8 | 6.4x | — | | H1 FY26 | Six deals incl. Swift, WDS, HSA | 84.6 | 59.6 (annualised) | 14.2 (annualised) | 6.0x | — | | Q3 FY26 | CDM (USD170m, c.USD80m revenue) and three others | c.169 (derived) | n/d | c.17 (derived from the issuer's c.40 for 15 deals) | c.10x | — | Aggregate FY21-FY25: about 1,267 of all-in price bought about 156 of entry profit (counting Peerless at its c.34 underwriting, not its windfall-inflated 59.9 pro forma): about 12.3% pre-tax, or 8.1x EBIT. Disposals recycled 46.6 in FY25 at about 6x (nine disposals in five years). Achieved multi-year deal yield, matched definitions (`dealYieldLow/Mid/High`): - Numerator: FY25 adjusted operating profit (342.7) less what the FY20 base would earn after compounding organic growth (1.637x), less the Peerless profit above a 40% margin (17.0). Low bound: base business at the FY25 group margin (198.1), so acquired profit 127.6. High bound: base business at its FY20 margin (142.6), so acquired profit 183.1. - Denominator: all-in deal spend FY21-FY24 plus half of FY25 = 1,209.4, however funded. - Result: 10.6% (low), 15.1% (high), 12.8% midpoint pre-tax; 9.6% after tax, a 10.4x purchase multiple of after-tax cash. The upside uses the high bound (8.8x); the downside uses a 9.5x EBIT entry with no growth after purchase (12.7x). - Limits: currency effects are not separated (FY24-FY25 translation cut revenue 5% in total); the issuer's organic measure includes acquired businesses' growth after their first year, which moves some deal growth into the base and understates the yield; edge-year timing is approximated by half-weighting FY25. Correction (2026-11-03, after the open-book calibration). The trap listed in stage 6 was named but not quantified: Peerless entered the issuer's organic measure in FY25 and its revenue gain over the FY24 pro forma (about 66.5, an estimate from the AR2025 sector mix) is about 4.9 of the FY25 11 points. Removing it from the FY20 base's organic factor (`organicFactorExPeerless`, 1.565 instead of 1.637) moves about 7.5 of FY25 profit from the base to acquisitions: the achieved yield becomes 11.3-15.7% (13.5% midpoint, 10.1% after tax, a 9.9x purchase multiple; `dealYieldMidAdj`). WCW's growth after its first year is still inside the base (no unit data), so this remains a lower bound. Cross-check: the last 15 deals' entry yield is 12.9% (`ltmEntryYield`); 12.9% grown about 4% for the 1.1 spend-weighted years a five-year exit captures is about 13.5%. Achieved (12.8%) is close to entry (12.3%), so in aggregate acquired businesses have earned roughly what was paid for: WCW and DICSA grew, T.I.E., R&G and the Nordics lagged, and Peerless's extra profit is a windfall removed from the numerator. Observed deployment pace: 244.8 a year net of disposals, 250.1 with FY25 gross of disposals (56.1 invested; the model has no disposal proceeds). The record is lumpy: best year FY21 1.8x the average (WCW, equity-funded), worst FY25 0.22x, longest quiet run one year (FY25; FY20 before it) (`dealPaceGross`, `dealBestYearRatio`, `dealWorstYearRatio`, `quietRunYears`); the twelve months to June 2026 ran at about 310 (15 deals). About a third of FY20-FY25 deal money came from equity raises, which the forecast does not repeat. # Thesis and valuation Current assessment:5November2026, after source reconciliation of the revenue-to-owner-cash bridge. Specialist inventory, technical service and approvals help manufacturers reach fragmented customers. Peerless benefits from aerospace fastener demand and scarcity, but temporary pricing cannot be treated as permanent owner earnings. The current yearly margin path already removes part of that windfall while service and inventory investment continue. ## Current valuation Saved quote GBP 70.45, market time 2026-11-04T17:15:00.000Z; fixed valuation date 5 November 2026 and original model exits. These are forward estimates. | Horizon | Bear | Base | Bull | Weighted cash-flow IRR | | --- | ---: | ---: | ---: | ---: | | 5 years | -5.57% | 7.06% | 14.92% | 6.85% | | 10 years | 1.33% | 9.48% | 14.51% | 9.57% | The H1 acquisition note supplies £12.5m missing sales and £2.5m missing operating profit directly. The prior £2.82m calendar approximation is retained only as an audit reference; live after-tax missing cash is now £1.875m. Q3 sales are estimated from CDM’s announced USD80m annual revenue, the fixed vintage FX, an explicit 18% CDM operating margin and 21% for residual deals within the existing £17m aggregate Q3 profit estimate. Only missing ownership months enter; estimated integration cash retains its ramp. ## Pre-mortem FY26 starting sales/profit remain guidance-based estimates until 17 November, not reported annual results. CDM and other Q3 deal margins are judgments and are tested over a range, not issuer disclosures. Minority detail, sustaining inventory, Peerless normalization and principal/supplier relationships remain uncertain. No new perpetual scarcity rent or acquisition synergy is added. The central estimate can fail even when a useful product or good acquisition history survives. Customer demand, supplier pricing, the capital needed for delivery and ownership/financing claims must all work together. The adverse case is a representative failure mechanism, not the largest possible loss. No new report or positive share-price move alone resolves these risks. ## Exit multiple The existing 22× base terminal owner-cash capitalization is retained as a conditional judgment after the independent debate. It needs durable profitable reinvestment after exit; tangible operating returns are not the same as goodwill-inclusive acquisition returns. Neither an unreconciled peer P/E nor the 18% CDM margin estimate establishes the terminal multiple. The revenue bridge does not supply independent proof of terminal value or remove the survival risks identified in the written debate. Shorter duration, poorer cash conversion and higher required returns remain relevant stresses. Historical source reproduction and current underwriting remain distinct. ## Bear case FY26 starting sales/profit remain guidance-based estimates until 17 November, not reported annual results. CDM and other Q3 deal margins are judgments and are tested over a range, not issuer disclosures. Minority detail, sustaining inventory, Peerless normalization and principal/supplier relationships remain uncertain. No new perpetual scarcity rent or acquisition synergy is added. The existing adverse operating and capital-allocation paths remain the coherent modeled stress. Matching sales to cash prevents a missing denominator from hiding those assumptions; it does not turn a proxy margin or assumed refinancing into a reported fact. Debt, deferred claims, operating liquidity and distributions remain funded in the same cash ledger. ## Variant view The market treats FY26's 26.5% margin and roughly 40% EPS growth as evidence that Diploma has become a structurally better compounder, and prices it at about 29x FY26E earnings. This research finds that about a quarter of FY26 profit comes from one business, Peerless, whose margin rose from about 30% to about 54% during an aerospace fastener shortage (stage 2, stage 3 independent checks), and that acquisitions have earned about 10% after tax on a matched basis, close to Diploma's own cost of capital (stage 6). The real engine (organic growth on very little capital) is excellent but grows at about 5% a year, which does not justify 31x normalised owner cash. The stock is a high-quality business priced for windfall earnings to persist. Catalyst: none needed for the view, but the FY26 results and first FY27 guidance on 17 November 2026 are the first test; management has already said FY27 reverts towards its 5% organic model and that margins will moderate. The source bridge correction changes the central estimate by only a small amount at the same price/date. It does not establish a new market mispricing or justify an unsupported advocate sensitivity. The [independent claim judgment](12-debate.md) still sets the evidence burden; the current numerical returns are the table above. ## Why the opportunity exists The following observations retain the measurement dates in their cited sources; no new holder, liquidity or short-interest survey is claimed. This is not a neglected stock. Market capitalisation is about £9.7bn (134.3m shares at 7,265p). Three-month average volume is about 651k shares (£47m a day). Excluding the 5%+ notified holders (Capital Group 12.96%, an LLC holder 5.27%, BlackRock 5.02%), the float of about 103m shares would turn over in about 158 trading days. It is a FTSE 100 member, with at least seven sell-side analysts on the Q3 call (Morgan Stanley, Deutsche Numis, BofA, BNP Paribas, Stifel, Kepler Cheuvreux, Berenberg) and an issuer-cited consensus. Short interest and retail ownership were not checked. The shares rose from a 4,970p low to 7,980p in October 2026 on repeated guidance upgrades (March, May, July), then fell about 9% in late October. So any mispricing is not neglect. It is extrapolation: four upgrades in a year driven by a shortage, priced as a permanent step. The opposite discount (a value opportunity) does not exist at this price; the honest conclusion is that the shares look fully valued to expensive, with a weighted five-year return of about 5%. ## What would prove us wrong FY2026 results on 17 November: replace guidance-based annual sales, disclosed acquisition contributions, Peerless margin/volume and cash conversion, net debt and deferred/minority claims. Compare future observations with the same ownership, cost and reporting periods used in this model. Preserve earlier frozen checks and score them as their reports arrive; do not edit their forecasts to match the result. Unavailable deal disclosure needs a bounded estimate and a revision trigger, not a claim of observed target profitability. [Source bridge and same-price/date changes](../updates/2026-11-05-revenue-bridge-closure.md). The earlier initial completion audit’s three revenue gaps are now closed with reported observations and explicit estimates; the remaining limits above are ordinary underwriting uncertainty, not missing forecast outputs. # Risks & disconfirming evidence ## Paths to permanent loss 1. **Paying for a windfall.** About 27% of FY26E profit comes from Peerless at a margin (about 54%) well above its pre-acquisition level (c.30%). If aerospace fastener supply normalises and manufacturers or primes take back the economics, group profit falls about 12% with no change elsewhere, and the multiple applied to the rest may fall too. The base deducts 34 of pre-tax profit; the downside about 73. 2. **Multiple compression.** At 31x normalised owner cash, the price needs the company's recent high multiples to persist. A move to 22x (still above its historical lows) erases most of five years of earnings growth; 16x produces a loss (bear case -5.9% a year). 3. **Disintermediation.** Distributors earn their margin because manufacturers and principals do not want small customers. Peerless's suppliers in a long market, or a Life Sciences principal opening its own subsidiary (the Nordics lost a key supplier in FY25), can remove a business's economics quickly. Diploma's own risk register names this. 4. **Acquisition discipline at a larger scale.** The achieved deal yield (about 10% after tax) is close to Diploma's cost of capital. Larger deals (CDM at about 2.1x revenue) and PE competition could push returns below it; an equity-funded deal at a high price would dilute holders, as the FY20 and FY23 raises did (+19% shares since FY19). 5. **End-market cycles.** Aerospace build-rate disruption, a US construction downturn (WCW), and UK/Australian industrial weakness (Seals) can coincide. FY20 showed operating profit falling about 10% in a severe shock, which is resilient but not immune. ## Unverified inputs (largest first) | Input | Status | How it gets verified | | --- | --- | --- | | Peerless FY25/FY26 revenue and margin | estimated from sector data (46-49% FY25, 53-61% H1 FY26) | management commentary; any disclosure of Peerless revenue; Controls margin trend | | FY26 operating profit (479.0) and September net debt (348.0) | estimated from guidance and H2 cash conversion | 17 November 2026 results | | Q3 FY26 deals' cost (169.3) and profit (about 17) | derived from issuer totals; CDM's profit undisclosed | FY26 acquisition note | | Organic return on new capital (0.7) | judgment between measured history and current ratios | working-capital intensity and FCF conversion in FY27 | | Exit multiples | judgment from history, peers and Koller | not verifiable; the main return driver | | Acquisition liabilities (39.0) | estimated from FY25 balance plus H1 movements | FY26 balance sheet | ## Disconfirming evidence to watch - Controls margin holding above 32% while Peerless grows about 10% a year (the windfall is structural). - Ex-Peerless organic growth sustained at 8%+ (upside path). - Rising working-capital intensity (inventory above 25% of revenue) without matching growth (the downside's 0.5 organic return). - Deal entry multiples drifting above 10x EBIT or acquired businesses shrinking after purchase (downside deal price). - New senior-finance changes, control discrepancies, restatements or changes to adjusted-profit definitions. The [current management record](03-management.md) distinguishes the issuer's explanation of the 2025 departure from independent assurance. No fieldwork was done, so supplier, customer and employee views are unknown; those are the evidence gaps that would most change confidence in risk 3. ## Surviving debate attacks — 5 November 2026 Low to moderate; valuable approvals/inventory service, unresolved Peerless normalization, current earning estimates and cohort sales. Reconstructed the independent isolated stress at the same quote/date/horizons; do not treat stress size as a probability or a measured current fact. Reconstructed the independent isolated stress at the same quote/date/horizons; do not treat stress size as a probability or a measured current fact. Reconstructed the independent isolated stress at the same quote/date/horizons; do not treat stress size as a probability or a measured current fact. Discriminating evidence: 17 November FY2026: Peerless margin/organic growth, cash conversion, debt and acquisition proforma revenue/profit. The [complete debate](12-debate.md) sizes these risks using the shared forecast; a sensitivity is neither a measured fact nor a calibrated probability. ## Revenue bridge closure and remaining uncertainty — 5 November 2026 The previously missing sales/cash perimeter is now modeled. FY26 starting sales/profit remain guidance-based estimates until 17 November, not reported annual results. CDM and other Q3 deal margins are judgments and are tested over a range, not issuer disclosures. Minority detail, sustaining inventory, Peerless normalization and principal/supplier relationships remain uncertain. No new perpetual scarcity rent or acquisition synergy is added. FY2026 results on 17 November: replace guidance-based annual sales, disclosed acquisition contributions, Peerless margin/volume and cash conversion, net debt and deferred/minority claims. This closes the missing forecast output rather than claiming that the target economics or refinancing outcomes are certain. Earlier source-gap statements in the dated debate describe its initial information set; current inputs are in 07-assumption-register.md. # Current assumption register Assessment:5November2026. Reported facts, issuer pro-forma estimates, normalization and forward judgments remain separately classified. ## Reported starting bridge The H1 acquisition note supplies £12.5m missing sales and £2.5m missing operating profit directly. The prior £2.82m calendar approximation is retained only as an audit reference; live after-tax missing cash is now £1.875m. Q3 sales are estimated from CDM’s announced USD80m annual revenue, the fixed vintage FX, an explicit 18% CDM operating margin and 21% for residual deals within the existing £17m aggregate Q3 profit estimate. Only missing ownership months enter; estimated integration cash retains its ramp. | Input | Value | Classification | Definition/evidence | | --- | ---: | --- | --- | | equityCashEarnings | 341.393 | underwriting estimate | FY26E after-financing owner cash before any Peerless normalisation: (FY26E adjusted operating profit 479.0 - run-rate interest 23.0) x 75% - minorities 0.6 = 341.4, an 18.9% margin on FY26E revenue. Since the revenue build (2026-11-03) the Peerless spot-price windfall fades through the yearly margin path (FY27 47%, FY28 40%) instead of a deduction from the base; the normalised 319.9 remains as a cross-check of the FY28 margin. Sources: dplm-fy25-prelim, dplm-h126, dplm-q326, dplm-ar25, dplm-peerless-rns, yahoo-fx | | shares | 134.317 | reported denominator | 134.317m ordinary shares in issue (AR2025 p.98); the employee trust's 81k shares are ignored. Sources: dplm-ar25 | | openingCash | 0.000 | underwriting policy | Zero investable cash. Diploma runs on its revolving facility; reported cash (68.7 in March) is operating cash inside net debt, which is carried as a claim. Sources: dplm-h126 | | revenue | 1807.675 | derived estimate | FY26E revenue 1,807.7 = FY26E adjusted operating profit / the guided 26.5% FY26 margin (July 2026 update, nine months reported). Replace with the reported FY26 figure on 17 November 2026. Sources: dplm-q326, dplm-fy25-prelim | ## Known and committed cohort bridge Revenue is consolidated gross; cash is on the model’s common-owner or cash-NOPAT perimeter. A reported sales observation is not a reported cash margin. Missing months exclude sales already in the starting baseline. Capital already paid or reserved is not purchased again. | Cohort | Missing annual sales | Missing/mature cash | Timing and basis | | --- | ---: | ---: | --- | | H1 FY26 bolt-ons, missing months | 12.500 | 1.875 | Issuer H1 note 10 reports 12.5m missing sales and £2.5m operating profit directly. Pro-forma numbers extrapolate actual post-close performance; these are estimates of missing ownership months, not newly reported full-year sales. After-tax profit is an owner-cash proxy with depreciation charged; no integration uplift. Add only the periods not already in FY26E. Sources: dplm-h126 | | Q3 FY26 deals incl. CDM, missing months | 67.096 | 9.563 | Estimated from the issuer's c.40 annualised profit on c.310 for the 15 LTM deals; CDM's own profit is undisclosed, so it ramps over a year. Missing revenue is estimated independently from CDM80m USD announced annual sales at fixed FX, explicit 18% CDM operating margin and 21% for the residual four deals, bounded by the existing issuer aggregate profit. Only75% absent ownership months enter; sales and cash retain the one-year ramp. The margin assumptions are not reported target margins. Sources: dplm-h126, dplm-q326, dplm-q326-transcript | ## Forward scenario mechanisms Specialist inventory, technical service and approvals help manufacturers reach fragmented customers. Peerless benefits from aerospace fastener demand and scarcity, but temporary pricing cannot be treated as permanent owner earnings. The current yearly margin path already removes part of that windfall while service and inventory investment continue. The existing 22× base terminal owner-cash capitalization is retained as a conditional judgment after the independent debate. It needs durable profitable reinvestment after exit; tangible operating returns are not the same as goodwill-inclusive acquisition returns. Neither an unreconciled peer P/E nor the 18% CDM margin estimate establishes the terminal multiple. ### downside Aerospace fastener supply catches up: Peerless revenue falls about 20% over FY27-FY28 and its margin returns to about 30% (owner-cash margin 17.3% in FY27, 16.3% after); UK and European industrial markets stay weak, so organic revenue is flat for two years, then grows 2.5%. Deals stop to FY25's 56 for two years (a dry spell longer than the record's), then return to 250, at 9.5x EBIT (2.0x revenue), growing 2.5% after purchase. Exit at 16x, near Diploma's trough trailing multiples (18-21x) less a further step for a slower business. | Driver | Value/path | Reason and sources | | --- | --- | --- | | roiic | 0.500 | Owner cash per unit of new working capital and capex: 0.5. About 13% after-tax margin on new revenue against 25% working capital (as in H1 FY26, when inventory rose 44.1 to secure supply); near the low end of the FY20-FY25 measured 37-75%. | | exitMultiple | 16.000 | 16x after-financing owner cash. Diploma's lowest trailing adjusted P/E in FY2017-FY2025 was 18-21x (FY20 trough about 18x). A slower business with 3% growth and no deal runway justifies about 16-17x at a 9% required return (Koller value driver). | | dilutionRate | 0.000 | No new shares: share-based pay (6.2 in FY25) stays charged in owner cash and awards are bought in the market by the employee trust. Weighted shares rose about 0.1% a year in FY24-H1 FY26. | | reinvestmentRate | 1.000 | All cash after dividends and organic investment is offered to acquisitions, but annual purchases are capped at the observed pace. No new borrowing in the model, although the board allows leverage up to 2x. | | organicGrowth, years 1–2 | 0.000 | Bear: aerospace fastener supply catches up and Peerless revenue falls about 20% over two years, while UK and Australian Seals stay weak; 2-3% growth elsewhere is offset, so group organic revenue is flat in FY27-FY28. Sources: dplm-q326-transcript, dplm-h126, dplm-fy25-prelim | | organicGrowth, years 3–10 | 0.025 | From FY29 2.5%: replacement and consumable demand (about 45% of profit) plus a weaker capex cycle; near FY24's 6% organic year less pricing and the Covid-year pattern. Sources: dplm-q326-transcript, dplm-h126, dplm-fy25-prelim | | organicMargin, years 1–1 | 0.173 | FY27 17.3%: Peerless margin falls to 40% on revenue down 10%; rest of the group held at 22.2% (it kept 19-24% sector margins through FY20). Same interest, tax and minority shares as the base. Sources: dplm-q326-transcript, dplm-h126, dplm-fy25-prelim, dplm-peerless-rns | | organicMargin, years 2–10 | 0.163 | From FY28 16.3%: Peerless at about 30%, its pre-acquisition margin, on revenue 20% lower; operating margin 23.0%. A cyclical reversal, not a permanent group margin collapse. Sources: dplm-q326-transcript, dplm-h126, dplm-fy25-prelim, dplm-peerless-rns | | priceToSales, years 1–10 | 1.995 | 2.0x revenue (priceToSalesDownside): 9.5x EBIT at a 21% margin, the price larger, defence-exposed assets like CDM (about 10-11x) command. Sources: dplm-q326-transcript, dplm-h126, dplm-fy25-prelim, dplm-q326-transcript | | acquiredMargin, years 1–10 | 0.158 | 15.75% owner-cash margin, as in the base: entry margins don't change with the scenario; the downside prices them dearer. Sources: dplm-q326-transcript, dplm-h126, dplm-fy25-prelim | | acquiredGrowth, years 1–10 | 0.025 | 2.5%: the downside rate of the non-Peerless portfolio, under the issuer's organic definition; T.I.E.-type laggards weigh on cohorts. Sources: dplm-q326-transcript, dplm-h126, dplm-fy25-prelim | | deploymentCapacity, years 1–2 | 56.000 | FY27-FY28: a two-year dry spell at FY25's 56, one year longer than the record's longest quiet run (FY25; FY20 before it). An aerospace and industrial downturn plus higher prices for larger assets (CDM about 10-11x) make Diploma walk away, as it did in FY25. Sources: dplm-ar25, dplm-fy25-prelim, dplm-h126, dplm-q326-transcript | | deploymentCapacity, years 3–10 | 250.000 | FY29-FY36: back to the through-cycle average of 250, not grown: a slower business with a smaller deal runway. Sources: dplm-ar25, dplm-fy25-prelim, dplm-h126, dplm-q326-transcript | ### base Revenue build from FY26E revenue (1,807.7) at an 18.9% owner-cash margin. Organic revenue grows 5% in FY27 (management's guide back towards its model), then 5.65% (sector rates weighted by revenue). Peerless's spot-price margin fades from 54% to 47% in FY27 and 40% in FY28 on 9% volume growth, taking the owner-cash margin to 17.6%; mix then lifts it towards 18.0%. Deal budget 250 a year in FY27-FY28 (the FY21-FY25 average), 280 in FY29-FY31 and 320 after; bought at 1.63x revenue (a 21% operating margin at the last 15 deals' 12.9% entry yield) and growing at the non-Peerless rate (4.4%, then 5.1%); about half a year's profit arrives in the purchase year; retained cash saves facility interest; new capital at an 0.6 owner-cash return. Exit at 22x after-financing owner cash. | Driver | Value/path | Reason and sources | | --- | --- | --- | | exitMultiple | 22.000 | 22x after-financing owner cash. Own history: trailing adjusted P/E averaged 25-32x in FY2017-FY2025 (excluding FY21's Covid distortion), with lows of 18-24x. Koller value driver at the exit: 5% growth, half organic at 0.8 and half acquired at about 9.7%, implies reinvesting about 29% of earnings; (1-0.29)/(8.5%-5%) = 20-21x, or 24x at 8%. 22x is below the average history because the history includes years when the market priced Peerless-type windfalls and faster deal growth. | | dilutionRate | 0.000 | No new shares; share-based pay stays charged. No repeat of the FY20/FY23 equity raises is modelled. | | reinvestmentRate | 1.000 | All surplus cash after dividends and organic investment goes to deals, capped at the observed pace. Unplaced cash is retained at face value (in practice it would repay debt). | | organicGrowth, years 2–10 | 0.056 | FY28 onwards: sector rates weighted by FY25 revenue, Controls 7% (Peerless near 10% after FY26, defence, WCW's datacentre and building cable), Seals 4% and Life Sciences 4% (opportunity ledger, 08-underwriting-plan) = 5.65% (organicRevenueGrowthBase). Below the FY21-FY25 record of about 10% a year, which included inflation and Peerless's shortage years. Sources: dplm-q326-transcript, dplm-h126, dplm-fy25-prelim | | organicMargin, years 1–1 | 0.183 | FY27 owner-cash margin 18.3% (ownerMarginY1): Peerless halfway through its spot-price fade, 54% to 47%, on revenue up 9%; the rest of the group at its FY26E 22.2% operating margin; interest held at 1.27% of revenue; 25% tax. Sources: dplm-q326-transcript, dplm-h126, dplm-fy25-prelim, dplm-peerless-rns | | organicMargin, years 7–10 | 0.180 | FY33-FY36 18.0%: the same mix drift continues (Peerless about 18% of revenue by FY36 against 14% today), operating margin about 25.2-25.5%; the average of yearly values of 17.9-18.1%. Still below FY26E's 18.9% owner-cash margin, which contains the spot-price windfall. Sources: dplm-q326-transcript, dplm-h126, dplm-fy25-prelim | | acquiredGrowth, years 1–1 | 0.044 | FY27 4.4% (restGrowthY1): the issuer counts acquired businesses as organic after their first year, and they resemble the non-Peerless portfolio, so they grow at its FY27 rate. Sources: dplm-q326-transcript, dplm-h126, dplm-fy25-prelim | | deploymentCapacity, years 1–2 | 250.140 | FY27-FY28: the through-cycle average of 250 a year (FY21-FY25, FY25 gross; dealPaceGross). Known deals are already in their year: CDM and the other Q3 FY26 deals (about 169) and the H1 FY26 deals (84.6) closed in FY26 and sit in the baseline and funded cohorts, not in this budget. No completed or financed FY27 deal is disclosed before the 17 November results. The LTM pace (about 310) is a recent high and is not extrapolated. Sources: dplm-ar25, dplm-fy25-prelim, dplm-h126, dplm-q326-transcript | | deploymentCapacity, years 3–5 | 280.000 | FY29-FY31: 280 a year. The average scaled with the group: owner cash grows about 5-6% a year, and the pipeline (about 50 live opportunities, typical deal 20-25m) and M&A staff grow with it. A level budget averages the lumpy years; it doesn't predict which year is quiet. Sources: dplm-ar25, dplm-fy25-prelim, dplm-h126, dplm-q326-transcript | ### upside Structural end markets (aerospace build backlog, defence, datacentres, energy) keep organic revenue growth at 7.5%, near the seven-year record; Peerless eases only to 47-50%, so the owner-cash margin stays about 18.4-18.7%. | Driver | Value/path | Reason and sources | | --- | --- | --- | | roiic | 0.900 | 0.9: incremental margins of 29-36% (FY24, FY25) on new revenue against about 22% working capital and capex, near the top of the measured range. | | exitMultiple | 27.000 | 27x, below the FY24-FY25 averages (31-33x trailing). Koller at 6% growth, 8.5-9% required return and a higher organic share gives 26-31x. | | dilutionRate | 0.000 | No new shares. | | reinvestmentRate | 1.000 | All surplus offered to deals, capped. | | organicGrowth, years 1–10 | 0.075 | 7.5% a year: near the seven-year organic record (7%) plus share gains in structural end markets (aerospace build backlog, defence, datacentres); H1 FY26 ex-Peerless growth was high single digit. Peerless about 10%, the rest about 7.1%. Sources: dplm-q326-transcript, dplm-h126, dplm-fy25-prelim | | organicMargin, years 1–1 | 0.185 | FY27 18.5%: Peerless eases only to 50% as volume holds. Sources: dplm-q326-transcript, dplm-h126, dplm-fy25-prelim, dplm-peerless-rns | | organicMargin, years 2–5 | 0.184 | FY28-FY31 18.4%: Peerless holds near 47% (management: structurally above the purchase level); the rest at 22.2%; average of yearly mix values 18.3-18.5%. Sources: dplm-q326-transcript, dplm-h126, dplm-fy25-prelim, dplm-peerless-rns | | organicMargin, years 6–10 | 0.187 | FY32-FY36 18.7%: Controls and Peerless keep gaining share; average of yearly mix values 18.6-18.8%. Sources: dplm-q326-transcript, dplm-h126, dplm-fy25-prelim | | acquiredMargin, years 1–10 | 0.158 | 15.75% owner-cash margin, as in the base. Sources: dplm-q326-transcript, dplm-h126, dplm-fy25-prelim | | acquiredGrowth, years 1–10 | 0.071 | 7.1%: the upside rate of the non-Peerless portfolio (7.5% group less Peerless at 10%). Sources: dplm-q326-transcript, dplm-h126, dplm-fy25-prelim | | deploymentCapacity, years 1–2 | 310.000 | FY27-FY28: the LTM pace (about 310 for 15 deals) holds as the pipeline 'bigger than ever' converts. The engine places only what cash allows, as no borrowing is modelled. Sources: dplm-ar25, dplm-fy25-prelim, dplm-h126, dplm-q326-transcript | | deploymentCapacity, years 3–5 | 380.000 | FY29-FY31: 380 a year, as larger platform deals (about 1.2-1.5x the average, like FY24's 311) recur. Sources: dplm-ar25, dplm-fy25-prelim, dplm-h126, dplm-q326-transcript | | deploymentCapacity, years 6–10 | 450.000 | FY32-FY36: 450 a year, a FY21-sized year on average, matching a group about twice FY21's size. Sources: dplm-ar25, dplm-fy25-prelim, dplm-h126, dplm-q326-transcript | ## Estimation limits and revision triggers FY26 starting sales/profit remain guidance-based estimates until 17 November, not reported annual results. CDM and other Q3 deal margins are judgments and are tested over a range, not issuer disclosures. Minority detail, sustaining inventory, Peerless normalization and principal/supplier relationships remain uncertain. No new perpetual scarcity rent or acquisition synergy is added. FY2026 results on 17 November: replace guidance-based annual sales, disclosed acquisition contributions, Peerless margin/volume and cash conversion, net debt and deferred/minority claims. Named arithmetic, stable fact IDs and source links are retained in valuation/calculations.json. Inputs were selected before inspecting the revised IRR; the earlier source result and prior models are not overwritten. ## 12 November share-plan notice The [original RNS body](https://www.investegate.co.uk/announcement/rns/diploma--dplm/block-listing-application-/9873209) applies to admit 249,058 shares under the 2020 Performance Share Plan, expected 16 November. It is not an exercise, settlement or updated voting-rights notice. `pspBlockListingApplication20261112` records the application; the live share count and existing economic compensation treatment stay unchanged. Reconcile actual issue/treasury movements, buybacks and charged compensation in the next accounts before adding dilution or a second replacement-cost charge. The host’s generated summary is not the original RNS. # Underwriting plan Written with the scenario inputs; returns were run once only to check the engine (see the assumption register for the two corrections made afterwards and their effect). ## Opportunity ledger | Opportunity | Starting point (FY25 / FY26E) | Customer need and why Diploma wins | Reachable revenue in five years | Investment and cash cost | Timing and limits | Evidence against | | --- | --- | --- | --- | --- | --- | --- | | Aerospace fasteners volume (Peerless, Clarendon, Swift) | Peerless c.196 / c.245; Clarendon c.110 | Certified, traceable parts delivered to subcontractors on schedule; a "10+ year backlog" of aircraft | Volume growth near 10% a year at Peerless after FY26 (management) | Inventory: Peerless carried 65.6 of inventory on 130 of revenue at purchase (50%) | Build rates depend on Boeing/Airbus ramps; fastener supply is catching up | Spot premiums fade as supply improves; manufacturers may sell direct | | Aerospace spot **price** (Peerless margin) | about 54% margin FY26E vs c.30% before purchase | Shortage pricing | Negative: base assumes the margin falls to 40% (already deducted) | — | FY27 onward | Management is already "easing off the margin" | | European and US defence (IS Group, Clarendon, Spring, CDM) | 6-7% of group revenue pro forma incl. CDM | Interconnect and fasteners for defence primes; new facility in the Czech Republic | Double-digit growth plausible from a small base | Inventory and facilities; CDM cost USD170m | Defence budgets rising in Europe; US is new (CDM) | Programme delays; defence primes' pricing power | | Datacentre and building cable (WCW, Shoal) | Datacentres c.15% of WCW, c.3% of group | Specified cable delivered to site; WCW grew 17% a year since 2020 | Mid to high single digits through the cycle | Inventory, copper price pass-through | Construction cycle; datacentre build-out may slow | Customer concentration in hyperscale builds; national distributors | | Seals aftermarket and OEM | 456 revenue, 2% organic FY25, 2% H1 FY26 | Repair parts with same-day availability; e-commerce two-thirds of Hercules Aftermarket sales | 3-5% a year if UK and Australia recover | Small | UK construction and agriculture; tariffs | Long UK weakness; R&G underperformance since 2022 | | Life Sciences | 232, 4-6% organic | Local market access for device and diagnostic makers | 3-5% a year | Small; agency renewals | Product-lifecycle refresh in FY26 | Supplier loss (Nordics FY25); principals going direct | | Bolt-on acquisitions | 244.8 a year average; c.310 LTM | Founders sell to a decentralised owner; Diploma walks away from many deals | 1,335 deployed over five years in the base (cash-funded) | All-in price at 9.9x after-tax cash (13.5% pre-tax achieved, corrected 2026-11-03 for Peerless growth counted as organic; 10.4x / 12.8% before) | Capacity grows 5% a year in the base | PE competition; bigger deals at higher multiples (Q3 deals about 10x) | Overlap: acquired businesses' later growth is inside the deal yield, not the organic rate, because the engine holds acquired cohorts flat; only the growth inside the measured FY21-FY25 window is captured (an omitted mechanism, see the 2026-11-03 calibration note). Peerless volume growth is part of the group organic rate; its price windfall is removed once, in the baseline, priced at the revenue Peerless reaches when its margin normalises (FY28). Base organic rate (revised 2026-11-03): weight the ledger's own sector rates by FY25 sector profit instead of adopting the issuer's 5% model. Controls (65% of sector profit) at 7%, Seals (23%) at 4%, Life Sciences (12%) at 4% gives 5.96% (`organicGrowthBase`). Evidence against: management expects FY27 to revert towards the 5% model after FY26's 14% and against tough comparisons; a single ten-year rate cannot show that dip (about -0.1 points a year at ten years). Absolute scale check (base, year 5 ≈ FY31): revenue at the FY26 perimeter is about 1,900 (FY26E about 1,800 plus the Q3 deals' missing months). Growing about 6% organically gives about 2,540 by FY31 (2,425 at the earlier 5%). About 1,335 of deals at about 10x EBIT and a 20% margin add about 670 of revenue. Total revenue is about 3,100, roughly 1.7x FY26E. At a normalised 24% margin that is about 740 of operating profit and about 540 of owner cash after interest and tax, consistent with the model's 514 in year 5 plus the retained cash it holds instead of repaying debt. Diploma had about 3,500 staff in July 2026; at constant revenue per employee it would need about 5,900 by FY31. Management calls this kind of scaling its main focus. The number of deals implied (about 15 a year at FY26 deal sizes) equals the record LTM pace. Management targets as bounds. The financial model (5% organic, 10% total growth, 20%+ margin, double-digit EPS growth, high-teens ROATCE, <2x leverage) is the issuer's medium-term ambition. The base sits about one point above the organic target (5.96%, from sector weights; the target was the base until 2026-11-03), below the 10% total-growth target (because no new debt or equity is used), and at a lower margin than FY26's 26.5% (because of the Peerless normalisation). The PSP's maximum vesting (13% EPS CAGR from FY25) is above the base. ## Scenario mechanisms | | Downside | Base | Upside | | --- | --- | --- | --- | | Peerless | margin to about 30% and spot volume falls; an extra 39 of pre-tax profit lost beyond the base normalisation | margin normalises to 40% by FY28 while revenue grows about 9% a year (excess of 28.7 deducted in the baseline; 34.4 before 2026-11-03) | margin holds nearer 47% (not modelled as a baseline add-back; covered by higher growth) | | Organic revenue growth (yearly path) | 0% FY27-FY28, then 2.5% | 5% FY27, then 5.65% | 7.5% | | Owner-cash margin (yearly path, from 18.9%) | 17.3% FY27, 16.3% after (Peerless to 30%) | 18.3% FY27, 17.6% FY28, 17.75% FY29-32, 18.0% FY33-36 | 18.5% FY27, 18.4% FY28-31, 18.7% FY32-36 | | Organic return on new capital | 0.5 (inventory-heavy growth) | 0.6 (0.7 before the 2026-11-03 claim reconstruction) | 0.9 (operating leverage) | | Deal budget (yearly path) | 56 in FY27-FY28 (dry spell), then 250 | 250 FY27-FY28, 280 FY29-FY31, 320 FY32-FY36 | 310 FY27-FY28, 380 FY29-FY31, 450 FY32-FY36 | | Financing and dilution | none | none | none | | Exit (after-financing owner cash) | 16x | 22x | 27x | Mechanisms behind each assumption: - **Downside.** The windfall reverses faster and further than the base (a supply-driven premium, with management already cutting spot prices), and Seals' UK/Australian weakness persists. That is a temporary under-use of a still-good franchise, not a permanent impairment: the group held 16%+ margins through Covid. The lower exit multiple comes from slower growth and a smaller deal runway, not from the lower earnings themselves. Deals continue because the board has kept buying through weak years (FY23-FY24). - **Base.** The rest of the portfolio grows at the profit-weighted sector rate (5.96%); the issuer's own 5% model, beaten in five of the last six years, is now a bound rather than the forecast; deals continue at the measured achieved yield and the observed pace. - **Upside.** Requires the structural end markets to keep ex-Peerless growth near high single digits (as in H1 FY26) and the deal engine to scale. The higher multiple needs its own mechanism: more organic growth at higher incremental returns raises the Koller value-driver multiple (6% growth at 8.5-9% gives 26-31x). It is not granted by growth alone. Funding: every scenario funds deals and organic investment from owner cash after dividends before any contribution, with a one-year delay; no scenario needs external finance (the engine would reject it). Adapter choice: `equity-acquirer`, because Diploma's owner cash after interest is the right claim, it pays a fixed progressive dividend, organic investment needs a return-on-capital charge, and deal capacity must be capped. The debt-funded `capital-acquirer` adapter was rejected because it forces zero organic growth and deploys to a debt ceiling every year without a capacity cap; neither fits Diploma. The cost of the choice: the model cannot borrow up to the board's 2x policy or issue shares, so it understates the deal capacity Diploma has actually used. ## Revenue build (3 November 2026) Owner cash is now revenue x margin on yearly paths, each with a written reason in `valuation/model.json` (`inputs.yearly`), computed in `calculations.json` (`fy26Revenue`, `startingOwnerMargin`, `ownerMarginY1`, `ownerMarginY2`, `organicRevenueGrowthBase`, `restGrowthY1/Y2`, `priceToSalesBase`). Starting point: FY26E revenue 1,807.7 and owner cash 341.4, an 18.9% margin that still contains the Peerless windfall. The fade is a margin path instead of a deduction. Base revenue reaches about 3,110 organic plus about 2,120 acquired by FY36 (about 5,230 in total, 2.9x FY26E), at a 17.1% blended owner-cash margin, because acquired businesses earn 15.75%. ## Revenue bridge closure — 5 November 2026 The H1 acquisition note supplies £12.5m missing sales and £2.5m missing operating profit directly. The prior £2.82m calendar approximation is retained only as an audit reference; live after-tax missing cash is now £1.875m. Q3 sales are estimated from CDM’s announced USD80m annual revenue, the fixed vintage FX, an explicit 18% CDM operating margin and 21% for residual deals within the existing £17m aggregate Q3 profit estimate. Only missing ownership months enter; estimated integration cash retains its ramp. 5-year base revenue 3221.658, operating owner cash 553.600 (GBPm). 10-year base revenue 5234.784, operating owner cash 894.324 (GBPm). The original customer/opportunity and acquisition-budget reasoning remains; this correction does not infer new demand from a matched denominator. FY26 starting sales/profit remain guidance-based estimates until 17 November, not reported annual results. CDM and other Q3 deal margins are judgments and are tested over a range, not issuer disclosures. Minority detail, sustaining inventory, Peerless normalization and principal/supplier relationships remain uncertain. No new perpetual scarcity rent or acquisition synergy is added. [Source arithmetic and old/new inputs](../updates/2026-11-05-revenue-bridge-closure.json). # Capital productivity and reinvestment Assessment:5November2026. These are shared-engine diagnostics from the same owner-cash projection shown in the dashboard, not a new historical measurement. Money is in GBP millions. | Horizon | Scenario | Return/proxy definition | Rate | Period-weighted reinvestment | New growth capital | Matching earnings | Funding residual | | --- | --- | --- | ---: | ---: | ---: | ---: | ---: | | 5 years | downside | Acquisition ROIIC · mature cash proxy | 7.89% | 51.68% | 909.371 | 1759.568 | 2.27e-13 | | 5 years | base | Acquisition ROIIC · mature cash proxy | 9.68% | 66.06% | 1504.750 | 2277.979 | 0.00e+0 | | 5 years | upside | Acquisition ROIIC · mature cash proxy | 10.00% | 73.15% | 1861.621 | 2544.862 | 4.55e-13 | | 10 years | downside | Acquisition ROIIC · mature cash proxy | 7.89% | 53.36% | 2265.756 | 4246.192 | 4.55e-13 | | 10 years | base | Acquisition ROIIC · mature cash proxy | 9.68% | 55.17% | 3406.122 | 6173.603 | 9.09e-13 | | 10 years | upside | Acquisition ROIIC · mature cash proxy | 10.00% | 59.60% | 4447.275 | 7462.098 | 0.00e+0 | The H1 acquisition note supplies £12.5m missing sales and £2.5m missing operating profit directly. The prior £2.82m calendar approximation is retained only as an audit reference; live after-tax missing cash is now £1.875m. Q3 sales are estimated from CDM’s announced USD80m annual revenue, the fixed vintage FX, an explicit 18% CDM operating margin and 21% for residual deals within the existing £17m aggregate Q3 profit estimate. Only missing ownership months enter; estimated integration cash retains its ramp. Reinvestment is total new growth investment divided by total matching earnings, not an average of annual percentages. Previously funded cohort contributions are not new investment returns. Acquisition cash-yield proxies and forecast-implied productivity are not measured historical ROIIC or project IRRs. All dividends, existing claims, liquidity and financing are reconciled before new purchase capital. FY26 starting sales/profit remain guidance-based estimates until 17 November, not reported annual results. CDM and other Q3 deal margins are judgments and are tested over a range, not issuer disclosures. Minority detail, sustaining inventory, Peerless normalization and principal/supplier relationships remain uncertain. No new perpetual scarcity rent or acquisition synergy is added. Historical measurements and definitions remain in 04-capital-allocation.md and valuation/calculations.json. [Dated source bridge and funding ledger](../updates/2026-11-05-revenue-bridge-closure.json). # Scenario probabilities Assessed 3 November 2026. Coarse, subjective weights on three joint business paths, defined in `research/08-underwriting-plan.md` before the weights were chosen. They are not fitted to any return, issuer guidance or external forecast; no investor reference was consulted (blind test). Stored in `valuation/model.json` → `scenarioProbabilities`. | Horizon | Downside | Base | Upside | | --- | ---: | ---: | ---: | | 5 years | 25% | 50% | 25% | | 10 years | 25% | 50% | 25% | **Five years.** The window spans the likely Peerless normalisation and at least one industrial cycle. Downside 25%: supply-driven spot premiums usually fade once manufacturers add capacity; Peerless already earns more than 50% against c.30% before purchase, management is already trimming spot prices, and Seals' UK weakness persists. Base 50%: diversified niches, 7% organic growth over seven years and stable 19-24% sector margins support profit-weighted sector growth of about 6% (5% before the 2026-11-03 revision, which had used the issuer's financial model) with deals at the measured yield. Upside 25%: aerospace backlog, defence, datacentres and energy, and high-single-digit ex-Peerless growth in H1 FY26, make sustained 8% growth and faster deal capacity plausible but not central. **Ten years.** The deal record and organic franchise matter more than the Peerless cycle, but there are more cycles and a growing risk that deal capacity cannot keep growing at the measured yield as the group gets bigger. There is no evidence-based reason to tilt the weights away from the five-year view, so they are unchanged; this is a judgment, not a default. Weighted returns at 7,265p (IRR of probability-weighted dated cash flows): 6.2% (5 years), 9.2% (10 years), after the 2026-11-03 calibration revision, engine-policy update, claim reconstruction, revenue build and the 4 November yearly deal budgets (blind version 4.9% and 8.0%). The weights were re-affirmed, not changed: the revision corrected a debt definition, the Peerless normalisation revenue, the base growth rule, the deal-yield measurement and (with new engine policies) acquired-cohort growth, part-year deal profit and a return on retained cash, but each path keeps its mechanisms. Limits: low confidence; the sources support the mechanisms, not the odds. Three paths approximate a wider distribution; a deep aerospace or industrial recession with a de-rating below 16x, or a large equity-funded deal, lies outside them. The weights are joint path weights, not independent driver probabilities. Price moves alone do not change them; the FY26 results (17 November 2026) and Peerless's FY27 trajectory are the next evidence. A local override of any scenario input suspends these weights until reviewed. ## Source bridge review — 5 November 2026 The revenue-to-cash coverage gap is closed. The H1 acquisition note supplies £12.5m missing sales and £2.5m missing operating profit directly. The prior £2.82m calendar approximation is retained only as an audit reference; live after-tax missing cash is now £1.875m. Q3 sales are estimated from CDM’s announced USD80m annual revenue, the fixed vintage FX, an explicit 18% CDM operating margin and 21% for residual deals within the existing £17m aggregate Q3 profit estimate. Only missing ownership months enter; estimated integration cash retains its ramp. Existing horizon-specific weights were reassessed and retained: the core joint demand/capital scenarios have not been replaced and exact numerical frequency evidence has not appeared. FY26 starting sales/profit remain guidance-based estimates until 17 November, not reported annual results. CDM and other Q3 deal margins are judgments and are tested over a range, not issuer disclosures. Minority detail, sustaining inventory, Peerless normalization and principal/supplier relationships remain uncertain. No new perpetual scarcity rent or acquisition synergy is added. These are subjective estimates, not measured odds. # Bull–bear review — Diploma PLC, 5 November 2026 Subsequent source closure: the missing revenue bridge identified in this debate has now been completed. [Current source-derived estimates, corrections and same-price/date effects](../updates/2026-11-05-revenue-bridge-closure.md). The original independent argument record below is retained with its information set; it is not a claim that all estimated target margins have become reported facts. Independent advocates prepared opening cases without seeing the other draft, then exchanged point-by-point cross-examinations. A separate judge who did not author the prior model assessed the competing claims. Source reconstructions retain their stated provenance; inaccessible private reports were not newly read. Quoted sensitivities test arguments and are not automatically adopted forecasts. Fixed quote GBP 70.45, market time 2026-11-04T17:15:00.000Z; valuation date 2026-11-05. Original forecast exits retained. | Horizon | Before bear/base/bull | After bear/base/bull | Weighted before → after | | --- | --- | --- | --- | | 5 years | -5.56% / 7.07% / 14.94% | -5.56% / 7.07% / 14.94% | 6.86% → 6.86% | | 10 years | 1.34% / 9.48% / 14.51% | 1.34% / 9.48% / 14.51% | 9.57% → 9.57% | ## Bull case As of 5 November 2026. No opposing advocate draft was read. Existing company research, fact/calculation ledgers and exact registered source locators were examined; original issuer sources additionally checked live: Lumine Q3, CSU Q2, Diploma Q3 transcript, NVIDIA CFO commentary, TerraVest Highland/KBK, WOSG FY26. Harvia Q3 PDF retrieval failed; its registered extraction remains the evidence limit. Private source interpretations below are the repository's documented reconstruction, not a new claim to have read an unavailable private original. The strongest bear case is paying a quality multiple for a temporary aerospace fastener shortage, while new acquisitions earn only around the cost of capital. Peerless's undisclosed present margin and revenue make group quality insufficient proof of durable current earning power. **DPLM-B1 — Organic demand may beat the subdued base.** The issuer's acquired Peerless had approximately 9% long-term organic growth before the shortage; aerospace production backlogs, defence and datacentre cable provide identifiable demand. Test the existing upside 7.5% yearly organic path, holding base margins and deals. Sources `dplm-peerless-rns`, pp.1–2; `dplm-q326-transcript`, pp.1–3; `dplm-h126`, sector reviews pp.7–11. H1 ex-Peerless growth supports the possibility, but the outside view must exclude inflation and shortage years; datacentres are only~3% of the group. **DPLM-B2 — Some Peerless uplift may survive.** Specialist inventory, approvals and service availability are customer value, not merely spot prices. Management says the eventual margin should be structurally above acquisition, below today (`dplm-q326-transcript`, PDF p.5). Test upside `organicMargin` only. `dplm-peerless-rns`, pp.1–2 gives~30% acquisition EBIT margin; the current~54% is reconstructed, not disclosed. Retaining stronger margin belongs in upside until actual disclosure resolves it. Test27× versus 22× as duration exposure, not an adopted multiple.22× needs about 14 years after exit at 10% growth, 20% blended marginal return and 8.5% required return in a finite-growth diagnostic. The advantage is service/approval density across small niches; principal loss, competition and larger deals can shorten it. A permanent high organic ROIIC cannot be transferred to acquisitions. Lack of funded-cohort sales and FY26 results not due until 17 November prevent high conviction. ### Isolated shared-engine effects Saved quote 70.45; market timestamp 2026-11-04T17:15:00.000Z. Fixed valuation date 2026-11-05; model ff 31833da2d3; exits remain the original model vintage. These are sensitivities, not joint forecasts or probability revisions. Baseline base IRR 7.07% / 9.48% (5/10 years). | Claim | Exact input/path | 5-year base | 10-year base | Change, pp (5/10) | | --- | --- | ---: | ---: | --- | | DPLM-B1 | `cases.base.inputs.yearly.organicGrowth` | 8.58% | 10.77% | +1.51 / +1.29 | | DPLM-B2 | `cases.base.inputs.yearly.organicMargin` | 7.73% | 9.79% | +0.66 / +0.31 | | DPLM-B3 | `cases.base.inputs.exitMultiple` | 11.39% | 11.52% | +4.31 / +2.04 | ## Bear case ### Strongest opposing case Diploma's service, approvals and inventory allow manufacturers to reach small customers economically, producing high tangible capital returns. Ex-Peerless organic growth was 10% at the Q3 update, and management describes high-single-digit to 10% longer-run Peerless volume growth, so attributing all FY26 strength to one temporary shortage would be wrong. [dplm-q326-transcript PDF pp.2–4.] ### Bear case The stock still prices a high-quality distributor as though unusually favorable aerospace pricing has become durable earning power. Peerless's estimated 53–61% first-half margin compares with about 30% before purchase. The estimate is reconstructed from Controls disclosure, not a disclosed Peerless P&L, so neither bull nor bear should pretend to know its exact margin. Management explicitly expects growth to moderate, is reducing margins and expects FY27 to return toward the usual financial model. That is stronger evidence than price momentum or the reputation of the acquirer. [dplm-peerless-rns pp.1–2; dplm-ar25 pp.22–23; dplm-h126 sector tables; dplm-q326-transcript pp.2–4.] The current base already normalizes some of this windfall through an annual margin path. The challenge is whether 17.75–18.03% owner-cash margins and 5.65% organic revenue growth sufficiently allow for lost supplier scarcity and inventory funding. Peerless's purchase inventory was roughly half annual sales. If primes or fastener manufacturers reclaim the spread as supply normalizes, replacement demand protects volume but cannot protect the distributor's price. The Nordics lost a Life Sciences principal: supplier relationships are valuable while suppliers want them, not owned in perpetuity. [dplm-fy24-prelim acquisition note 11; dplm-fy25-prelim Life Sciences review; research/01-business.md representative units.] The matched deal yield is around 10% after tax, much less than the return on tangible operating capital. Paying for goodwill and funding inventory matters. A22x base exit can be rational only with further profitable growth beyond 2031. The present Koller check uses perpetual 5–6% growth and 8–8.5% required returns; it does not establish a finite duration of the advantage or a matched peer cash denominator. The low current forward return leaves little cushion if the duration shortens. [research/04-capital-allocation.md matched deal arithmetic; dplm-fy21-prelim through dplm-fy25-prelim acquisition notes; research/05-thesis.md Exit multiple.] ### Contested claims - DPLM-B1 —Test16x exit against 22x base; 16x is an existing downside stress below the Covid-era observed P/E range, not a sourced forecast of owner-cash multiples. - DPLM-B2 —Test16.28% owner-cash margin from year 2, the present downside's normalization, instead of later recovery to 18.03%. Does inventory service create durable economics once scarcity rents fade? - DPLM-B3 —Test acquisition revenue price 1.995x against 1.6275x with acquired cash margin 15.75% unchanged. This isolates seller competition/price from target quality. The issuer's Q3 transcript was reopened for this review. FY26 results are due 17 November 2026; FY26 profit, September debt, Peerless margins and deal earnings remain estimates until then. The [11 November correction](../updates/2026-11-11-evening-review.md) removes the false claim that the annual report left the finance departure unexplained; the financial verdict is unchanged. ### Shared-engine effects Fixed saved quote GBP70.45; market timestamp 2026-11-04T17:15:00.000Z; valuation date 5 November 2026; fixed original exit calendar. Base before: 7.07% / 9.48% (5/10years). These are one-change sensitivities to the base, except named bundles; existing scenario probabilities are not revised or claimed as validated. | Claim | Change | 5-year base IRR | 10-year base IRR | 5-year effect | 10-year effect | | --- | --- | ---: | ---: | ---: | ---: | | DPLM-B1 | Exit16x | 0.75% | 6.46% | -6.32pp | -3.02pp | | DPLM-B2 | Owner cash margin 16.28% from year 2 | 5.57% | 8.69% | -1.50pp | -0.80pp | | DPLM-B3 | Acquisition price/revenue1.995x | 6.05% | 8.64% | -1.03pp | -0.84pp | ## Rebuttals ### Bull response 5 November 2026. Written after independently frozen openings were exchanged. Every numbered bear claim is addressed below; original bull openings and sensitivities remain unchanged. 1. **Bear16× versus 22×: concede duration; reject arbitrary central 16×.** A high tangible ROIC cannot by itself capitalise bought goodwill at the same return. My22× diagnostic's20% marginal return is illustrative blended productivity, not measured group new-money RONIC. Organic and acquisition capital must be separated. `dplm-h126`p5 and acquisition notes support acquired EBIT yields~13% before tax, while stage 6 after-tax deal cash is~10%. Both22× and 16× require explicit forward economics; historical P/E is not the same denominator. 2. **Bear16.28% owner-cash margin: partly rebut, partly concede.** The issuer explicitly expects Peerless margin above acquisition level (`dplm-q326-transcript`, PDFp5), not a return to~30%. The present base already assumes 40% Peerless and a fall in owner-cash margin. But the exact starting~54% and eventual 40% are reconstructions, not disclosed target accounts. My stronger margin path belongs in upside; no evidence supports promoting 47–54% permanently. Bear16.28% is an adverse combination, not a newly observed normalization. 4. **Bull7.5% organic attacked: concede no complete group bridge.** Datacentres~3% cannot alone generate group 7.5%; issuer ex-Peerless10% in the Q3 call is a useful current observation, not future proof.7.5% upside requires Controls/aerospace/defence and the rest of portfolio to persist together. Retain5→5.65% central until a full unit/customer bridge supports more. FY26 results 17 November must replace estimates. Conviction moderate on the business, low on a cheap stock at the current quote. ### Bear response 5 November 2026, after independent openings were exchanged. 1. **Bull7.5% organic: possibility conceded, central upgrade rejected.** Issuer Q3 ex-Peerless organic 10% is a favorable observation, and certified inventory service can grow. Datacentres at 3% of group cannot alone explain 7.5% sustained group growth; aerospace, defence, Seals and Life Sciences must jointly contribute. The bull concedes the complete bridge is missing. Retain the current 5→5.65% base rather than projecting one exceptional year (`dplm-q326-transcript`, PDFpp.2–4; sector ledger in 08-underwriting-plan). 2. **Bull higher structural Peerless margin: direction conceded, exact level unresolved.** Management expects margin above the~30% acquisition level, supporting some structural improvement (`dplm-q326-transcript`, PDFp.5; `dplm-peerless-rns`, pp.1–2). It does not establish 47–54% forever. I concede the existing base already removes part of the windfall and that 16.28% is an adverse group margin, not an observed central level. Supplier economics and principal loss remain live counterarguments. 3. **Bull22/27x exit: blended productivity needs a denominator.**20% blended marginal return cannot be inferred from 60% tangible organic returns when bought goodwill earns~10%. State the mix of new capital, not the share of reported growth (`dplm-fy25-prelim` funds flow/acquisition notes; matched stage 6 ledger). The bull concedes the 20% was illustrative. Likewise I concede 16x is a stress and not an independently derived central terminal value. The agreed current cash denominator differs from statutory P/E peers. 4. **Bear deal-price stress: bounded, not central.** CDM~2.1x revenue is an observed expensive deal; LTM profit 40m/price310m and the multi-deal record support differentiated targets. I concede one transaction does not establish 1.995x for the whole future deal population. Retain it as competition/mix stress (`dplm-h126`, p.5; `dplm-q326-transcript`, p.2; acquisition notes). No new numerical central input follows from this debate before 17 November FY26 results. The judgment should retain separate windfall and volume mechanisms, avoid double normalization, lower conviction in exact margins and publish the negative 16x stress alongside the favorable organic case. Main attacks on price/value survive without inventing a harsher base. ## Judgment Existing estimates receive no presumption merely because they are already in the model. Positive company evidence, the relevant outside view and the capital/funding bridge determine each verdict; neither arbitrary conservatism nor an unsupported favorable sensitivity replaces it. The evidence and reconstruction below explain numerical changes or explicit no-change conclusions claim by claim. | Contested claim | Verdict | Evidence/locator | Reason and model action | | --- | --- | --- | --- | | bull advocate: DPLM-B1: cases.base.inputs.yearly.organicGrowth | reject | dplm-q326-transcript PDF pp.2–5; dplm-peerless-rns pp.1–2; dplm-h126 sector tables/note10; dplm-fy25-prelim Life Sciences and acquisition notes. Ex-Peerless growth and aerospace backlog support base5.65%; 7.5% remains upside because some revenue carries shortage/price effects. | Reconstructed the isolated proposed input/path from independent opening and shared engine, at frozen saved quote/date/horizons; source mechanism separated from unsupported numerical extrapolation. | | bull advocate: DPLM-B2: cases.base.inputs.yearly.organicMargin | reject | dplm-q326-transcript PDF pp.2–5; dplm-peerless-rns pp.1–2; dplm-h126 sector tables/note10; dplm-fy25-prelim Life Sciences and acquisition notes. Management expects margin above entry and below current; central partial normalization has direct support. Reconstructed54% target margin is not disclosed. | Reconstructed the isolated proposed input/path from independent opening and shared engine, at frozen saved quote/date/horizons; source mechanism separated from unsupported numerical extrapolation. | | bull advocate: DPLM-B3: cases.base.inputs.exitMultiple | reject | dplm-q326-transcript PDF pp.2–5; dplm-peerless-rns pp.1–2; dplm-h126 sector tables/note10; dplm-fy25-prelim Life Sciences and acquisition notes. Low organic tangible-capital needs and service approvals support a conditional premium; 22x requires roughly 14 post-exit years at 10%growth/20%blendedreturn/8.5%requiredreturn.27x lacks independent runway support. | Reconstructed the isolated proposed input/path from independent opening and shared engine, at frozen saved quote/date/horizons; source mechanism separated from unsupported numerical extrapolation. | | bear advocate: DPLM-B1: Exit16x | reject | dplm-q326-transcript PDF pp.2–5; dplm-peerless-rns pp.1–2; dplm-h126 sector tables/note10; dplm-fy25-prelim Life Sciences and acquisition notes. 16x is a coherent short-duration sensitivity, but approval/service economics and management-led normalisation support retaining conditional 22x, with duration unresolved rather than asserted as fact. | Reconstructed the independent isolated stress at the same quote/date/horizons; do not treat stress size as a probability or a measured current fact. | | bear advocate: DPLM-B2: Owner cash margin 16.28% from year 2 | reject | dplm-q326-transcript PDF pp.2–5; dplm-peerless-rns pp.1–2; dplm-h126 sector tables/note10; dplm-fy25-prelim Life Sciences and acquisition notes. Base annual margin already gives back scarcity rents; management says margin should remain above acquisition, supporting partial rather than full margin surrender.16.28% stays adverse. | Reconstructed the independent isolated stress at the same quote/date/horizons; do not treat stress size as a probability or a measured current fact. | | bear advocate: DPLM-B3: Acquisition price/revenue1.995x | reject | dplm-q326-transcript PDF pp.2–5; dplm-peerless-rns pp.1–2; dplm-h126 sector tables/note10; dplm-fy25-prelim Life Sciences and acquisition notes. Higher prices may reduce future returns; matched deal population supports 1.6275x currently, not universal 1.995x. Keep seller-price stress visible. | Reconstructed the independent isolated stress at the same quote/date/horizons; do not treat stress size as a probability or a measured current fact. | Retain existing coarse horizon-specific weights after reassessment: scenarios already represent these adverse/favorable operating mechanisms. Numerical odds are subjective, not verified frequencies; unresolved gaps prevent high conviction. No tuning to return. ## Conviction Low to moderate; valuable approvals/inventory service, unresolved Peerless normalization, current earning estimates and cohort sales. These are forward estimates with surviving attacks, not confirmed outcomes. Neither numerical calibration, a completed checklist nor passing software tests establishes investment skill. Reported facts stay separate from forward judgments, and adverse paths are not worst-case bounds. **What would change the view:** 17 November FY2026: Peerless margin/organic growth, cash conversion, debt and acquisition proforma revenue/profit. [Exact inputs, isolated sensitivities, model hashes and same-price/date before/after returns](../updates/2026-11-05-debate-review.json). ## Public Primary & Secondary Sources - `dplm-fy25-prelim` (primary): Diploma FY2025 preliminary results (year to 30 September 2025) — https://www.diplomaplc.com/media/ejrpfglw/diploma-plc-fy25-preliminary-results-final-1.pdf (published 2025-11-18) - `dplm-ar25` (primary): Diploma Annual Report and Accounts 2025 — https://www.diplomaplc.com/media/5s3fi0yk/diploma-plc-annual-report-and-accounts-2025.pdf (published 2025-12-08) - `dplm-h126` (primary): Diploma half year results, six months to 31 March 2026 — https://www.diplomaplc.com/media/5m0fmngw/diploma-plc-half-year-results-2026.pdf (published 2026-05-19) - `dplm-h126-slides` (primary): Diploma H1 2026 results presentation — https://www.diplomaplc.com/media/gvgpf035/diploma-plc-h126-slides.pdf (published 2026-05-19) - `dplm-fy25-slides` (primary): Diploma FY2025 results presentation — https://www.diplomaplc.com/media/crxjpp0u/diploma-plc-fy25-slides-final.pdf (published 2025-11-18) - `dplm-q126` (primary): Diploma Q1 2026 trading update — https://www.diplomaplc.com/media/3wbn0p3a/diploma-plc-q1-2026-trading-update.pdf (published 2026-01-14) - `dplm-tu-mar26` (primary): Diploma trading update: significant upgrade to FY26 guidance — https://www.diplomaplc.com/media/2badvq4n/diploma-trading-update-180326-final.pdf (published 2026-03-18) - `dplm-q326` (primary): Diploma Q3 2026 trading update (RNS) — https://www.investegate.co.uk/announcement/rns/diploma--dplm/q3-trading-update-/9671242 (published 2026-07-16) - `dplm-q326-transcript` (primary): Diploma Q3 2026 trading update call transcript — https://www.diplomaplc.com/media/ym1ao0mc/diploma-q3-transcript-160726.pdf (published 2026-07-16) - `dplm-fy24-prelim` (primary): Diploma FY2024 preliminary results — https://www.diplomaplc.com/media/1aajhnat/fy24-full-year-results.pdf (published 2024-11-19) - `dplm-ar24` (primary): Diploma Annual Report 2024 — https://www.diplomaplc.com/media/vdyeuqza/diploma-2024-annual-report.pdf (published 2024-12-05) - `dplm-fy23-prelim` (primary): Diploma FY2023 preliminary results — https://www.diplomaplc.com/media/2beds2ll/diploma-plc-preliminary-results-for-year-ended-30-september-2023.pdf (published 2023-11-20) - `dplm-fy22-prelim` (primary): Diploma FY2022 preliminary results — https://www.diplomaplc.com/media/zwlfmexp/diploma-plc-fy22-prelim.pdf (published 2022-11-21) - `dplm-fy21-prelim` (primary): Diploma FY2021 preliminary results — https://www.diplomaplc.com/media/rgzlescm/rns-fy-2021-final.pdf (published 2021-11-22) - `dplm-ar20` (primary): Diploma Annual Report 2020 — https://www.diplomaplc.com/media/dgqhbpca/diploma_ar20_full.pdf (published 2020-11-27) - `dplm-peerless-rns` (primary): Diploma: acquisition of Peerless Aerospace Fastener LLC — https://www.diplomaplc.com/media/0dbbx2ui/acquisition-of-peerless-diploma-plc.pdf (published 2024-03-27) - `dplm-calendar` (primary): Diploma financial calendar — https://www.diplomaplc.com/investors/financial-calendar/ (published 2026-11-03) - `dplm-ir` (primary): Diploma results, reports and presentations index — https://www.diplomaplc.com/investors/results-reports-and-presentations/ (published 2026-11-03) - `dplm-pdmr-jan26` (secondary): Director/PDMR shareholding notice: CEO sale, 16 January 2026 — https://www.ticker.app/lse/DPLM/rns/2026-01-16/DirectorPDMR-Shareholding/2954P (published 2026-01-16) - `yahoo-dplm` (market-data): Yahoo Finance market data for DPLM.L — https://finance.yahoo.com/quote/DPLM.L/ (published 2026-11-03) - `yahoo-fx` (market-data): Yahoo Finance GBP exchange-rate history — https://finance.yahoo.com/quote/GBPUSD%3DX/ (published 2026-11-03) - `yahoo-peers` (market-data): Yahoo Finance peer quotes (trailing statutory P/E), 2 November 2026 — https://finance.yahoo.com/ (published 2026-11-03) - `rs-h1-fy2027` (primary): RS Group half-year results to30September2026 — https://www.rsgroup.com/media/b5oltgec/h1-fy27-rns-final.pdf (published 2026-11-05) - `spirax-trading-2026-11-11` (competitor-primary): Spirax Group trading update for four months to 31 October 2026 — https://content.spiraxgroup.com/-/media/engineering/documents/results-and-agm-notices/2026/results/trading-update-11-november-2026.ashx?rev=8452eb586ecd4c00ae5074c87f74d4e9&hash=5AB450C79F02339E9EE4E10D0B943F51 (published 2026-11-11) - `dplm-block-listing-2026-11-12` (primary): Diploma12November block-listing application — https://www.investegate.co.uk/announcement/rns/diploma--dplm/block-listing-application-/9873209 (published 2026-11-12) --- # Harvia (HARVIA.HE) - Legal Name: Harvia Oyj - Exchange: Helsinki (Finland) - Sector: Sauna & wellness equipment - Currency: EUR - Reviewed: 2026-11-15 - Model Framework: Organic growth and reinvestment - Financial Period: TTM to September 2026; September financial position plus known October dividend; fixed October 2026 forecast vintage (anchored 2026-10-26) - Research Status: 8 of 11 research checks complete. Still to complete: Security and evidence base; Opportunities and funded scenarios; Variant view and catalysts - Static Report URL: https://longview-investing.vercel.app/research/harvia.html - Interactive Model URL: https://longview-investing.vercel.app/#/company/harvia - JSON API: https://longview-investing.vercel.app/api/companies/harvia ## Executive Summary Replacement demand, distribution gains and high returns on organic investment drive compounding. Growth must fund working capital and capacity before cash reaches shareholders. ## Key Operational Watch Items - Organic growth by geography - Margin recovery and capacity utilization - Maintenance, leases and working capital ## Valuation Scenario Assumptions ### Downside Case Adverse adoption/destocking with 0% initial growth recovering toward 5%, no recovery from current 22.68% EBITDA margin, and unchanged measured capital per sales euro. Terminal11.89x is a finite 15-year/3%nominal growth/10%required-return judgment, followed by no-growth value; no claim that this is a loss bound. - Key Drivers: Near-term organic growth: 0.0% | Long-run organic growth: 5.0% | Normalized EBITDA margin: 22.7% | Return on incremental organic capital: 26.1% | Annual dividend growth: 5.0% | Exit EV / steady-state cash NOPAT: 11.89× | Additional annual share dilution: 0.0% - Revenue growth: 0.0%, 0.6%, 1.1%, 1.7%, 2.2% - Margin: 22.7%, 22.7%, 22.7%, 22.7%, 22.7% ### Base Case Regional demand bridge supports initial 10% growth fading to 9%, about EUR349m in year 5: current TTM sales allocated by FY2025 regional weights, North America15%, Europe5%, APAC/MEA10%. Steam is included in the regions. Retain the 25.7% cycle-margin recovery and measured capital intensity; growth above this needs additional identified opportunities. - Key Drivers: Near-term organic growth: 10.0% | Long-run organic growth: 9.0% | Normalized EBITDA margin: 25.7% | Return on incremental organic capital: 38.4% | Annual dividend growth: 5.0% | Exit EV / steady-state cash NOPAT: 20.00× | Additional annual share dilution: 0.0% - Revenue growth: 10.0%, 9.9%, 9.8%, 9.7%, 9.6% - Margin: 23.3%, 23.9%, 24.5%, 25.1%, 25.7% ### Upside Case Faster near-term adoption begins at 17% and fades to the same 9% long-run rate. Keep 29.2% normalized EBITDA, 45% ROIIC and 20x cash NOPAT. Favorable demand does not itself establish a higher terminal multiple or a permanently longer growth runway. No M&A uplift is assumed. - Key Drivers: Near-term organic growth: 17.0% | Long-run organic growth: 9.0% | Normalized EBITDA margin: 29.2% | Return on incremental organic capital: 45.0% | Annual dividend growth: 5.0% | Exit EV / steady-state cash NOPAT: 20.00× | Additional annual share dilution: 0.0% - Revenue growth: 17.0%, 16.1%, 15.2%, 14.3%, 13.4% - Margin: 24.0%, 25.3%, 26.6%, 27.9%, 29.2% ## Research Notebook # Business model Harvia makes sauna heaters, control units, heater stones, spare parts, sauna rooms, Scandinavian hot tubs and steam equipment. It sells mostly to retailers, distributors, wholesalers, sauna builders and DIY chains in nearly 100 countries, and directly to consumers in North America and, from 2026, Germany and Austria (`harvia-annual-report-2025`, pp.8 and 141). Factories: Muurame, Finland (heaters, the largest); Lewisburg, West Virginia (Almost Heaven rooms, expansion completing Q4 2026); Driedorf, Germany (EOS premium heaters and controls); Guangzhou, China; Romania; Italy (Phoenix controls); Sastamala, Finland (Kirami tubs). Money is EUR millions unless stated. ## Evidence plan Primary: financial statements bulletins FY2019–FY2025 (`harvia-fsb-2019` … `harvia-fsb-2025`), the 2025 annual report (`harvia-annual-report-2025`), the Q3 2026 interim report and presentation (`harvia-q3-2026`, `harvia-q3-2026-presentation`), and the acquisition releases. Outside observations: Masco's price for Sauna360 (`masco-10k-2023-sauna360`), Tulikivi results (`tulikivi-fy2025`). No fieldwork, dealer survey or customer interview has been done. ## Value-creation mechanism Harvia turns capital into owner cash by manufacturing a small, high-margin component — the electric heater and its control unit — that goes into almost every new or renovated sauna, whoever builds the room, and then selling the room, stones, spares and accessories around that installed base through a dealer network it already owns. The step where most value is created is the heater/control sale into a fragmented, specified market: the product costs little capital to make (group materials and services were 35.8% of FY2025 revenue, so gross margin before labour was 64.2%), the installed base of over 18 million saunas generates replacement demand, and Harvia's estimated "clearly over 20%" share of heaters and components (`harvia-q3-2026` p.4) lets one factory serve many channels. Growth capital (capacity and inventory) is needed, but each organic sales euro has needed about EUR0.49–0.63 of growth capital through the cycle (`07-assumption-register.md`, 3 November). The weak step is overhead: since 2021 the profit gap has opened in personnel and other operating costs, not in product margin. ## Business-model profile Comparable firms are chosen by these traits, not by the industry label. **Growth vehicle: Organic.** Revenue moved from 179.1 in 2021 to 198.9 in 2025 through the post-pandemic downturn and recovery. Acquisitions are occasional bolt-ons for a product or region (Almost Heaven 2018, Kirami 2021, ThermaSol 2024), not a recurring capital program: acquisition cash was 5.5% of revenue over 2023–2025, almost all ThermaSol. Growth comes from capacity, distribution and price, with growth capital earning about 34% after tax across 2019–2025. The model forecasts organic growth only; ThermaSol is in the starting base and further deals are not forecast, because their timing and price cannot be estimated from the record. That omission is upside, not a base-case assumption. **Capital intensity: 4.6% of sales, asset-light.** Capital spending is about 4.6% of revenue over 2023–2025, raised by the Lewisburg factory expansion (completing Q4 2026) on top of maintaining factories in Finland, Germany, China, Romania and Italy. Growth needs plant capacity and inventory (tangible capital 86.7 in 2025), which earns about 35% after tax. The share sits just under the 5% boundary, so the asset-light label is marginal. **Comparables.** A comparable firm makes branded durable consumer products sold through retailers, builders and dealers, grows organically through capacity, distribution and price, buys only occasional bolt-ons, and carries moderate plant and inventory. Its revenue follows housing, renovation and discretionary-spending cycles. Not the label: "Leisure" or "building products" groups Harvia with commodity fixtures; its heaters and controls are a niche brand franchise earning about a 20% EBIT margin. Declared in `valuation/model.json` (`businessProfile`). `npm run check` measures deal and capital intensity from the reported facts, and the outside view grades the forecast only against firms that share them. ## Economic snapshot Definitions. Adjusted EBIT as reported (items affecting comparability were 0.8 in 2025, 1.6 in 2024). Tax at the model's 23% normalized rate (FY2025 effective rate was 20.4%: 6.776 / 33.182). Tangible capital = PP&E + right-of-use assets + net working capital (inventories + trade and other receivables − trade and other payables); excess cash is excluded. Return on tangible capital (ROTC) capitalizes leases; return on capital employed (ROCE) here includes goodwill and acquired intangibles, which is how a buyer of the whole company pays. Sources: balance sheet and income statement `harvia-annual-report-2025` pp.134–135; bulletins for earlier years. FY2025 capital base (31 December 2025): | Item | EUR m | | --- | ---: | | PP&E | 31.6 | | Right-of-use assets | 7.3 | | Net working capital (inventory 52.1 + receivables 24.6 − payables 28.6) | 47.9 | | **Tangible capital** | **86.7** | | Intangible assets (acquired brands, customer relationships, software) | 19.2 | | Goodwill | 89.3 | | Loans from credit institutions | 95.4 | | Lease liabilities | 7.5 | | Cash | 45.2 (none treated as excess while net debt is 57.7) | Returns, FY2025: adjusted EBIT 39.1 → NOPAT 30.1. ROTC on year-end tangible capital 34.7%; on average tangible capital (84.0) 35.8%. ROCE including goodwill and intangibles (195.4) 15.4%. The issuer's adjusted ROCE (pre-tax, excluding goodwill but including intangibles) was 41.3% (`harvia-annual-report-2025` p.55); at 23% tax that is about 32%, consistent with our ROTC once the 19.2 of intangibles is included. Seven-year history (PP&E + NWC as tangible capital, leases excluded before 2024 for comparability): | FY | Revenue | Gross margin* | Opex ratio** | Adj. EBIT margin | Inventory days*** | Tangible capital | After-tax ROTC | ROCE incl. goodwill | Incremental EBIT margin | | --- | ---: | ---: | ---: | ---: | ---: | ---: | ---: | ---: | ---: | | 2019 | 74.1 | 60.3% | 38.7% | 18.7% | 171 | 31.4 | 34% | 11.7% | — | | 2020 | 109.1 | 61.5% | 37.3% | 22.4% | 180 | 34.9 | 54% | 17.8% | 30% | | 2021 | 179.1 | 60.9% | 31.9% | 26.4% | 240 | 69.9 | 52% | 25.3% | 33% | | 2022 | 172.4 | 59.3% | 35.3% | 21.1% | 236 | 72.4 | 39% | 19.2% | n/m (sales fell) | | 2023 | 150.5 | 62.7% | 37.2% | 22.4% | 231 | 63.0 | 41% | 19.0% | n/m (sales fell) | | 2024 | 175.2 | 64.3% | 40.6% | 21.2% | 287 | 73.1 | 39% | 17.4% | 14% | | 2025 | 198.9 | 64.2% | 42.0% | 19.6% | 267 | 79.4 | 38% | 17.8% | 8% | \* 1 − materials and services / revenue. \*\* Employee benefit expenses + other operating expenses, / revenue. \*\*\* Year-end inventory / materials and services × 365. Sources: income statements and balance sheets in `harvia-fsb-2019` … `harvia-fsb-2025`. What the snapshot says: - Product economics improved: gross margin rose from about 60% to 64% between 2019–2022 and 2024–2025, even through the 2025 tariffs and a weak dollar. - Overhead absorbed all of it and more. The opex ratio rose ten points from 2021 to 2025. Employee costs grew from 30.6 to 42.0 and other operating expenses from 26.5 to 41.4, while revenue grew 19.8. Q3 2026 repeated it: personnel and other costs grew faster than revenue, headcount 816 vs 728 (`harvia-q3-2026` pp.1–3, 8). - Inventory is heavy: 230–290 days of materials cost since 2021, versus 170–180 before the pandemic. Working capital, not machinery, is the main growth capital. - Incremental margins fell from about 30% (2020–2021, existing overhead) to 8–14% (2024–2025, while overhead was being built). Normalizing for the unconsolidated part of ThermaSol (consolidated from 31 July 2024) moves 2024–2025 by under a point and doesn't change this. - Current price: at EUR39.70 (3 November 2026, `harvia-q3-2026` share count) market capitalization is about 742 and EV about 803. Trailing September 2026 cash NOPAT on the engine definition ((49.6 − 1.26% × 218.9) × 0.77) is 36.1, so EV is 22× today's earning power and about 17× earning power at a 29.2% EBITDA margin. ## Economic segmentation Harvia reports revenue by region and product group but no segment profit. Profit has to be inferred from which units Harvia has bought and at what margins, and from what moved the group margin. | Economic unit | FY2025 revenue (share) | Demand driver | Estimated EBITDA margin | Estimated share of FY2025 EBITDA (46.5) | Trend | | --- | ---: | --- | --- | ---: | --- | | Heaters, controls, IR components | 107.9 (54%) | New saunas and replacement of the installed base; specified by builders and dealers | High: residual after the units below, about 27% | about 63% | Growing; +14.6% in 9M 2026 | | Accessories, stones, spare parts and services | 26.2 (13%) | Installed base usage and replacement; no new room needed | High, similar to heaters | about 15% | Growing 16–17% in 9M 2026 | | Saunas (rooms, cabins) and Scandinavian hot tubs | 47.5 (24%) | New-room purchases, mostly North America (Almost Heaven, D2C and retail) and Finland (Kirami); discretionary | Estimate 15% (Kirami was bought at a 15.0% adjusted EBITDA margin, `harvia-kirami-2021`) | about 15% | Growing fastest in NA, +18.8% in 9M 2026 | | Steam products | 17.2 (9%) | US residential construction and renovation; commercial projects | Estimate 15–17% (ThermaSol was bought at 17.2%) | about 6% | Declining: −8.2% in 9M 2026 | Method: allocate the room/tub and steam units the margins at which Harvia bought comparable businesses, and attribute the rest of group EBITDA to heaters, controls and aftermarket. Heaters, controls and the aftermarket are then about two thirds of revenue and about 78% of EBITDA (36.5 on 134.1 of revenue, a 27% margin). Adverse allocation: give rooms and steam the full group margin of 23.4%; heaters and aftermarket still produce about two thirds of EBITDA. Either way, revenue share overstates the profit share of rooms and steam. These margins are estimates from acquisition disclosures, not reported segment data. Regional view (FY2025 revenue / 9M 2026 growth): North America 75.8 (38%) / +21.9%; Continental Europe 55.6 (28%) / +5.1%; Northern Europe 46.6 (23%) / +12.8%; APAC & MEA 21.0 (11%) / +9.6% (`harvia-fsb-2025` p.5, `harvia-q3-2026` p.5). Growth comes from North America, where the mix is richer in rooms and steam. Northern and Continental Europe are mostly heater deliveries from Muurame, which is why the Q2 2026 IT stoppage hit those regions (`harvia-q3-2026` p.5). Group growth led by North America therefore tends to dilute the margin even if each product earns what it did before. ## Unit economics Unit: one euro of additional organic sales, which is how Harvia adds value. It sells through existing dealers and factories, and one more sauna heater shipped from Muurame or room shipped from Lewisburg needs inventory, receivables and occasionally a capacity step. The P&L and capital per EUR100 of incremental sales, from the group accounts: | Per EUR100 of incremental sales | 2019→2021 (existing overhead) | 2023→2025 (building overhead) | Source / basis | | --- | ---: | ---: | --- | | Materials and services | 39 | 31 | Incremental: gross margin on total sales rose from 62.7% to 64.2% over 2023→2025 | | Gross profit | 61 | 69 | | | Incremental personnel and other operating cost | 27 | 57 | ΔOpex / Δsales: 2019→2021 +28.4 / +105.0; 2023→2025 +27.4 / +48.4 | | Depreciation increment | 3 | 2 | ΔD&A / Δsales | | Other operating income and comparability items | 0 | +1 | Residual to reported adjusted EBIT | | Incremental operating profit | 32 | 11 | Reported incremental adjusted EBIT margin | | NOPAT at 23% tax | 24 | 9 | | | Growth capital (working capital + capex above the recurring charge) | 49 | 49 | Full-cycle measured EUR0.49 per organic sales euro, FY2019→FY2025, lagged (`calculations.json` `capitalPerOrganicSale2019to2025`) | | of which working capital | about 29 | about 29 | Working-capital investment / organic sales change, FY2019–FY2025 | | **After-tax return on the unit's capital** | **about 50%** | **about 18%** | NOPAT / growth capital | Product economics got better between the two columns (incremental gross margin 69 against 61); overhead absorbed the gain and more. At the measured capital intensity, an incremental sales euro returns about 50% after tax if overhead is held, and under 20% while overhead grows at today's pace. The model's 45% organic ROIIC, using the 29.2% normalized margin as the steady state, sits close to the first column. It assumes the overhead now being built is a one-off step that later sales absorb. The FY2021 build (capex 11.8, inventory +22.6) was put in place when organic sales rose 43%. Sales then fell 7.7% and 9.4% in 2022–2023 as dealers destocked, and the capacity looked like dead money: ROTC fell from 52% to 39–41%. Measured from FY2019 to FY2025 across that cycle, growth capital of 40.5 supported 82.6 of organic sales at the same start and end NOPAT margin (about 17%), a 34% after-tax return at actual margins (`07-assumption-register.md`, 3 November). The project still earned its cost of capital several times over through a full boom and bust. Failure case: steam. Steam revenue then fell 8.2% in 9M 2026 because US residential construction and renovation, which drive steam showers, weakened (`harvia-q3-2026` pp.3–5). A steam shower is part of a bathroom build. The buyer is a contractor responding to housing turnover, not a consumer adopting sauna, so the replacement-and-adoption lever behind heaters doesn't apply. The same fixed overhead earns less on it. ## Independent checks 1. Reported capital returns. Harvia's adjusted ROCE excluding goodwill was 41.3% in FY2025 and 39.9% on a trailing basis in September 2026 (pre-tax, average capital). Our after-tax ROTC of 35–36% grossed up for tax (about 46%) and including 19.2 of intangibles agrees with it. The unit model's 49% "existing overhead" return and the 34–44% measured cycle return bracket the group's tangible return, so the unit model reconciles with group accounts. 2. Private-market price of the closest competitor. Masco paid about EUR124m net of cash for Sauna360 (Tylö, Helo, Finnleo, Kastor, Amerec) in 2023 and recorded USD60m of goodwill on USD136m (`masco-10k-2023-sauna360`). Sauna360's revenue isn't disclosed in the filing, so its margin can't be checked. The price shows that the second global heater platform changed hands for about one sixth of Harvia's current EV. 3. Scale check against a sub-scale peer. Tulikivi, a listed Finnish stove and fireplace maker with a sauna-heater line, had EUR29.5m of 2025 sales and a comparable operating loss of EUR0.1m (`tulikivi-fy2025`). A small heater and stove maker without Harvia's dealer reach and volume doesn't earn Harvia's margins. That supports the view that Harvia's returns come from scale and distribution rather than from the product category itself. This is one observation, not a peer set. 4. Market-share arithmetic. The issuer estimates a EUR4bn global sauna market and puts its share above 5% (`harvia-q3-2026` p.4). Harvia's 2025 revenue of 198.9 is 5.0% of 4bn, consistent with the estimate; its heater and component share is "clearly over 20%". The estimate is management's, not independent. Adverse allocation: assign all of the 2023–2025 opex increase to the growth program and none to inflation; incremental NOPAT is then 9 per 100 against 49 of capital, an 18% return. That still beats the cost of capital, but it is less than half the model's 45%. The conclusion that growth creates value survives. The conclusion that it creates value at 45% depends on overhead being absorbed. Where the consensus differs. The issuer's own long-term target is an adjusted EBIT margin "exceeding 20%" (`harvia-q3-2026` p.2), about 24% EBITDA at today's D&A. The model's 29.2% EBITDA is about 25.5% adjusted EBIT, achieved only in 2021. The observation that would settle it is the FY2026 and FY2027 opex ratio: if personnel and other operating costs grow more slowly than revenue once Lewisburg and the IT platform are complete, the 29.2% path is plausible. If the ratio stays near 42%, the unit return is about 20%. # History, market & cycle Harvia was a Finnish heater maker (founded 1950, Muurame), listed on Nasdaq Helsinki in March 2018. It became an international sauna company through acquisitions: Sentiotec (Austria, 2016), Almost Heaven (US rooms, December 2018), EOS (German premium heaters, controls and steam, April 2020), Kirami and Sauna-Eurox (Finland, 2021), Phoenix El-Mec (Italian controls, 2023) and ThermaSol (US steam, July 2024). Russia was exited in 2022. Matias Järnefelt replaced Tapio Pajuharju as CEO in 2023. The record that follows separates structural adoption from the pandemic pull-forward and the dealer destocking that followed. Money is EUR millions. ## Demand drivers and cycle Who decides. The end user is a homeowner, or a hotel, spa or gym. The payer is the homeowner, either directly (D2C, retail, DIY chains) or through a builder or renovator. The heater brand is often chosen by the dealer, sauna builder or electrician who specifies it, and the consumer chooses the room. Harvia sells mostly to retailers, distributors and builders on contracts that are usually annual (`harvia-annual-report-2025` p.141). That gives dealers timing power: they can destock without losing the brand. Demand split (judgment from the product and regional data; the issuer doesn't split demand this way): - Replacement and aftermarket. A heater, its elements and stones wear out in an installed base the issuer puts at over 18 million saunas, more than half in Finland, Germany, Russia and the US (`harvia-q3-2026` p.4). Spare parts, services, accessories and stones were 26.2 (13%) in 2025 and grew 16–17% in 9M 2026. Replacement is the least cyclical demand and needs no new room. - Growth (adoption). New saunas in North America and APAC, driven by health awareness rather than consumer confidence, according to management. North American revenue rose from 62.0 (2024) to 75.8 (2025), +21.9% in 9M 2026, while US consumer confidence and residential construction were weak (`harvia-q3-2026` pp.3–4). This is most of the incremental demand and the most uncertain part. - Regulation-driven. Little demand is created by rules, but certification is a barrier to entry: a heater sold in the US needs ETL/UL listing. Tulikivi announced US ETL approval for its electric heaters in 2026 (`tulikivi-us-etl-2026`), which shows the barrier can be crossed. Product-safety recalls (Sauna360's 2026 infrared recall, `sauna360-cpsc-2026`) show that regulators can hit a competitor's channel. - Discretionary. Rooms, outdoor saunas, hot tubs and steam showers are big-ticket purchases tied to renovation and housing. Steam fell 8.2% in 9M 2026 with US residential construction (`harvia-q3-2026` p.5). The measured cycle (issuer organic growth at comparable currency; `harvia-fsb-2020` … `harvia-fsb-2025`, `harvia-q3-2026`): | Year | Organic growth | Revenue | Adj. EBIT | Adj. EBIT margin | Inventory (year end) | What happened | | --- | ---: | ---: | ---: | ---: | ---: | --- | | 2020 | +28.0% | 109.1 | 24.4 | 22.4% | 20.7 | Pandemic home-improvement surge; EOS acquired | | 2021 | +43.3% | 179.1 | 47.3 | 26.4% | 46.1 | Peak; capacity and inventory build | | 2022 | −7.7% | 172.4 | 36.5 | 21.1% | 45.3 | Dealer destocking; Russia exit; energy price spike in Europe | | 2023 | −9.4% | 150.5 | 33.7 | 22.4% | 35.5 | Trough; inventory released 10.1 of cash | | 2024 | +12.9% | 175.2 | 37.1 | 21.2% | 49.2 | Recovery led by North America (+62.7% in Q4); ThermaSol | | 2025 | +14.4% | 198.9 | 39.1 | 19.6% | 52.1 | All regions grow; Northern Europe back to growth after two years of decline | | 9M 2026 | +15.9% | 165.2 | 31.0 | 18.7% | — | North America +21.9%; Q2 IT stoppage shifted about 4 of deliveries to Q3 | Peak to trough (2021→2023), revenue fell 16% and adjusted EBIT 29%, but the margin never went below 21% and operating cash flow peaked in 2023 (39.1) as inventory was released. This was a two-year organic decline of about 16%, not a collapse. By 2025 organic sales were about 8% above the 2021 peak (1.000 × 0.923 × 0.906 × 1.129 × 1.144 ≈ 1.08), and 2026 added another 16% at comparable currency. The market misperception to test is that the sauna boom was a pandemic fad. The organic index above refutes it for the group: the base exceeded its 2021 peak within four years. The right series to keep testing is North American organic growth against US residential construction. If NA sauna demand follows housing, the "health adoption" explanation is wrong. So far it has diverged: rooms and heaters grew while steam, the housing-linked product, fell. Seasonality matters for reading quarters: Q3 is the weakest cash quarter because working capital is built before winter (operating free cash flow −0.1 in Q3 2026). ## Competition Where competition happens: - Premium and professional (Germany, spas, hotels). EOS competes with Klafs and others in commercial projects. These are project sales, which are lumpier (Middle East projects were postponed in 2026). - Rooms and outdoor saunas (North America). Almost Heaven competes with Finnleo (Sauna360), many infrared brands and imported barrel saunas. Competition here is on price, delivery and online reach; about 70% of what Harvia sells in the US is made there, which shields it from tariffs relative to importers (`harvia-q3-2026` p.11). Campaign pricing is visible (`almost-heaven-black-friday-2026`), so this is the most price-competitive unit. - Steam (US). ThermaSol competes in a contractor channel tied to bathroom renovation; demand is weaker. Trend in the number of competitors: consolidation at the top (Masco–Sauna360 in 2023; Harvia's six deals), fragmentation at the bottom (imported rooms; new US entrants in heaters). Advantage in one sentence: Harvia makes the highest-volume heater range in the industry in its own factories, sells it through the widest dealer network, and builds most US product in the US, so it has the lowest unit cost and the fastest availability in the component dealers specify most. Evidence that it lasts: gross margin rose from about 60% (2019–2022) to 64% (2024–2025) through tariffs, the weak dollar and campaign pricing, so it isn't buying volume with price. Sub-scale Tulikivi lost money in 2025 (`tulikivi-fy2025`). The share estimates rose from 13% (Harvia alone, 2020) to over 20% (2025). The competitor who could hurt: Masco, through Sauna360. It owns the second heater platform, a US plumbing and wellness channel (Watkins hot tubs), and a balance sheet many times Harvia's. It could cut heater prices in North America or bundle saunas with hot tubs through Watkins dealers. Why it hasn't (judgment): Sauna360 is a small bolt-on inside Masco's plumbing segment, bought to enter a category, and still being integrated into Watkins. The 2026 Sauna360 infrared recall points to execution problems, not a price campaign. Harvia's rising gross margin is the observation that would change first if Masco started a price war. ## Counterparty incentives | Counterparty | What they want | What Harvia gives them | Their outside options | What would change the arrangement | | --- | --- | --- | --- | --- | | Largest customer (about 11% of 2025 net sales, 9% in 2024; not named, `harvia-annual-report-2025` p.141) | Margin per shelf or web listing, sell-through, campaign support, reliable US supply | A brand consumers search for, US-made product (tariff-safe), Black Friday campaign volume | Sauna360/Finnleo, Huum, imports, private label | A private-label or alternative-supplier program; a demand for more promotional funding. Dependence runs both ways: one customer at 11% is a risk, but the concentration is also how Harvia reaches US consumers at scale | | Dealers, distributors, sauna builders (many, annual contracts) | Availability, margin, a brand that installers trust, low warranty returns | The full range (heater, controls, room, stones, spares) from one supplier; delivery speed | Switch heater brand at the next contract | A prolonged delivery failure (the Q2 2026 Muurame IT stoppage postponed about 4 of deliveries; recovered in Q3) or a better-margin rival. Dealers showed their power in the 2022–23 destocking | | Consumers (D2C) | Price, delivery, health experience | Almost Heaven brand, direct web sales | Infrared and imported rooms | Price competition in rooms; fading health narrative | | Suppliers (steel, electronics, transport) | Volume | Scale purchasing | Many | No single-source dependence is disclosed; Harvia says it can shift production between factories and change suppliers for many key materials (`harvia-q3-2026` p.11). Unlike TerraVest's boiler business, no supplier controls a key component | | Regulators (US safety certification, CPSC, EU rules) | Safe products | Certified products | — | A heater or room safety recall at Harvia would damage the dealer channel the advantage rests on | | Shareholders | — | — | — | No controlling holder: the largest, Onvest Oy, held 4.43% at end-2025; ten largest registered holders 27.4% in September 2026 (`harvia-q3-2026` p.8). No one can force a sale or block a strategy | Read in both directions, the main dependence is on dealers and the largest retailer, and it is mutual: they need the category leader's range and US supply, and Harvia needs their shelves. Nothing found suggests a counterparty is positioned to take Harvia's share of value. Sources: as cited; channel structure from `harvia-annual-report-2025` pp.8 and 141. # Management & incentives Capital allocation is decided by a seven-member board with no controlling shareholder and a CEO hired from outside in 2023. Several long-serving managers from the pre-IPO team are still in the management team and hold most of the insider stake. Judge capital allocation by the cash returns of completed investments and by what the next factory or acquisition costs; a high group ROCE doesn't establish either. Sources: `harvia-annual-report-2025` (governance statement and remuneration report, pp.29–52 and 130), `harvia-q3-2026` pp.9–10, `harvia-manager-lengauer-2026-10-30`. Values at EUR39.70 (3 November 2026). ## Ownership and incentives Holdings at 31 December 2025, including controlled companies (`harvia-annual-report-2025` pp.32–39, 130): | Person | Role | Since | Harvia shares | Value at EUR39.70 | Notes | | --- | --- | --- | ---: | ---: | --- | | Matias Järnefelt | CEO | 2023 | 0 | 0 | Earned 8,032 gross LTI shares for 2023–2025, delivered net of tax in March 2026; no open-market purchase disclosed | | Ari Vesterinen | CFO | 2014 | 151,650 | 6.0m | Pre-IPO management | | Markus Wörmanseder | Head of Products and Solutions | 2017 | 106,960 | 4.2m | | | Timo Harvia | Head of Innovation and Technology | 2014 | 103,316 | 4.1m | Founding family; also represents Tiipeti Oy (342,790 shares, 1.83%) on the Nomination Board, not disclosed as a controlled company | | David Ahonen | Head of APAC & MEA | 2014 | 82,983 | 3.3m | | | Anssi Pelkonen | Head of Northern Europe | 2014 | 63,052 | 2.5m | | | Mika Suoja | Head of Operations | 2016 | 35,000 | 1.4m | | | Päivi Juolahti | Head of Marketing and Brands | 2020 | 3,371 | 0.1m | | | Nathan Hagemeier, Ivan Sabato, Philipp Krauth | Heads of North America, Continental Europe; EOS | 2024–25 | 0 | 0 | New regional leaders hold nothing | | Olli Liitola | Director; Chair 2014–2024 | 2014 | 42,174 | 1.7m | | | Hille Korhonen | Director | 2021 | 5,407 | 0.2m | | | Heiner Olbrich | Director; Chair until Nov 2025 | 2022 | 3,000 | 0.1m | Left the board at the 2026 AGM | | Catharina Stackelberg-Hammarén | Chair from Nov 2025 | 2023 | 1,836 | 0.1m | | | Anders Holmén | Director | 2021 | 1,986 | 0.1m | Managing Director, Fyrklöver-Invest | | Markus Lengauer | Deputy Chair from April 2026 | 2024 | 880 + 660 bought 30 Oct 2026 at EUR38.10 | 0.1m | The only open-market purchase found in 2026 | | Petri Castrén | Director, audit chair | 2025 | 301 | 0.0m | CFO of Kemira | | **Board, CEO and management team total** | | | **604,076 (3.2%)** | **24.0m** | 507,816 (2.7%) a year earlier | Pay. CEO total remuneration paid in 2025 was EUR839,153: fixed salary 566,545 (up 11%, partly because a Finland-wide personnel bonus was converted into base salary), 2024 short-term incentive 243,556 and a 29,052 performance bonus. The 2025 STI earned 213,717 (54% of maximum). STI: up to 70% of fixed salary, 70% on consolidated adjusted operating profit and 30% on personal targets (2025: steam business, Continental European D2C, infrared strategy). LTI: target 70% and maximum 140% of fixed salary in shares; criteria for 2024–2026 and 2025–2027 are absolute TSR 50%, revenue growth 40% (payable only above an undisclosed EBIT-margin threshold) and CO2 reduction 10%. The plan has no holding period or shareholding requirement (`harvia-annual-report-2025` pp.46–52). Board: chair EUR63k, others EUR49k on average in 2025 (up 30%), about 40% of monthly fees paid in shares. Whose interests dominate: - CEO: stake EUR0, annual pay EUR0.84m. If the stock compounds 10% a year, his maximum LTI shares per plan (11,894) are worth about EUR0.5m at today's price, roughly 60% of one year's cash pay. Pay, which grows with revenue growth and adjusted operating profit, dominates his economics. The TSR leg is the only per-share measure. - Long-serving managers (CFO, Wörmanseder, Timo Harvia, Ahonen, Pelkonen): EUR2.5–6.0m each. At 10% compounding the CFO's stake gains about EUR0.6m a year, probably more than his salary (not disclosed individually). Their interests are closer to owners'. - Board: small holdings; the chair holds about EUR0.1m. Shareholders discipline the board through the Nomination Board of the four largest holders (2026: representatives including Nordea, Evli and Onvest), not through director ownership. No compensation metric measures return on capital, cash conversion or per-share value other than TSR. Revenue growth and adjusted operating profit can both be raised by spending: overhead to grow sales, or acquisitions bought with debt. ## Track record and references Two eras. Pajuharju era (CEO to May 2023; IPO 2018). Revenue 74.1 (2019) to 179.1 (2021) through Almost Heaven, EOS and Kirami and the pandemic surge. Measured outcomes: Almost Heaven, taken over in December 2018 without cash consideration, became the core of a North American region with 75.8 of 2025 revenue. Kirami cost 9.5 in total for 22.1 of revenue and 3.3 of adjusted EBITDA (about 2.9×). EOS cost 19.7 for 78.6% in 2020 on a 25.5 debt-free value (17.3 of 2019 revenue), but the 21.4% minority was then bought in July 2022 for 19.0 at the same EBITDA multiple. That implies a 100% value about 3.5× the 2020 valuation, set on boom-period EBITDA just before the downturn (`harvia-fsb-2020`, `harvia-fsb-2022`). Inventory was built to 240 days in 2021 and then released through 2023. Russia (EOS Russia and Harvia operations) was exited in 2022 with the brand rights retained. Adjusted EBIT margin peaked at 26.4% in 2021. Pajuharju still held 139,000 shares at end-2025 (`harvia-annual-report-2025` p.130); his sales history wasn't examined. Järnefelt era (from 2023). Revenue 150.5 (2023) to 218.9 trailing September 2026 (+45%, of which ThermaSol about 13). Adjusted EBIT margin 22.4% (2023) → 19.6% (2025) → 18.7% (9M 2026). Adjusted ROCE excluding goodwill 54.5% (2023) → 41.3% (2025) → 39.9% (September 2026). Headcount 605 → 816. Decisions: - Worked: North American growth (+62.7% in Q4 2024, +22.1% in 2025, +21.9% in 9M 2026); APAC (China, Japan, Australia); Northern Europe returned to growth in 2025; gross margin rose to 64%. - Failed or not yet proven: ThermaSol (about 11–13× EBITDA in 2024, steam −8.2% in 9M 2026); the Muurame IT/ERP cut-over (about EUR4m of Q2 2026 deliveries postponed); a North American revenue-recognition correction in Q3 2026 (−0.7 revenue, −0.8 EBIT); North America head Jennifer Thayer resigned in May 2025, interim cover until Nathan Hagemeier joined in November 2025; opex outgrew revenue in 2024, 2025 and 9M 2026. - Shareholder return: from EUR17.68 at end-2022 to EUR39.70 plus about EUR2.84 of dividends paid 2023–2026, about 26% a year. Most of it is re-rating from a trough (EV about 12× cash NOPAT at end-2022 against 22× today), not per-share earnings growth: EPS rose from EUR1.25 (2023) to EUR1.41 (2025). Since end-2019 (EUR10.45): price ×3.8 plus about EUR4.3 of dividends, about 23% a year over 6.8 years. References. None collected. No former employee, dealer, competitor or customer has been interviewed; the issuer's annual customer and employee surveys (cited in `harvia-fsb-2025`) are self-reported. Bench depth: the CFO and four other managers have 9–12 years in the management team; the North American and Continental European leadership changed in 2025. This is a gap in the research. Insider trades. The only 2026 open-market trade found is Lengauer's purchase of 660 shares at EUR38.10 on 30 October 2026 (`harvia-manager-lengauer-2026-10-30`), about EUR25,000, the day after the Q3 fall. That is too small to signal much. Management sales over the full period weren't reviewed (open check). ## Capital allocation principles Governing objective as stated: average annual revenue growth of 10%, adjusted operating margin above 20%, net debt/adjusted EBITDA below 2.5×, and a regularly increasing dividend paid twice a year (`harvia-q3-2026` p.2). These are growth and margin targets, not per-share value or return-on-capital targets. 1. Zero-based allocation. Mixed. Capital has moved toward North America (Lewisburg expansion, ThermaSol, D2C, IT), and Harvia exited Russia (2022) and sold its Estonian operations (2025). But the 2023–2026 overhead build was allocated across all regions at once, and no business has been shrunk to fund another. 2. Funding strategies, not projects. Partly. The four strategic priorities (full sauna experience, strategic markets, key channels, operations) are funded as a program; no project-level returns are reported. The cost of the program shows up as opex, so it bypasses capital-budgeting discipline. 3. Capital has a cost. Share compensation is small: about 79,000 maximum LTI shares per plan (0.4% of shares), settled from treasury shares bought in the market (17,000 at EUR39.01 in February 2026), so dilution is offset with cash, a financing cost the cash flow statement records as a buyback. Debt is used for acquisitions (EOS 2020, EOS minority 2022, ThermaSol 2024: each year net borrowing was about 20). Leverage 1.2× against a 2.5× ceiling. 4. Zero tolerance for bad growth. Weak evidence. Opex was cut in the 2022–23 downturn (60.9 → 56.0), so management can retrench. Since then three years of opex growing faster than revenue haven't been reversed, and ThermaSol hasn't been reported on separately. 5. Price versus value. Passive. No buyback at the 2022 trough (EUR17.68, about 12× cash NOPAT), when the stock was cheap against the business; buybacks are only for incentive delivery. No equity was issued at the 2021 peak (EUR58.70, 29×). The 2022 EOS minority purchase followed a pre-agreed option formula at a boom multiple, so price discipline was given up in advance. Does management behave as if long-term value per share is the goal? The dividend has risen every year since the IPO and acquisitions have mostly been small and cheap (EOS majority, Kirami). But incentives and stated targets point at growth and margin, and the current program is buying growth with overhead at a lower incremental return than the business historically earned (`01-business.md`, unit economics). The record so far is "good business, competent operators, unproven allocators of the new overhead". ## Integrity and governance Evidence: - Board independence: all seven directors elected in 2025 were assessed independent of the company and of its major shareholders (`harvia-annual-report-2025` pp.32–33). One share class; no dual-class or super-voting control. - Ownership is dispersed: the ten largest registered holders owned 27.4% in September 2026; nominee and foreign holders 38.4%; households about 19% at end-2025. A Shareholders' Nomination Board of the four largest holders proposes directors. - Adjustments: items affecting comparability were small (0.8 in 2025, 1.6 in 2024; 0.3 in 9M 2026), mostly transaction and restructuring costs. No pattern of adding back recurring costs. - Related party: the 2022 EOS minority was bought from EOS's managing director, a management-team member, at a pre-agreed formula; disclosed. - Pay: CEO fixed salary rose 11% in 2025 and the board's average fee 30%; the 2025 STI paid 54% of maximum with operating profit below target. The remuneration report was approved in an advisory vote at the 2025 and 2026 AGMs (vote counts not reviewed). - Controls: the Q3 2026 North American revenue-recognition correction (EUR0.7m revenue) is small but is a control finding in the fastest-growing region, one quarter after the Muurame IT cut-over. Deloitte is the auditor. Agency costs (Mauboussin and Callahan pp.47–48): - Control of resources: present. Revenue growth is 40% of the LTI and operating profit 70% of the STI, with no return-on-capital measure, while the CEO owns no shares. This rewards growth spending and acquisitions whether or not they earn their cost. It is the agency cost that matters here. - Risk tolerance: low concern. Leverage is modest and the CEO's undiversified exposure is pay, not stock. - Time horizon: LTI periods are three years with no holding requirement, shorter than a five- to ten-year investment horizon. Valuation treatment: the model charges all opex as reported (no add-back of growth spending) and gives no value to future acquisitions. The agency cost appears as the risk that the 29.2% margin recovery never comes because spending keeps rising with revenue. Since 3 November 2026 the base and downside hold the margin at the 25.7% cycle average, which matches management's stated reinvestment policy; only the upside assumes the 29.2% normalization (`05-thesis.md`, `06-risks.md`). # Capital allocation Where Harvia's cash went over FY2021–FY2025, what each use earned, and how the model funds growth. Money is EUR millions. Facts are in `financials/facts.json` (cash-flow statements of `harvia-fsb-2021` … `harvia-fsb-2025`); arithmetic is in `valuation/calculations.json`. ## Sources and uses of cash | | FY2021 | FY2022 | FY2023 | FY2024 | FY2025 | Five years | | --- | ---: | ---: | ---: | ---: | ---: | ---: | | Cash flow before working capital (pre-tax) | 51.2 | 42.4 | 39.6 | 42.6 | 46.4 | 222.3 | | Working capital (− = investment) | −20.9 | −5.4 | +7.8 | −2.9 | −5.2 | −26.6 | | Income taxes and operating interest | −8.5 | −12.7 | −8.3 | −8.0 | −6.2 | −43.7 | | **Net cash from operating activities** | **21.8** | **24.3** | **39.1** | **31.7** | **35.0** | **151.9** | | Capex net of asset sales | −11.7 | −3.5 | −3.0 | −6.1 | −14.3 | −38.7 | | *memo: D&A* | *5.8* | *6.5* | *6.3* | *7.0* | *7.4* | *33.0* | | Lease repayments | −0.4 | −0.5 | −0.8 | −0.9 | −1.0 | −3.6 | | Interest paid (net of interest received in investing) | −1.9 | −1.0 | −2.9 | −1.8 | −4.2 | −11.8 | | **Free cash flow to equity before acquisitions** | **7.9** | **19.3** | **32.4** | **22.8** | **15.4** | **97.8** | | Acquisitions net of cash (and divestments) | −7.6 | +0.1 | −2.8 | −24.8 | −1.2 | −36.3 | | EOS minority purchase | — | −19.0 | — | — | — | −19.0 | | Dividends | −9.5 | −11.3 | −12.0 | −12.7 | −14.0 | −59.5 | | Treasury share purchases (for incentive delivery) | −2.5 | −0.3 | — | — | — | −2.8 | | Net borrowing (excluding leases) | −0.4 | +20.9 | −1.9 | +20.1 | −0.3 | +38.4 | | **Net change in cash** | **−12.1** | **+9.6** | **+15.7** | **+5.3** | **−0.0** | **+18.5** | | FX on cash | +0.3 | +0.2 | −0.4 | +0.5 | −1.2 | −0.6 | | Cash at year end | 15.5 | 25.3 | 40.6 | 46.4 | 45.2 | 27.3 → 45.2 | Each column reconciles to the reported net change in cash within rounding. FCF to equity is the same definition as the `ownerCashPerShare` history series. What the table shows about priorities: - The dividend comes first and rises every year: 59.5 over five years, 61% of FCF to equity, paid through the 2022–23 downturn. - Acquisitions (55.3 including the EOS minority) were funded with debt: net borrowing of 38.4 came in 2022 (EOS minority) and 2024 (ThermaSol). - Organic growth capital appears inside operating cash: net capex was 38.7 against 33.0 of D&A, and working capital absorbed 26.6, of which 20.9 was the 2021 inventory build. - There have been no buybacks for value. Treasury purchases only deliver incentive shares. - Cash built from 27.3 (end-2020) to 45.2 while gross bank debt rose to 95.4. Net debt was 57.7 at end-2025 (1.2× adjusted EBITDA). ## Investment rates per sales change Window FY2019→FY2025: reported sales +124.8, of which organic +82.6 (issuer organic growth rates; acquired businesses count as organic after twelve months). Rates are computed in `valuation/calculations.json`. | Rate | Definition | FY2019→FY2025 | Step | | --- | --- | ---: | --- | | Incremental M&A rate | Acquisition cash ÷ Δsales | 43.7% (58.9% incl. the EOS minority) | `acquisitionRate2019to2025` | | Expensed growth (intangible) rate, upper bound | Δ(personnel + other operating cost) ÷ Δsales | 43.9% | `expensedCostRate2019to2025` | | — of which FY2023→FY2025 | same | 56.7% | `expensedCostRate2023to2025` | | Fixed capital rate | (Capex − D&A) ÷ Δsales | 3.1% | `fixedCapitalRate2019to2025` | | Working capital rate | Cash-flow working-capital investment ÷ Δsales | 19.4% (29.2% of organic Δsales) | `workingCapitalRate2019to2025` | Where Harvia really invests: - Not in fixed assets. Capex barely exceeds D&A over the window, though D&A includes acquired-intangible amortization and right-of-use depreciation. Excluding those, organic fixed investment is about 0.20 per organic sales euro (0.49 total growth capital less 0.29 of working capital). - In working capital: 0.29 per organic sales euro, because inventory runs at 230–290 days of materials. - In acquisitions: 44 cents per sales euro. - Increasingly in overhead. Over FY2019→FY2025, personnel and other operating costs rose 54.8, 44% of the sales increase. Under the current program (FY2023→FY2025) they absorbed 57% of each incremental sales euro, against 27% over FY2019→FY2021. R&D expensed is only 1.6–2.1 a year (`harvia-annual-report-2025`), so the growth investment is mainly selling, marketing, IT and regional organizations, expensed as incurred. This rate includes inflation and acquired overhead (ThermaSol), so it is an upper bound on investment. It is still the main inflection point in the record. Who was in charge when the pattern changed. The working-capital surge (2021) and the debt-financed EOS minority (2022) were under Tapio Pajuharju. The overhead build (2023–2026), ThermaSol (2024), Lewisburg and the IT platform (2025–2026) are under Matias Järnefelt. ## Reinvestment and returns Owner earning power on the model's cash definition ((adjusted EBITDA − 1.26% × revenue) × 0.77): 12.4 (FY2019) → 33.9 (FY2025) → 36.1 (trailing September 2026). The 2.7× rise came from acquisitions, organic growth and FY2019's lower margin. Start and end NOPAT margins are almost equal (16.7% vs 17.0%). Reinvestment rate. Organic growth capital FY2019–FY2024 (one-year lag) was 40.5 against 165.6 of matching cash NOPAT (FY2019–FY2024): 24%. Acquisitions (73.6 of cash including the EOS minority) take total reinvestment to 69%. Dividends took 56% of FCF to equity over FY2019–FY2025 (73.6 of 132.0). Incremental returns (3 November 2026 filing test; table in `07-assumption-register.md`): - Full cycle FY2019→FY2025, lagged: EUR0.49 of growth capital per organic sales euro, which is 44% in engine units (21.5% steady cash margin) and 34% at actual margins (`measuredRoiic2019to2025`). - FY2020→FY2025: EUR0.63, 34% in engine units at the 29.2% margin (`measuredRoiic2020to2025`). The downside uses the same EUR0.63 at its 25.7% margin: 29.8% (`downsideRoiicAtCycleMargin`). - Three-year windows swing from −7% to 67% with utilization. They measure operating leverage, not durable capital efficiency, and aren't used. Operating leverage versus capital efficiency. Incremental adjusted EBIT margins were 30–33% in 2020–2021 and 8–14% in 2024–2025 (`01-business.md`). Capital per sales euro hasn't worsened. What changed is the operating leverage, through overhead. The 45% base ROIIC is a statement about capital, and the measurements support it. The 29.2% margin is a statement about overhead, and the recent record doesn't (`06-risks.md`). Identity check (growth ≈ reinvestment rate × return on new capital). 24% × 34–44% predicts 8–11% organic growth a year. The geometric mean of reported organic growth FY2020–FY2025 is 12.1%. The gap of 1–4 points is explained by acquired businesses' growth counting as organic after year one (EOS, Kirami and ThermaSol's later growth was bought with acquisition capital, not organic capital) and by 2020–2021 demand filling existing capacity. The identity roughly holds, and it says the measured organic ROIIC is, if anything, slightly flattered. Mix and return by use: - Organic: 34–44% after tax at full cycle (above). - Acquisitions: see below. Kirami is the clear success on price (about 2.9× EBITDA); ThermaSol (about 11–13×) is the weakest so far. - Payouts: 56% of FCF to equity over FY2019–FY2025 (61% over FY2021–FY2025); no buybacks for value. - Capacity projects: the FY2021 build earned its cost through the 2022–23 bust (stage 3 cohort, `01-business.md`). Lewisburg's cost isn't separately disclosed and its return is unmeasured. ## Acquisition record | Deal | Date | Stake | Cash price | Target revenue at purchase | Target EBITDA / multiple | Later results | | --- | --- | --- | ---: | ---: | --- | --- | | Sentiotec (Harvia Austria) | Nov 2016 | 100% | n/a (pre-IPO) | n/a | n/a | Controls and heaters; now part of Continental Europe | | Almost Heaven Saunas (US rooms) | Dec 2018 | business assets | No cash consideration (`harvia-fs-2018` p.28) | 2017 c.USD9m | n/a | Became the base of a North American region with 75.8 of 2025 revenue (incl. ThermaSol) | | EOS Group (Germany premium/pro heaters, controls, steam; Russia) | Apr 2020 | 78.6% Germany, 80% Russia | 19.7 (cash flow 18.1 net) | 17.3 (2019) | Debt-free value 25.5 for 100% = 1.5× sales; EBITDA not disclosed | Russia sold Nov 2022. Revenue per issuer increased particularly in heaters and controls | | EOS minority | Jul 2022 | 21.4% | 19.0 | — | Same EBITDA multiple as 2020 → implied 100% value about 88.8, about 3.5× the 2020 value | Price fixed on boom EBITDA; debt-funded | | Kirami (Finland hot tubs) | May 2021 | 100% | 7.0 + 2.5 earn-out (paid Dec 2023) = 9.5 | 22.1 (LTM Apr 2021) | 3.3 adj. EBITDA (15.0%) → 2.9× | Hot tubs and saunas group +1.6% in 2025, +18.8% in 9M 2026 (not separable) | | Sauna-Eurox (Finland stones) | Aug 2021 | 100% | part of 7.6 of 2021 acquisition cash | c.3.2 | n/a | Stones and accessories +8.3% in 2025 | | Phoenix El-Mec (Italy controls) | 2023 | 100% | c.0.3 (2023 acquisition cash 2.8 less the 2.5 Kirami earn-out) | c.1.4 | n/a | Small | | Not itemized | 2025 | — | 1.3 | — | — | Estonia operations sold in 2025 | Aggregate (FY2020–FY2025): 73.6 of cash (including the EOS minority) for about 57 of acquired revenue: EOS 17.3, Kirami 22.1, Sauna-Eurox 3.2, Phoenix 1.4, ThermaSol 13.3. That is about 1.3× sales. Known EBITDA at purchase: Kirami 3.3 and ThermaSol 2.3. EOS's isn't disclosed, so an aggregate EBITDA multiple and a matched multi-year deal yield can't be computed. The bounds that can be stated: - Price paid per sales euro (1.3×) is well below Harvia's own EV/sales today (about 3.7×). - The acquired units are mostly lower-margin categories (rooms, tubs, steam) than heaters; the largest by margin (EOS) is undisclosed. - Group adjusted EBITDA margin at the window's start (2019: 22.9%) and end (2025: 23.4%) is similar. Integration didn't lift the group margin, and it didn't dilute it either. Deployment pace: about 12 a year on average (73.6 / 6), lumpy: three years with near-zero deals. The model assumes no future acquisitions and no acquisition benefit. Management says it is "prepared to act swiftly" on M&A (`harvia-fsb-2025` p.3); debt headroom to the 2.5× ceiling is roughly 63 at trailing adjusted EBITDA (2.5 × 49.6 − 60.8). ## How the model funds growth The model estimates incremental capital as incremental revenue × normalized cash-earnings margin ÷ organic ROIIC. Working capital is funded first within that requirement; growth capex is the remaining positive amount, so working capital isn't charged twice. If working capital alone exceeds the requirement, it is still fully funded. Base and upside use 45%. The downside keeps the measured FY2020–FY2025 capital intensity (EUR0.63 per organic sales euro), which is 29.8% ROIIC at its 25.7% cycle margin (`07-assumption-register.md`). Recurring capital (maintenance, right-of-use and capitalized development cash) uses the source's FY2026 EUR2.9m / EUR230m sales ratio (1.26%) as a company-specific scaling estimate, not a generic maintenance percentage. Reported depreciation is kept separately for the tax and FCF bridge. Dividends start at the known EUR0.77 and grow 5% a year. No acquisition benefit and no buyback accretion are assumed. Cash left after investment, financing and dividends reduces net bank debt, and accumulated cash enters terminal equity once. Lease cash is already charged in recurring capital, so lease debt isn't deducted again. Measured diagnostics: `10-capital-efficiency.md`. # Thesis and valuation Assessment:5 November 2026. Replacement heaters and controls support recurring demand; North American sauna adoption adds growth. Distribution, availability and manufacturing capacity matter more than exclusive technology. Growth still needs working capital and productive capacity, while personnel and other overhead have outgrown sales. A good product franchise does not by itself establish the speed of adoption or margin recovery. The current scenarios are a separately dated judgment, not a newly attributed author forecast. ## Valuation summary Quote EUR 40.65, market time 2026-11-04T16:29:38.000Z; valuation date 5 November 2026. Fixed original exits:26 October 2031/2036. Current reported anchors, normalized earning power and forward estimates remain separate. | Horizon | Bear | Base | Bull | Weighted cash-flow IRR | | --- | ---: | ---: | ---: | ---: | | 5 years | -6.47% | 14.01% | 22.79% | 12.45% | | 10 years | 0.93% | 13.28% | 18.37% | 12.53% | Base initial growth is 10%, fading to 9% over ten years, supporting EUR348.956m sales in year 5. FY2025 regional weights are explicitly allocated to current TTM sales: North America15% growth, Europe5%, APAC/MEA10% produces approximately EUR348.5m after five years. Steam overlaps those regions. Base EBITDA recovers to the 25.7% cycle average; upside alone retains 29.2%. The downside uses 0→5% growth and current 22.683% EBITDA, with measured capital per sales euro unchanged. ## Pre-mortem The main failure is mistaking a plausible mechanism for an evidenced quantity. Adoption could fade while permanent overhead remains; low maintenance estimates and incentive-driven expansion could overstate owner cash. A supplier/competitor price response can destroy the assumed margin and a bad acquisition can consume cash already needed for growth. The adverse scenario represents a specific failure path, not a maximum loss or a frequency estimate. The independent advocates’ strongest surviving attacks are recorded in 12-debate.md. ## Exit multiple The live base/upside use 20× cash NOPAT, after recurring economic capital and operating tax, with debt/claims deducted once. Installed-base replacement, distribution and high incremental productivity support a conditional premium. An illustrative finite calculation requires roughly 12 postexit years at 9% cash growth, 38.5% productivity and 9% required return; that duration is uncertain, not measured. Historical trading multiples are context rather than proof. The adverse 11.8947× is calculated from 15 finite postexit years at 3% nominal growth, 10% required return and 26.0803% productivity, then no-growth value. Its exact annual cash receipts and capital costs are in the model-change record. ## Bear case North American adoption slows, dealers destock and overhead fails to absorb. Flat initial sales gradually recover toward 5% growth while EBITDA stays at the reported 22.68% margin. Capital per incremental sales euro remains EUR0.6325, so model productivity becomes 26.08%; reducing the margin does not silently make growth cheaper. There are no acquisitions. Product recalls, aggressive price competition or a debt-funded bad acquisition can produce losses outside this representative case. ## Variant view The opportunity-backed base no longer requires unidentified sales beyond the regional ledger. Product capital efficiency is supported, while the recovery of overhead and long-run adoption remain contested. At the saved price, a14.01% five-year base offers less margin of safety than the previous 17.81% estimate, and the revised adverse result is negative. There is no hard rerating catalyst or basis to assume the market simply misunderstands the business. Evidence of slower overhead growth and funded regional adoption is needed to strengthen the view. ## Why the opportunity exists The following liquidity/holder observations retain their original 3 November measurement window; they are context, not a new float or short-interest survey. - Size and liquidity: market capitalization about EUR742m. On Nasdaq Helsinki, 9M 2026 trading was EUR260.6m and 6.63m shares, about EUR1.4m or 35,000 shares a day (`harvia-q3-2026` p.8). Shares also trade on Cboe, Aquis, the London Stock Exchange and ITG Posit, which isn't counted here. - Float turnover: excluding the two most recently flagged 5%+ holders (Alecta 5.08%, April 2025; Nordea Funds 5.02%, August 2025) and insiders (3.2%), about 16.2m shares float, about 460 days of Helsinki volume. Liquid enough for institutions; not neglected. - Holders: 33,504 registered shareholders; households about 19% (end-2025); nominee and foreign 38.4% (September 2026, down from 47.7% a year earlier). Domestic funds and pension insurers dominate the register. - Coverage: covered by Nordic sell-side and commissioned research (Inderes publicly); the full analyst list and consensus weren't retrieved. Short interest wasn't checked (FIN-FSA register). - Recent events that drove sellers: the 29 October 2026 Q3 report (adjusted EBIT margin 17.7% vs 19.1%; North American revenue-recognition correction; Q2 delivery catch-up) cut the price from EUR44.05 to EUR38.15 in a day. The year's range was EUR31.05–47.65. Neglect or justified discount? Neither: a widely held, liquid Nordic small-cap at a multiple near its own median. The Q3 drop is a short-term seller event on a real issue (overhead), not evidence of mispricing by itself. ## What would prove us wrong | Observation | Refutes | Where and when | | --- | --- | --- | | Opex (personnel + other operating) grows faster than revenue in FY2026 and again in 2027 | Any margin recovery above the 25.7% cycle average (the upside path) | FY2026 results 10 February 2027; Q1–Q3 2027 interim reports | | Trailing adjusted EBITDA to September 2027 below 56 (downside band; base 60.4) | Margin recovery and base growth | Q3 2027 report, already a frozen prospective check | | Gross margin (1 − materials/revenue) falls below 62% | The "overhead, not product" diagnosis; points to price competition | Each annual income statement | | North American organic growth below 10% for two consecutive quarters while US housing is flat | The health-adoption explanation of NA demand | Quarterly regional revenue tables | | Largest customer share falls sharply or a private-label sauna program appears | Counterparty incentives (stage 4) | Annual report note 2.1; retailer announcements | | Another acquisition above about 10× EBITDA, or one funded beyond 2× leverage | Capital-allocation discipline | Stock exchange releases | | Working capital per organic sales euro rising above 0.4 | The measured capital intensity behind the 45% ROIIC | Annual cash-flow statements | | CEO buys shares in size or the LTI adds a return-on-capital metric | (Favorable) the agency-cost concern | Managers' transactions; 2027 remuneration report | Sources: as cited; `harvia-fsb-2019` … `harvia-fsb-2025`, `harvia-annual-report-2025`, `harvia-q3-2026`, `harvia-q3-2026-presentation`, `masco-10k-2023-sauna360`, `tulikivi-fy2025`. Formulas are in `valuation/calculations.json`; inputs and their evidence in `valuation/model.json`. The [independent debate](12-debate.md) links each surviving attack to the relevant reporting event. Earlier frozen predictions remain unchanged; new debate-vintage forecasts below are stored separately rather than backfitted to realized outcomes. [Same-price/date returns, sensitivities and claim judgments](../updates/2026-11-05-debate-review.json); [annual model-change arithmetic](../updates/2026-11-05-debate-model-change.json). # Risks and disconfirming evidence Assessment: 5 November 2026. The independent debate sizes the risks below against current scenarios. These are business and financing mechanisms, not forecasts of exact failures or a worst-case loss bound. ## Surviving attacks Rebase regional weights without adding steam twice. Sales opportunity rather than a desired IRR determines growth. The reference author forecast/calibration remains intact. Evidence: harvia-q3-2026 pp.6–8; research/08-underwriting-plan.md opportunity ledger. Use0→5% growth and current 22.68% EBITDA. Derive26.08% model productivity at unchanged 0.6325 capital per organic sales euro; do not retain convenient 30% ROIIC. Finite15year/3%growth/10%requiredreturn tail gives 11.8947x. The earlier 12.3–12.35x was an illustrative diagnostic, not the exact recomputation. Evidence: harvia-q3-2026 pp.4,8,11–13; research/06-risks.md; named capitalPerOrganicSale2020to2025. Historical share-price median is not the proof. Shorter duration remains an adverse path and prevents high conviction. ## Evidence gaps and conviction Moderate in installed-base/replacement economics; low in exact adoption speed, overhead absorption and postexit duration. The revised adverse path represents surviving attacks rather than merely labeling them. Reassessed changed joint scenarios. Retain coarse existing horizon weights; there is no new empirical frequency evidence for a precise odds change. Slower central demand and no adverse margin recovery are expressed in operating paths, not by tuning probabilities to returns. ## What would change the view 10February2027 FY2026 results: regional growth, gross margin/opex, Lewisburg cost/ramp and cash conversion; subsequently reported FY2027 regional sales and operating cash. [Full bull case, bear case and rebuttals](12-debate.md); [same-price/date financial effects](../updates/2026-11-05-debate-review.json). Existing frozen predictions are preserved, with the new debate vintage dated separately. # Thesis-to-model register Current assessment:5 November 2026. The source calibration is preserved separately. Reported facts, normalization estimates and new forecast judgments keep their own evidence labels. ## Starting financial bridge | Model input | Value | Classification | Economic definition and evidence | | --- | ---: | --- | --- | | revenue | 218.878 | derived reported | FY2025 + 9M2026 − 9M2025 using detailed reported figures. The Q2 deliveries completed in Q3 are already included; no second backlog addition. This is a TTM proxy, not reported October revenue. Sources: harvia-q3-2026 | | ebitda | 49.649 | derived reported | Matching trailing adjusted EBITDA from the detailed thousands table. Includes the North America timing correction; no additional EUR0.8m addback or continuing growth-opex addback. Sources: harvia-q3-2026 | | shares | 18.830 | dilution proxy | September issued shares less treasury shares plus Q3 share-plan dilution allowance (0.140m). Current proxy; not a full future grant-by-grant valuation. Sources: harvia-q3-2026 | | netDebt | 60.828 | pro forma estimate | September bank debt less September cash, plus the October dividend once. Economic lease cash is charged in recurring capital, so lease debt is excluded. Unreported post-September operating cash is not invented. Sources: harvia-q3-2026, harvia-dividend-calendar | | depreciation | 8.223 | derived reported | FY2025 + 9M2026 − 9M2025 income-statement D&A. Used for tax/FCF, separately from economic sustaining capital. Sources: harvia-q3-2026 | | financingExpense | 3.410 | derived reported proxy | Trailing finance costs less finance income, excluding fair-value movements. Hold constant as a refinancing assumption, with no additional credit for growing cash. Sources: harvia-q3-2026 | | workingCapitalRate | 0.259 | derived current ratio | September working capital / updated TTM sales. This seasonal high is a conservative funding proxy, not a recurring quarterly cash outflow or free opening cash release. Sources: harvia-q3-2026 | | dividendPerShare | 0.770 | reported policy anchor | 0.77 annual dividend paid in 2026. Future growth and payment dates are explicitly estimated. Sources: harvia-dividend-calendar | | otherClaims | 1.635 | carried book-value estimate | September current and non-current employee-benefit obligations, carried once as a separate terminal claim. Minority earnings are already removed; no second NCI book-value deduction. Sources: harvia-q3-2026 | ## Scenario assumptions Base initial growth is 10%, fading to 9% over ten years, supporting EUR348.956m sales in year 5. FY2025 regional weights are explicitly allocated to current TTM sales: North America15% growth, Europe5%, APAC/MEA10% produces approximately EUR348.5m after five years. Steam overlaps those regions. Base EBITDA recovers to the 25.7% cycle average; upside alone retains 29.2%. The downside uses 0→5% growth and current 22.683% EBITDA, with measured capital per sales euro unchanged. The live base/upside use 20× cash NOPAT, after recurring economic capital and operating tax, with debt/claims deducted once. Installed-base replacement, distribution and high incremental productivity support a conditional premium. An illustrative finite calculation requires roughly 12 postexit years at 9% cash growth, 38.5% productivity and 9% required return; that duration is uncertain, not measured. Historical trading multiples are context rather than proof. The adverse 11.8947× is calculated from 15 finite postexit years at 3% nominal growth, 10% required return and 26.0803% productivity, then no-growth value. Its exact annual cash receipts and capital costs are in the model-change record. ### base Regional demand bridge supports initial 10% growth fading to 9%, about EUR349m in year 5: current TTM sales allocated by FY2025 regional weights, North America15%, Europe5%, APAC/MEA10%. Steam is included in the regions. Retain the 25.7% cycle-margin recovery and measured capital intensity; growth above this needs additional identified opportunities. | Driver | Value/path | Reason and sources | | --- | --- | --- | ### downside Adverse adoption/destocking with 0% initial growth recovering toward 5%, no recovery from current 22.68% EBITDA margin, and unchanged measured capital per sales euro. Terminal11.89x is a finite 15-year/3%nominal growth/10%required-return judgment, followed by no-growth value; no claim that this is a loss bound. | Driver | Value/path | Reason and sources | | --- | --- | --- | | targetMargin | 0.227 | Adverse adoption/destocking with 0% initial growth recovering toward 5%, no recovery from current 22.68% EBITDA margin, and unchanged measured capital per sales euro. Terminal11.89x is a finite 15-year/3%nominal growth/10%required-return judgment, followed by no-growth value; no claim that this is a loss bound. Sources: harvia-fsb-2019, harvia-fsb-2020, harvia-fsb-2021, harvia-fsb-2022, harvia-fsb-2023, harvia-fsb-2024, harvia-fsb-2025, harvia-annual-report-2025, harvia-q3-2026 | ### upside Faster near-term adoption begins at 17% and fades to the same 9% long-run rate. Keep 29.2% normalized EBITDA, 45% ROIIC and 20x cash NOPAT. Favorable demand does not itself establish a higher terminal multiple or a permanently longer growth runway. No M&A uplift is assumed. | Driver | Value/path | Reason and sources | | --- | --- | --- | ## Evidence that would change the inputs 10 February 2027 FY2026 results: regional growth, gross margin/opex, Lewisburg cost/ramp and cash conversion; subsequently reported FY2027 regional sales and operating cash. Reassessed changed joint scenarios. Retain coarse existing horizon weights; there is no new empirical frequency evidence for a precise odds change. Slower central demand and no adverse margin recovery are expressed in operating paths, not by tuning probabilities to returns. [Detailed claim-by-claim judgment](12-debate.md). All shared-engine calculations retain stable fact IDs and audit links; original source arithmetic is not a current forecast. # Underwriting plan This plan replaces `08-underwriting-method.md` (driver map, 27 October) and `12-q3-2026-underwriting-plan.md` (Q3 decision record, 29 October). The opportunity ledger and scenario review were written on 3 November 2026 after the stage 2–7 work, before returns were recalculated. Money is EUR millions. ## Opportunity ledger | Opportunity | Starting point | Customer need and why Harvia wins | Reachable revenue in five years | Investment and cash cost | Timing and limits | Evidence against | | --- | --- | --- | --- | --- | --- | --- | | North American adoption: rooms and heaters (volume, share) | NA revenue 75.8 (2025), +21.9% in 9M 2026; Almost Heaven D2C, retail and dealers | Home sauna for health; Harvia offers rooms plus heaters, US-made (about 70% of US end products), and the largest retailer relationship (about 11% of group sales) | About 150–170 if NA grows 15% a year, half of the base case's increment | Lewisburg expansion (in 2025–26 capex of 14.8 and 7.9; cost not itemized); working capital at about 0.29 per sales euro | Lewisburg completes Q4 2026. US consumer confidence low, housing weak | Steam (housing-linked) is falling; room pricing is campaign-driven; new entrants (Tulikivi ETL heaters, imports); Sauna360/Masco | | European replacement and renovation (replacement, share) | Northern Europe 46.6 (+12.8% 9M 2026), Continental 55.6 (+5.1%) | Replacement of heaters in a large installed base; Harvia's share of heaters and components is over 20% | 4–6% a year from replacement and modest share gains: about 125–135 in 2031 | Low: Muurame has spare capacity from 2021; working capital | Finnish construction weak; Northern Europe fell in 2023 and 2024 | Energy prices cut sauna use (2022); Q2 2026 IT disruption showed delivery risk | | APAC & MEA (new markets) | 21.0 (2025), +9.6% in 9M 2026 | Growing awareness in China, Japan, Australia | 30–35 at 10% a year | Sales organizations (Japan company set up 2023), Guangzhou factory | Middle East (about 2% of group revenue) on hold | Small base; project-driven | | Price and mix (price) | Gross margin 64.2% (2025) vs about 60% in 2019–22 | Premium EOS, digital controls (MyHarvia sensor), heater/control mix | Supports the margin; not a separate revenue line | Product development (R&D expensed 2.1 a year) | Campaigns and tariffs offset part | Room mix in NA dilutes; Black Friday discounting | | Overhead absorption (margin) | Opex 42% of sales (2025); adjusted EBITDA margin 22.7% trailing | Operating leverage once the IT platform, regional teams and Lewisburg stop growing faster than sales | 29.2% EBITDA needs an opex ratio near 36%: opex growing 4–5 points a year slower than revenue for about four years | None in capital; it is a spending choice | The program continued in Q3 2026 (headcount +12%) | Four years of opex outgrowing revenue; no metric in pay rewards it | | Steam (ThermaSol, EOS) | 17.2 (2025), −8.2% in 9M 2026 | Steam rooms and showers in US renovation and commercial projects | Flat to slightly up | Little | Tied to US housing | Already declining; possible goodwill pressure | | Acquisitions | 73.6 deployed FY2020–FY2025, about 12 a year, lumpy | Fragmented sauna and spa market; Harvia paid about 1.3× sales | Not modeled | Debt headroom about 63 to the 2.5× ceiling | Management says it is ready to act | ThermaSol at about 11× EBITDA has underperformed; EOS minority bought at a boom multiple | Overlap. NA heater revenue sits in the first row, not in price and mix; European replacement excludes NA. The margin line is a cost line, not revenue. Acquisitions are excluded from all cases: cash that can't be placed reduces net debt, following the source and the company's dividend policy. Management targets are a cross-check, not the demand forecast. The issuer targets 10% average annual revenue growth and adjusted EBIT above 20%. The current base now starts at 10% growth and fades to 9%, because the regional opportunity bridge supports that scale independently. The 25.7% cycle EBITDA margin remains partial overhead recovery; the 29.2% author normalization remains favorable only. Current returns and sensitivities are in the dated debate review rather than earlier-price examples. ## Scenario mechanisms Current5 November 2026 judgment: Base initial growth is 10%, fading to 9% over ten years, supporting EUR348.956m sales in year 5. FY2025 regional weights are explicitly allocated to current TTM sales: North America15% growth, Europe5%, APAC/MEA10% produces approximately EUR348.5m after five years. Steam overlaps those regions. Base EBITDA recovers to the 25.7% cycle average; upside alone retains 29.2%. The downside uses 0→5% growth and current 22.683% EBITDA, with measured capital per sales euro unchanged. North American adoption slows, dealers destock and overhead fails to absorb. Flat initial sales gradually recover toward 5% growth while EBITDA stays at the reported 22.68% margin. Capital per incremental sales euro remains EUR0.6325, so model productivity becomes 26.08%; reducing the margin does not silently make growth cheaper. There are no acquisitions. Product recalls, aggressive price competition or a debt-funded bad acquisition can produce losses outside this representative case. The forward opportunity ledger determines the operating path before return comparisons. It is not a projection of past CAGR or management targets. Capital, timing and ownership claims remain matched to the revenue/cash perimeter. Reassessed changed joint scenarios. Retain coarse existing horizon weights; there is no new empirical frequency evidence for a precise odds change. Slower central demand and no adverse margin recovery are expressed in operating paths, not by tuning probabilities to returns. ## Current absolute scale and terminal opportunities 5-year base: revenue 348.957, owner cash 65.526, exit cash earnings 65.526 (EURm). 10-year base: revenue 542.406, owner cash 101.938, exit cash earnings 101.938 (EURm). The live base/upside use 20× cash NOPAT, after recurring economic capital and operating tax, with debt/claims deducted once. Installed-base replacement, distribution and high incremental productivity support a conditional premium. An illustrative finite calculation requires roughly 12 postexit years at 9% cash growth, 38.5% productivity and 9% required return; that duration is uncertain, not measured. Historical trading multiples are context rather than proof. The adverse 11.8947× is calculated from 15 finite postexit years at 3% nominal growth, 10% required return and 26.0803% productivity, then no-growth value. Its exact annual cash receipts and capital costs are in the model-change record. No return target determines this path. Source locators and disagreements are in 12-debate.md and 07-assumption-register.md. # Method self-evaluation - 27 October 2026 This review reread the investment-cycle and valuation arguments, inspected the original page 96 table, and checked the H1 2026 issuer figures. It is not an independent blind replication or a new complete company/industry review. ## What failed The prior prompt required source reproduction but did not require an explicit economic test distinct from that reproduction. The existing check rebuilt cash NOPAT and FCF from source operating inputs. That proved the earnings definitions and arithmetic, while still accepting future EBITDA and investment rows as inputs. It did not prove that the research process would choose a coherent forecast or downside. The current downside also permanently reduced normalized EBITDA from 29.2% to 25.5% without identifying separate evidence of impaired unit economics. The upside combined faster growth, a stronger terminal growth rate and 22x valuation. Both departed from the causal case more than their labels made apparent. ## Revised test `valuation/method-check.json` holds a separate, simplified ten-year economic reconstruction at the original EUR38 price and FY2025 opening balance sheet. The engine generates future revenue, margins, NOPAT, investment and cash. Numerical benchmarks are used only after forecasting. A regression test changes those benchmarks and confirms that the generated forecast is unchanged. The reconstruction still accepts explicit author judgments: normalized margin, organic return on capital, tax, dividends and valuation. Those judgments are explained in the driver register; the test does not claim to derive their exact values uniquely from observations. Historical figures, source-derived recurring-capital and depreciation proxies, simplified growth fade and margin timing remain visible. The range is a sensitivity to a documented 15%-17% near-term growth cadence, not a probability interval or optimized fit. The initial high-cadence reconstruction gave 17.34%, above 16.7%. Its ending revenue was EUR673.5m versus EUR637.2m in the report. That exposed an overly strong simplified growth path, not an excuse to alter taxes or the multiple. Keeping the source-supported cadence range gives the following comparison: | Check | Reconstructed result | Author reference | Interpretation | | --- | --- | --- | --- | | Exact printed valuation bridge | 16.7303% | 16.7% | Matches printed rounding. | | Exact operating-input cash bridge | 16.7349% | 16.7% | Matches printed rounding; annual maximum difference EUR0.139m. | | Driver-generated base range | 16.37%-17.34% | 16.7% | Economically consistent range; no claim of an exact independently rediscovered point. | | Ending revenue | EUR616.9m-EUR673.5m | EUR637.2m | Original outcome lies within the mechanically generated output range. | | Ending cash NOPAT | EUR132.7m-EUR144.9m | EUR137.3m | Same. | | Cumulative FCF | EUR638.9m-EUR674.9m | EUR649.9m | Same; future FCF was generated, not inserted. | | Ending net cash | EUR391.0m-EUR427.0m | EUR411.1m | Same; distributions and retained cash reconcile. | Run `npm run thesis:check -- harvia` to reproduce the numerical consistency check. `npm test` also verifies that slower growth reduces required investment without silently damaging normalized margins, worse capital efficiency consumes more cash, and a changed expected return or cash target cannot feed back into forecast inputs. ## Remaining limits The exact growth and margin cadence is not determined by the prose alone. A range containing the source result is weaker evidence than independently producing the exact annual curve. This check supports causal consistency; the separate original table remains the precise reference. No tolerance was widened to relabel the economic reconstruction as an exact match. The input range was assembled during an open-book review after the report was known. The grower adapter does not explicitly model capacity cohorts or ramp delays. The report's project analysis informs its ROIIC judgment, but the saved check is not a new project-level replication. A later capacity-cohort model requires its own evidence and tests. No M&A uplift or scenario probabilities are inferred. Current five-year returns use updated financials, a different starting price/date, a fresh forecast window and dated dividends. They cannot be made to equal the historical 16.7% merely by treating that figure as a target. # Capital productivity and reinvestment Assessment:5 November 2026. Values below come directly from the same shared forecast used by the dashboard. Money is in EUR millions. | Horizon | Scenario | Capital-return definition | Return/proxy | Period-weighted reinvestment | Growth capital | Matching earnings | Funding residual | | --- | --- | --- | ---: | ---: | ---: | ---: | ---: | | 5 years | downside | Organic ROIIC · model implied | 26.08% | 4.27% | 7.842 | 183.809 | 2.13e-14 | | 5 years | base | Organic ROIIC · model implied | 38.45% | 24.28% | 63.679 | 262.278 | 2.84e-14 | | 5 years | upside | Organic ROIIC · model implied | 45.00% | 31.54% | 107.805 | 341.817 | 1.42e-14 | | 10 years | downside | Organic ROIIC · model implied | 26.08% | 9.76% | 38.559 | 394.971 | 3.55e-14 | | 10 years | base | Organic ROIIC · model implied | 38.45% | 22.84% | 158.382 | 693.578 | 5.68e-14 | | 10 years | upside | Organic ROIIC · model implied | 45.00% | 24.78% | 249.659 | 1007.608 | 5.68e-14 | Historical capital productivity is distinct from this table’s model-implied organic productivity. FY2019–2025 and FY2020–2025 lagged capital/sales measurements remain in calculations.json. Base capital intensity remains approximately 0.49 per incremental sales euro; adverse remains 0.6325. Lower adverse EBITDA changes productivity to 26.08%, not the required physical/working capital per sales euro. Total earning-power change can include margin recovery and is not a newly measured ROIIC. Reinvestment equals total new growth investment divided by total matching earnings over the period, not the average of annual percentages. Cash generation less distributions/old claims plus opening cash drawn and net borrowing reconciles to new investment. No extra borrowing or invented working-capital release is introduced. All six funding ledgers reconcile within 0.000001 model millions. Sources and definitions:04-capital-allocation.md, valuation/calculations.json and the shared ledger; docs/method/cash-flow-and-reinvestment.md. [Dated review](../updates/2026-11-05-debate-review.md). # Scenario probabilities Assessment:5 November 2026, after the independent debate. Low-confidence subjective joint-path weights; no claim of empirical calibration or source-author odds. | Horizon | Bear | Base | Bull | | --- | ---: | ---: | ---: | | 5 years | 25.00% | 55.00% | 20.00% | | 10 years | 30.00% | 45.00% | 25.00% | Reassessed changed joint scenarios. Retain coarse existing horizon weights; there is no new empirical frequency evidence for a precise odds change. Slower central demand and no adverse margin recovery are expressed in operating paths, not by tuning probabilities to returns. Base initial growth is 10%, fading to 9% over ten years, supporting EUR348.956m sales in year 5. FY2025 regional weights are explicitly allocated to current TTM sales: North America15% growth, Europe5%, APAC/MEA10% produces approximately EUR348.5m after five years. Steam overlaps those regions. Base EBITDA recovers to the 25.7% cycle average; upside alone retains 29.2%. The downside uses 0→5% growth and current 22.683% EBITDA, with measured capital per sales euro unchanged. North American adoption slows, dealers destock and overhead fails to absorb. Flat initial sales gradually recover toward 5% growth while EBITDA stays at the reported 22.68% margin. Capital per incremental sales euro remains EUR0.6325, so model productivity becomes 26.08%; reducing the margin does not silently make growth cheaper. There are no acquisitions. Product recalls, aggressive price competition or a debt-funded bad acquisition can produce losses outside this representative case. Moderate in installed-base/replacement economics; low in exact adoption speed, overhead absorption and postexit duration. The revised adverse path represents surviving attacks rather than merely labeling them. Review on 10 February 2027 FY2026 results: regional growth, gross margin/opex, Lewisburg cost/ramp and cash conversion; subsequently reported FY2027 regional sales and operating cash. Weights sum to 100% independently for each horizon. These paths represent outcome regions and omit permanent-loss tails. Price alone does not change business-path odds; local scenario overrides suspend researched weights. Weighted IRR is calculated from probability-weighted dated shareholder receipts, not by averaging scenario IRRs. [Evidence and rebuttals](12-debate.md); method:docs/method/scenario-probabilities.md. # Bull–bear review — Harvia, 5 November 2026 Independent advocates prepared opening cases without seeing the other draft, then exchanged point-by-point cross-examinations. A separate judge who did not author the prior model assessed the competing claims. Source reconstructions retain their stated provenance; inaccessible private reports were not newly read. Quoted sensitivities test arguments and are not automatically adopted forecasts. Fixed quote EUR 40.65, market time 2026-11-04T16:29:38.000Z; valuation date 2026-11-05. Original forecast exits retained. | Horizon | Before bear/base/bull | After bear/base/bull | Weighted before → after | | --- | --- | --- | --- | | 5 years | 5.33% / 17.81% / 22.79% | -6.47% / 14.01% / 22.79% | 16.34% → 12.45% | | 10 years | 8.24% / 15.71% / 18.37% | 0.93% / 13.28% / 18.37% | 14.73% → 12.53% | ## Bull case As of 5 November 2026. No opposing advocate draft was read. Existing company research, fact/calculation ledgers and exact registered source locators were examined; original issuer sources additionally checked live: Lumine Q3, CSU Q2, Diploma Q3 transcript, NVIDIA CFO commentary, TerraVest Highland/KBK, WOSG FY26. Harvia Q3 PDF retrieval failed; its registered extraction remains the evidence limit. Private source interpretations below are the repository's documented reconstruction, not a new claim to have read an unavailable private original. The strongest bear case is that growing North American adoption is being purchased with recurring overhead, with a dealer/consumer cycle and campaign pricing layered on top. The base's15% initial growth also exceeds the revenue independently reachable from its own opportunity ledger. The existing ledger forecasts NA150–170, Europe125–135 and APAC30–35. These are regions: steam already sits inside them and cannot be added again. Allocate TTM218.878 by FY2025 regional weights 75.8/46.6/55.6/21.0 (explicit unchanged-mix proxy), then NA15%, combined Europe5%, APAC10% for five years: about 348.5m. The engine's10% initial growth fading 9% produces 348.96m versus 415.23m currently. I concede 15% base lacks~66m of named opportunities; a10% initial base is justified by the bottom-up bridge, not management authority. **HARV-B2 — Margin absorption is the strongest legitimate upside.** Lewisburg capacity, IT implementation and regional teams can cease outgrowing sales; heater/control mix and higher gross margin can release operating leverage. Test29.2% EBITDA with ROIIC adjusted to preserve the base's measured growth-capital intensity; no free productivity uplift. The outside view is 25.7% through-cycle EBITDA; management's continuing growth-spend preference prevents making 29.2% central today. **HARV-B3 —20× has an economic duration cost.** Heaters participate in rooms sold by others; replacement and distribution can outlast initial adoption.20× at 9% required return, 9% post-exit growth and 38.5% marginal return requires~12 additional high-return years. A15× test quantifies shortened runway. It is not a recommendation to retain 20× merely because historical multiples averaged more. Distribution concentration, competing heaters and health-fashion risk can truncate the duration. No acquisition upside is credited. ### Isolated shared-engine effects Saved quote 40.65; market timestamp 2026-11-04T16:29:38.000Z. Fixed valuation date 2026-11-05; model 09878f061ae1; exits remain the original model vintage. These are sensitivities, not joint forecasts or probability revisions. Baseline base IRR 17.81% / 15.71% (5/10 years). | Claim | Exact input/path | 5-year base | 10-year base | Change, pp (5/10) | | --- | --- | ---: | ---: | --- | | HARV-B1 | `cases.base.inputs.growthRate` | 19.33% | 16.67% | +1.52 / +0.96 | | HARV-B2 | `cases.base.inputs.targetMargin + roiic (constant capital intensity)` | 21.19% | 17.38% | +3.38 / +1.67 | | HARV-B3 | `cases.base.inputs.exitMultiple` | 11.65% | 13.03% | -6.15 / -2.68 | | HARV-B4 | `cases.base.inputs.growthRate` | 14.01% | 13.28% | -3.80 / -2.43 | ## Bear case ### Strongest opposing case Heaters and controls participate in both new sauna adoption and later replacement; Harvia's distribution and US manufacturing can earn attractive returns without owning the whole room market. Gross margin is strong and the matched capital-intensity reconstruction through a boom and bust supports economically attractive organic growth. ### Bear case The present base assumes more demand than its own opportunity ledger explains. Starting from EUR218.9m trailing sales, identified North American adoption, European replacement/share gains, APAC and flat steam total onlyEUR320–360m in 2031. The15→9% base reaches roughlyEUR415m. Filling the gap needs faster conversion, share or pricing with named evidence; saying the latest organic growth was 16% is trend extrapolation, not a reachable-demand bridge. A10→9% sensitivity is consistent with the ledger's scale, chosen independently of the issuer's similar target. The downside is internally inconsistent with the written bear. The modeled downside still grows 7% every year and lifts 22.7% EBITDA to 25.7%; it is slower compounding, not the described dealer destocking/adoption pause with persistent overhead. Yet Harvia itself had−7.7% and−9.4% organic years in 2022–2023. A meaningful adverse path can hold sales flat initially, retain current overhead and use a trough multiple without alleging permanent product obsolescence. [harvia-fsb-2022 p.2; harvia-fsb-2023 p.2; research/05-thesis.md Bear case; valuation/model.json downside.] High gross margin does not automatically translate into owner-cash margin. Personnel and operating cost grew faster than sales for several years, despite a record gross margin. Continued investment is necessary to create the assumed adoption; simultaneously counting rapid growth and removing its sales/service cost overstates operating leverage. The base's25.7% is already a recovery. The upside 29.2% requires overhead absorption not yet observed. [harvia-fsb-2023 through harvia-fsb-2025 income statements; harvia-q3-2026 financials pp.15–23; harvia-annual-report-2025 remuneration pp.46–52.] A20x exit exceeds the 11.1x no-growth value at 9% required return and requires 4–5% growth with attractive reinvestment after 2031. The current record does not establish how many additional years of North American adoption/share gain support that premium. The2022 trough 12.3x shows the price can discount a good product franchise severely when channel conditions reverse. Historical medians are observations, not a terminal entitlement. [research/05-thesis.md Exit multiple; harvia-fsb-2019…2025.] ### Contested claims - HARVIA-B1 —Reduce base initial growth to 10%, fading 9%, to close the identified scale gap; if the bull supplies incremental reachable demand, reconsider. - HARVIA-B2 —Keep EBITDA at today's22.7%, rather than assuming recovery. This tests persistent investment cost; it is not a claim that gross product economics weakened. - HARVIA-B3 —Test15x on otherwise unchanged base: terminal assumptions matter much more than the notebook's stale 15.5% sensitivity suggests at the present model/quote. - HARVIA-B4 —Use the already-written bear as an explicit stress:0→5% growth, 22.7% margin, 30% ROIIC,12.3x. A negative return is not an excuse to discard it. It is not a probability-calibrated worst case. The current issuer PDF was inaccessible through the web tool in this run; page-specific figures above come from the existing source-linked reconstruction. No new channel sample was performed. ### Shared-engine effects Fixed saved quote EUR40.65; market timestamp 2026-11-04T16:29:38.000Z; valuation date 5 November 2026; fixed original exit calendar. Base before: 17.81% / 15.71% (5/10years). These are one-change sensitivities to the base, except named bundles; existing scenario probabilities are not revised or claimed as validated. | Claim | Change | 5-year base IRR | 10-year base IRR | 5-year effect | 10-year effect | | --- | --- | ---: | ---: | ---: | ---: | | HARVIA-B1 | Base growth 10% fading 9% | 14.01% | 13.28% | -3.80pp | -2.43pp | | HARVIA-B2 | EBITDA stays at current margin | 14.72% | 14.19% | -3.09pp | -1.52pp | | HARVIA-B3 | Base exit 15x | 11.65% | 13.03% | -6.15pp | -2.68pp | | HARVIA-B4 | Written bear:0→5% growth, 22.7% margin, 30% ROIIC,12.3x exit | -5.87% | 1.25% | -23.68pp | -14.46pp | ## Rebuttals ### Bull response 5 November 2026. Written after independently frozen openings were exchanged. Every numbered bear claim is addressed below; original bull openings and sensitivities remain unchanged. 1. **Bear10→9% growth: concede, with an independently rebuilt bridge.** FY2025 regions sum 199.0m: NA75.8, Northern Europe46.6, Continental55.6, APAC/MEA21.0. Allocate current TTM218.878 at those regional weights (an explicit unchanged-mix estimate), then compound NA15%, combined Europe5%, APAC10% for five years:~348.5m. Steam is inside regions, never an extra 17m. Shared engine 10→9% gives 348.96m; present 15→9% gives 415.23m. No named~66m incremental opportunity bridges the excess. My17% is an upside sensitivity only; recalibrate its absolute market/share implications rather than treating a positive IRR as evidence. 2. **Bear flat 22.7% margin: concede as adverse; partly rebut as central.** Continued growth overhead is observed; gross margin and completion of current capacity/IT support possible absorption (`harvia-q3-2026`, pp6–8; `harvia-fsb-2025`, p8; source reconstruction).25.7% cycle-average base is a reasoned partial recovery, not proof; 29.2% belongs in upside until costs demonstrably lag sales. Do not eliminate necessary growth opex while retaining all demand. 3. **Bear15× exit: concede duration sensitivity; no automatic haircut.**20× is supported only if replacement, distribution and attractive incremental economics persist for roughly 12 post-exit high-return years in my stated diagnostic. No source empirically establishes that life.15× belongs to shorter/lower-return paths. A method correction should state the duration risk; replacing 20× merely because a lower multiple lowers returns is not analysis. 4. **Bear written adverse path versus modeled mild downside: concede substantive inconsistency.**0→5% growth with current 22.7% EBITDA better matches adoption pause/dealer destocking than 7% perpetual growth plus margin recovery. Preserve the measured~0.63 capital per organic sales euro: ROIIC must fall with margin (~26.1%, not an independently chosen 30%).12.3× is an observed 2022 trough stress, not an economic entitlement. A finite diagnostic at 10% required return, 3% nominal post-exit growth,~26% incremental return and 15years gives~12.35×; that duration is a disclosed judgment, not evidence precise 12.3× is right. Reassess weights after changing scenario definitions. Conviction moderate in product economics, lower in the old forecast. ### Bear response 5 November 2026, after independent openings were exchanged. 2. **Bull margin absorption: upside conceded, free capital productivity rejected.** Finished capacity/IT and strong gross margin support possible operating leverage, but current growth overhead is recurring until shown otherwise.25.7% cycle-average partial recovery is defensible central judgment; 29.2% remains upside. Preserve measured growth capital/sales when margins move; this also corrects my opening's convenient 30% ROIIC in the genuine-bear stress. 3. **Genuine downside refinement.** Current reported 22.6834% EBITDA versus 25.70545% old downside implies new ROIIC26.08033483% at unchanged 0.6324826082 capital per incremental sales euro:29.7594% ×(22.6834%−1.26087%)/(25.70545%−1.26087%). With0→5% growth, current margin, 12.3x exit and existing dividend path, shared engine gives−5.9204%/+1.1734% over 5/10years atEUR40.65,5 November. This is a judgment stress grounded in observed destocking and overhead, not a predicted worst case. Both advocates agree old 7% growth plus margin recovery failed the written adverse mechanism. 4. **Bull20x duration: risk conceded, automatic haircut rejected.** Historical median is not an entitlement; the bull's12post-exit high-return years are an explicit forward judgment. My15x test also does not establish central fair value. For the adverse 12.3x, an independent finite diagnostic at 10% required return, 3% post-exit growth, 26.08% new-capital return and 15years produces about 12.35x, followed by no-growth value. A perpetual cross-check at 2.7% nominal replacement growth yields 12.28x. These chosen duration/growth judgments corroborate the historical 12.3x trough; neither is a reported forecast. Use the finite one consistently, rather than claiming the trough price proves intrinsic value (`harvia-fsb-2022`, source calibration; stage 9 terminal definition). Reassess joint scenario probabilities after changing base/downside definitions. The original investor's calibration remains untouched; the material changes are this study's revised current demand and risk judgments. Gross-margin deterioration, unabsorbed overhead and lost dealers remain the most important refutations. Current issuer-PDF access failure limits this to the registered extraction and independent arithmetic, not new channel research. ## Judgment Existing estimates receive no presumption merely because they are already in the model. Positive company evidence, the relevant outside view and the capital/funding bridge determine each verdict; neither arbitrary conservatism nor an unsupported favorable sensitivity replaces it. The evidence and reconstruction below explain numerical changes or explicit no-change conclusions claim by claim. | Contested claim | Verdict | Evidence/locator | Reason and model action | | --- | --- | --- | --- | | bear advocate: Current base exceeds independently identified regional demand | adopt | harvia-q3-2026 pp.6–8; research/08-underwriting-plan.md opportunity ledger | Rebase regional weights without adding steam twice. Sales opportunity rather than a desired IRR determines growth. The reference author forecast/calibration remains intact. | | bear advocate: The written genuine bear is absent from the actual scenario | adopt-modified | harvia-q3-2026 pp.4,8,11–13; research/06-risks.md; named capitalPerOrganicSale2020to2025 | Use0→5% growth and current 22.68% EBITDA. Derive26.08% model productivity at unchanged 0.6325 capital per organic sales euro; do not retain convenient 30% ROIIC. Finite15year/3%growth/10%requiredreturn tail gives 11.8947x. The earlier 12.3–12.35x was an illustrative diagnostic, not the exact recomputation. | Reassessed changed joint scenarios. Retain coarse existing horizon weights; there is no new empirical frequency evidence for a precise odds change. Slower central demand and no adverse margin recovery are expressed in operating paths, not by tuning probabilities to returns. ## Conviction Moderate in installed-base/replacement economics; low in exact adoption speed, overhead absorption and postexit duration. The revised adverse path represents surviving attacks rather than merely labeling them. These are forward estimates with surviving attacks, not confirmed outcomes. Neither numerical calibration, a completed checklist nor passing software tests establishes investment skill. Reported facts stay separate from forward judgments, and adverse paths are not worst-case bounds. **What would change the view:** 10 February 2027 FY2026 results: regional growth, gross margin/opex, Lewisburg cost/ramp and cash conversion; subsequently reported FY2027 regional sales and operating cash. [Exact inputs, isolated sensitivities, model hashes and same-price/date before/after returns](../updates/2026-11-05-debate-review.json). ## Public Primary & Secondary Sources - `harvia-h1-2026` (primary): Harvia H1 2026 financial review — https://harviagroup.com/wp-content/uploads/2026/08/Harvia-Plc-Interim-Report-Q2-2026-ENG.pdf (published 2026-08-06) - `harvia-dividend-calendar` (primary): Harvia official dividend calendar — https://harviagroup.com/harvias-share/dividend/ - `harvia-report-calendar` (primary): Harvia Q3 2026 publication notice — https://harviagroup.com/harvia-will-publish-its-interim-report-for-january-september-2026-on-thursday-29-october-2026-at-around-900-a-m-eet/ (published 2026-10-22) - `harvia-newsroom` (primary): Harvia reports, presentations and issuer releases — https://harviagroup.com/reports-and-presentations/ - `capital-method-roiic` (methodology): Mauboussin and Callahan: Return on Invested Capital — https://www.morganstanley.com/im/publication/insights/articles/article_returnoninvestedcapital.pdf (published 2022-10-06) - `capital-method-cash` (methodology): Dan Shuart, Eagle Point Capital: A Framework For Spotting Value Traps — https://eaglepointcapital.substack.com/p/spotting-value-traps-the-anti-compounders (published 2023-08-26) - `scenario-method` (methodology): Aswath Damodaran: Probabilistic Approaches in Valuation, chapter 33 — https://pages.stern.nyu.edu/~adamodar/pdfiles/val3ed/c33.pdf - `harvia-releases` (primary): Harvia issuer release index — https://harviagroup.com/releases/ - `pool-q3-2026` (primary): POOLCORP Q3 2026 results — https://ir.poolcorp.com/pool-corporation-reports-third-quarter-results-confirms-annual-earnings-guidance-range/ (published 2026-10-22) - `kesko-september-2026` (primary): Kesko September 2026 sales — https://www.kesko.fi/en/media/news-and-releases/investor-news/2026/keskos-sales-in-september/ (published 2026-10-14) - `sauna360-cpsc-2026` (primary): CPSC Sauna360 infrared/hybrid sauna recall — https://www.cpsc.gov/Recalls/2026/Sauna360-Recalls-Infrared-Saunas-Hybrid-Infrared-Saunas-and-Infrared-Kits-Due-to-Risk-of-Serious-Injury-or-Death-from-Fire-and-Burn-Hazards (published 2026-09-17) - `masco-q3-2026-calendar` (primary): Masco Q3 2026 publication notice — https://investor.masco.com/news-stream/news-details/2026/Masco-Corporation-Announces-Date-for-Earnings-Release-and-Conference-Call-for-2026-Third-Quarter-2026-4aOPe3SJ_Z/default.aspx (published 2026-09-28) - `harvia-q3-calendar-gnw-2026` (primary): Harvia Q3 publication notice — issuer wire copy — https://rss.globenewswire.com/news-release/2026/10/22/3386121/0/en/harvia-will-publish-its-interim-report-for-january-september-2026-on-thursday-29-october-2026-at-around-9-00-a-m-eet.html (published 2026-10-22) - `harvia-ownership-index` (primary): Harvia flagging notification register — https://harviagroup.com/shareholders/flagging-notifications/ - `masco-q3-2026-release` (primary): Masco Q3 2026 results — original issuer release — https://www.businesswire.com/news/home/20261028013042/en/Masco-Corporation-Reports-Third-Quarter-2026-Results (published 2026-10-28) - `masco-q3-2026-10q` (primary): Masco Q3 2026 Form 10-Q — https://www.sec.gov/Archives/edgar/data/62996/000006299626000033/mas-20260930.htm (published 2026-10-28) - `harvia-q3-2026` (primary): Harvia Q3 2026 interim report — https://harviagroup.com/wp-content/uploads/2026/10/Harvia-Plc-Interim-Report-Q3-2026-ENG.pdf (published 2026-10-29) - `harvia-q3-2026-presentation` (primary): Harvia Q3 2026 results presentation — https://harviagroup.com/wp-content/uploads/2026/10/Harvia-Plc-Q3-2026-ENG.pdf (published 2026-10-29) - `kesko-q3-2026` (primary): Kesko Q3 2026 interim report — https://www.kesko.fi/en/media/news-and-releases/news/2026/kesko-interim-report-1.1.30.9.2026-keskos-profit-improved-significantly--technical-trade-continues-as-the-driver/ (published 2026-10-29) - `harvia-manager-lengauer-2026-10-30` (primary): Harvia manager transaction — Markus Lengauer — https://harviagroup.com/harvia-plc-managers-transactions-lengauer/ (published 2026-10-30) - `pentair-q3-2026` (primary): Pentair Q3 2026 results — https://investors.pentair.com/news-releases/news-release-details/pentair-reports-third-quarter-2026-results (published 2026-10-27) - `almost-heaven-fall-sale-2026` (primary): Almost Heaven Fall Favorites promotion — https://almostheaven.com/collections/fall-favorites-sale - `jyu-sauna-humidity-2026` (primary): University of Jyväskylä sauna-study summary — https://www.jyu.fi/fi/uutinen/tutkimus-loyly-lisaa-saunan-kuormittavuutta-kosteus-nostaa-syketta-ja-kehon-lampotilaa (published 2026-07-10) - `almost-heaven-black-friday-2026` (primary): Almost Heaven Black Friday campaign terms — https://almostheaven.com/collections/black-friday-now - `tulikivi-us-etl-2026` (primary): Tulikivi electric sauna heaters meet U.S. ETL requirements — https://tulikivigroup.com/en/tulikivis-electric-sauna-heater-collection-meets-us-etl-requirements/ (published 2026-10-20) - `tulikivi-financial-calendar-2026` (primary): Tulikivi financial reporting in 2026 — https://tulikivigroup.com/en/financial-reporting-in-2026/ (published 2025-11-27) - `harvia-fsb-2019` (primary): Harvia financial statements bulletin 2019 — https://harviagroup.com/wp-content/uploads/2020/02/Harvia-Oyj-Financial-Statements-Bulletin-2019-ENG_FINAL.pdf - `harvia-fsb-2020` (primary): Harvia financial statements bulletin 2020 — https://harviagroup.com/wp-content/uploads/2021/02/Harvia-Oyj-Financial-Statements-Bulletin-2020-ENG.pdf - `harvia-fsb-2021` (primary): Harvia financial statements bulletin 2021 — https://harviagroup.com/wp-content/uploads/2022/02/Harvia-Plc-Financial-Statements-Bulletin-2021-ENG.pdf - `harvia-fsb-2022` (primary): Harvia financial statements bulletin 2022 — https://harviagroup.com/wp-content/uploads/2023/02/Harvia-Plc-Financial-Statements-Bulletin-2022-ENG.pdf - `harvia-fsb-2023` (primary): Harvia financial statements bulletin 2023 — https://harviagroup.com/wp-content/uploads/2024/02/Harvia-Plc-Financial-Statements-Bulletin-2023-ENG.pdf - `harvia-fsb-2024` (primary): Harvia financial statements bulletin 2024 — https://harviagroup.com/wp-content/uploads/2025/02/Harvia-Plc-Financial-Statements-Bulletin-2024-ENG.pdf - `harvia-fsb-2025` (primary): Harvia financial statements bulletin 2025 — https://harviagroup.com/wp-content/uploads/2026/02/Harvia-Plc-Financial-Statements-Bulletin-2025-ENG.pdf - `harvia-fs-2021` (primary): Harvia Board of Directors report and consolidated financial statements 2021 — https://harviagroup.com/wp-content/uploads/2022/03/Harvia_EN_FinancialStatements_2021_web.pdf - `harvia-fs-2018` (primary): Harvia consolidated financial statements 2018 — https://harviagroup.com/wp-content/uploads/2020/02/Consolidated-Financial-Statements-2018.pdf - `harvia-annual-report-2025` (primary): Harvia Annual Report 2025 (incl. Board report, IFRS statements, governance statement, remuneration report) — https://harviagroup.com/wp-content/uploads/2026/03/Harvia-Plc-Annual-Report-2025.pdf - `harvia-kirami-2021` (primary): Harvia acquires still-water hot tub manufacturer Kirami (release, 28 May 2021) — https://harviagroup.com/harvia-strengthens-its-position-in-the-sauna-and-spa-market-by-acquiring-still-water-hot-tub-manufacturer-kirami-3561 - `harvia-kirami-earnout-2023` (primary): Harvia to pay additional purchase price of the Kirami acquisition in advance (December 2023) — https://harviagroup.com/harvia-to-pay-additional-purchase-price-of-the-kirami-acquisition-in-advance-9281 - `masco-10k-2023-sauna360` (primary): Masco Corporation Form 10-K FY2023 (Sauna360 acquisition) — https://www.sec.gov/Archives/edgar/data/62996/000006299624000006/mas-20231231.htm - `sauna360-watkins-2023` (secondary): Watkins Wellness (Masco) expands into sauna through acquisition of Sauna360 — https://eurospapoolnews.com/actualites_piscines_spas-en/77915-watkins,wellness,announces,pending,acquisition,sauna360,group.htm - `tulikivi-fy2025` (primary): Tulikivi Corporation financial statements release 1–12/2025 (6 March 2026) — https://www.inderes.fi/sv/releases/tulikivi-corporation-financial-statements-release-1-122025-investments-in-strategic-projects-continued-despite-challenging-market-conditions - `tulikivi-guidance-2026-10-28-finnish` (primary competitor release): Tulikivi 2026 guidance reduction — original Finnish release — https://tulikivigroup.com/tulikivi-laskee-ja-tarkentaa-ohjeistustaan-vuodelle-2026/ (published 2026-10-28) - `tulikivi-q3-2026` (primary): Tulikivi Q3/9M2026 interim report — https://ml-eu.globenewswire.com/Resource/Download/82f9268e-9945-4233-960c-a658e854b35d (published 2026-11-06) - `harvia-fenix-support-2026-11-12` (primary): Harvia Fenix 2026 product support and remote-use licences — https://support.harvia.com/hc/en-gb/articles/25175666407196-Harvia-Fenix-2026 (published 2026-11-12) --- # IEH Corporation (IEHC) - Legal Name: IEH Corporation - Exchange: OTCQX (United States) - Sector: High-reliability interconnects - Currency: USD - Reviewed: 2026-11-15 - Model Framework: Qualified connector production and cash - Financial Period: Reported TTM to 30 September 2026; September balance sheet; known October option award; rolling November forecast years (anchored 2026-11-13) - Research Status: 8 of 11 research checks complete. Still to complete: Security and evidence base; Opportunities and funded scenarios; Variant view and catalysts - Static Report URL: https://longview-investing.vercel.app/research/ieh.html - Interactive Model URL: https://longview-investing.vercel.app/#/company/ieh - JSON API: https://longview-investing.vercel.app/api/companies/ieh ## Executive Summary Qualified connectors benefit from missile replenishment and medical orders. Backlog conversion, pricing and factory absorption can lift cash, but the valuation already requires a successful ramp; controls, concentration and working capital limit the margin of safety. ## Key Operational Watch Items - Backlog conversion and repeat orders - Gold repricing and factory operating margin - Receivables, inventory, customer advances and cash - IT controls, export-control cyber exposure and option grants ## Valuation Scenario Assumptions ### Downside Case Order deferrals, persistent gold costs and underused capacity. - Key Drivers: Capital spending / revenue: 3.0% | Operating working capital / revenue: 38.0% | Normalized operating tax rate: 25.0% | Exit / cash operating earnings: 7.50× | Annual share dilution: 0.0% | Retained investment value / carrying cost: 0.0% | Stress loss / contingent exposure: 0.0% - Revenue growth: 17.0%, -11.6%, -7.9%, 0.0%, 2.9% - Operating margin: 7.0%, 5.0%, 3.0%, 5.0%, 6.0% - Fundable net-debt capacity: 0.00, 0.00, 0.00, 0.00, 0.00 ### Base Case Backlog converts with funded capacity and cost absorption; later growth fades. - Key Drivers: Capital spending / revenue: 4.0% | Operating working capital / revenue: 32.0% | Normalized operating tax rate: 25.0% | Exit / cash operating earnings: 8.50× | Annual share dilution: 0.0% | Retained investment value / carrying cost: 0.0% | Stress loss / contingent exposure: 0.0% - Revenue growth: 30.6%, 18.8%, 10.5%, 6.3%, 4.5% - Operating margin: 10.0%, 12.0%, 13.0%, 14.0%, 14.0% - Fundable net-debt capacity: 0.00, 0.00, 0.00, 0.00, 0.00 ### Upside Case Broader design wins, rapid conversion and strong execution, with extra spending. - Key Drivers: Capital spending / revenue: 5.0% | Operating working capital / revenue: 30.0% | Normalized operating tax rate: 25.0% | Exit / cash operating earnings: 8.50× | Annual share dilution: 0.0% | Retained investment value / carrying cost: 0.0% | Stress loss / contingent exposure: 0.0% - Revenue growth: 41.5%, 26.9%, 18.2%, 12.8%, 9.1% - Operating margin: 12.0%, 15.0%, 17.0%, 18.0%, 18.0% - Fundable net-debt capacity: 0.00, 0.00, 0.00, 0.00, 0.00 ## Research Notebook # IEH: a qualified connector factory Research date: 13 November 2026. USD millions unless stated. New independent underwriting from public evidence; no investor return target was supplied. Source IDs below resolve to `sources/registry.json`. ## Value-creation mechanism IEH earns money by designing a reliable electrical contact into a customer's equipment, qualifying the connector, then repeatedly manufacturing it to the required specification. Multiple spring-wire contact paths resist vibration and reduce insertion force. The valuable step is dependable production of an already qualified design: changing a small component can create a large testing and failure risk for the customer. That is an inference about incentives, not a measured switching-cost premium. Qualification does not protect a supplier that misses delivery or prices itself out of the next design. [ieh-fy2026, Item 1; smiths-hypertac] ## Business-model profile Growth is organic: qualified-programme volume, price, capacity and mix. Restated sales fell from $34.72m in FY2021 to $19.14m in FY2023, then recovered to $29.42m in FY2026. There were no business acquisitions in the reviewed cash statements. Management's interest in M&A is an unpriced option, not a forecast deal budget. Reported price and volume are not separately disclosed; no invented organic split is presented. Cash PP&E spending was 2.30% of sales over FY2024–26, but FY2022 capex reached $2.65m, 10.9% of sales, around the Allentown build. Leased premises, production labour, inventories and receivables make the business more demanding than the recent capex percentage suggests. Compare small qualified component factories with similar customer concentration, input costs and programme cycles. Amphenol's scale and acquisition machine are not IEH economics. The broad statistical cohort is a check on growth, not an industry valuation peer. [ieh-fy2022–2026] ## Economic snapshot | Observation | Value | Definition | | --- | ---: | --- | | TTM revenue / EBIT | 36.753 / 0.939 | FY2026 + current H1 − comparative H1 | | H1 FY2027 gross / operating margin | 30.62% / 8.26% | Gross profit before separately stated D&A; EBIT after D&A | | Q2 gross / operating margin | 28.29% / 9.78% | Not the 12.74% net margin inflated by tax benefit | | Operating WC | 11.786 | AR + net stock + prepaids − AP − advances − other operating accruals | | Net PP&E / ROU assets | 3.045 / 1.390 | September balances | | Equipment / lease debt | 0.382 / 1.624 | Rent remains expensed; do not deduct lease debt again | | Unrestricted / restricted cash | 9.004 / 1.000 | Only unrestricted cash above the $2.5m reserve enters excess cash | | H1 annualized AR / inventory days | 56 / 103 | Closing balances divided by twice H1 sales / COGS, ×365 | At a judged 25% normal tax, trailing NOPAT is $0.704m. Lease-expensed tangible operating capital is WC + PP&E = $14.831m; adding required cash gives $17.331m. Current returns are about 4.75% before that reserve, or 4.06% including it. The latter decomposes into 1.92% NOPAT margin × 2.12 sales/capital. These are point-in-time proxies, not matched historical cohort returns. A lease-capitalized diagnostic adds the $1.390m ROU asset and estimated annualised lease-interest component of $0.086m (=2×(H1 straight-line lease expense $0.294524m − ROU amortisation $0.251437m)) before tax: roughly 4.74% on $16.221m, before required cash. Changing lease convention does not create cash. [ieh-q2fy27, statements and Notes 2,5; calculations.json] ## Economic segmentation Defence was 72.9% of H1 sales ($15.11m), commercial aerospace 19.5% ($4.04m), space 1.5% ($0.31m), and other 6.1% ($1.26m). Medical is inside other; it is not an additional segment to add on top. These shares reveal demand exposure, not profit shares. Programme, product and customer margin tables are unavailable, so no claim that defence supplies a measured percentage of profit is made. The chain runs from metals, plating, tooling and labour through IEH and representatives to equipment makers and procurement agencies. Primes control programme selection and budgets; IEH supplies a small reliability-critical item. Manufacturing competence and requalification friction may leave IEH a useful return, but there is no verified measurement of industry profit-pool shares or content per missile. Model the whole factory, then test exposure by programme. [ieh-q2fy27, Notes 2,10; ieh-fy2026, Item 1] ## Unit economics The relevant unit is a recurring qualified programme batch through the shared factory. Contract prices, connector counts, scrap and labour hours are not disclosed. A transparent $1m-of-sales allocation illustrates the economics without pretending to be an observed connector P&L: | Per $1m sales | H1 observed group allocation | Mature base allocation estimate | | --- | ---: | ---: | | Gross contribution before D&A | 306,150 | 330,000 | | SG&A, including awards/commissions | (203,503) | (167,929) | | D&A | (20,070) | (22,071) | | EBIT | 82,577 | 140,000 | | Normalized tax at 25% | (20,644) | (35,000) | | NOPAT before growth investment | 61,933 | 105,000 | Manufacturing wages and materials sit in COGS; they are not added again. All rent and recurring equity compensation remain charged. Mature 14% EBIT assumes repricing/throughput and lower overhead per dollar, not a reported product margin. At 32% WC, every $1m of additional annual sales needs about $320,000 of operating capital, plus machinery. Existing spare capacity can make the first increment unusually profitable; that does not establish the next factory's return. The FY2022 expansion followed by FY2023–24 losses is the failure case. Undisclosed project cohorts prevent a measured project IRR. [ieh-fy2022–2026; ieh-q2fy27] ## Independent checks First, H1 sales increased $7.336m while gross contribution increased $3.597m and EBIT $2.571m: a 35.05% incremental EBIT margin, largely recovery and absorption. That reconciles the unit story with the group accounts but cannot be applied indefinitely. Second, BAE's APKWS expansion and US Army PAC-3 contract actions independently corroborate programme demand; they do not disclose IEH content or allocation. Third, Smiths sells competing hyperboloid products, refuting a technology monopoly. An adverse allocation that retains H1's 20.35% SG&A ratio rather than assuming leverage leaves only about 10.4% EBIT at a 33% gross margin. That is a material valuation sensitivity. No independent factory visit, channel sample or customer interview was performed. [army-pac3-apr26; bae-apkws-oct26; smiths-hypertac] # Demand, competition and counterparties ## Demand drivers and cycle Missiles and guidance kits are consumed, replenished and redesigned; aircraft programmes need initial fitment and maintenance, while medical devices need reliable repeated connection. The end user is the military, aircraft operator or medical provider; procurement agencies and OEMs pay, and engineering/quality teams choose qualified parts. Reliability under shock is the job. Rival contacts, alternative interconnect architectures, redesign, inventory drawdown and delayed procurement can all substitute for a new IEH purchase. The cycle is visible in restated revenue: $34.7m FY2021, $24.3m FY2022, $19.1m FY2023, $21.5m FY2024, $28.8m FY2025 and $29.4m FY2026. Sales recovery is not proof of a secular CAGR. September backlog of $53.751m is expected to ship over 6–24 months and remains subject to cancellation or postponement. The November release says backlog now exceeds $56m without matching the September measurement date; the forecast uses the dated filing figure, never both. Named $5.1m and $7.9m Patriot, $6.5m APKWS and $3.6m medical orders are evidence within the book, not additions to it. [ieh-fy2022–2026; ieh-q2fy27, MD&A; ieh-exhibit-2026-03-26/2026-05-18/2026-09-02/2026-09-25/2026-11-13] Funding matters. The Army's April $4.7bn action and Lockheed's July undefinitized multiyear award support actual procurement, beyond a defence-spending aspiration. BAE's October framework targets up to 35,000 APKWS kits annually but says a contract is still being negotiated. Its November update reports additional munitions funding. Those are customer-side corroboration, not guaranteed IEH purchase commitments, appropriations through every future year or verified exposure to a particular PAC-3 variant. Initial seven-year frameworks expire around 2032–33; years 6–10 fade to mature growth, and the downside models replenishment loss as a step down. No customer budget × guessed connector content is used to create a false demand ceiling. [army-pac3-apr26; lmt-pac3-jul26; bae-apkws-oct26; bae-nov26] ## Competition Smiths Hypertac provides the clearest technology counterexample: it markets hyperboloid contacts for the same harsh-environment job. IEH's claim to be the only independent US producer of hyperboloid PCB connectors is narrow, not a claim to own the technology. The annual report describes competitive bids and product-performance/production competition. Verified market shares and like-for-like price premiums are unavailable. [smiths-hypertac; ieh-fy2026, Item 1] Rivalry is meaningful at new design wins; qualification may reduce substitution within an existing programme. Entry into precision manufacturing is possible, but qualification, process consistency and delivery history take time. Buyer power is strong: one customer supplied 29.9% of H1 sales, and two supplied 37.6% in Q2. Supplier pricing matters through gold, imported parts and labour; three vendors supplied 40.2% of H1 purchases. Low-end commodity connectors can address less demanding uses, but a cheaper contact must satisfy the customer's failure constraints to displace IEH. None of these facts proves a rising or falling industry-wide force over time. The plausible power is switching cost within a qualified design. Its benefit is repeat orders; the barrier is requalification and failure risk. The size and duration of a price advantage have not been measured. Process competence is a strength, not independently established process power. No scale, network or brand premium is assumed. These limits are why higher sales can support an upside operating case without automatically earning a premium terminal multiple. ## Counterparty incentives Primes need an approved small component delivered on schedule to protect much larger programme cash flows. Maintaining a competent supplier can be cheaper than requalification; they also have an incentive to dual-source, squeeze prices and redesign at the next platform generation. IEH's leverage weakens if it misses shipments. The alignment supports some repeat demand, while concentration puts order timing and price recovery in the downside. Anonymous customer percentages cannot establish that any one prime is the disclosed largest customer. Metal suppliers want price coverage, not fixed-price exposure to a defence backlog. IEH carries stock and has increased vertical integration and purchasing to reduce disruption; that consumes cash and cannot eliminate gold risk. Labour wants wages and job security; about half the manufacturing/testing workforce is union-covered, with the agreement expiring 31 March 2027. The family wants business continuity and retains meaningful ownership, but outside holders still bear compensation and control risk. Each relationship affects margin, WC, capex or allocation; none merits an unsupported numerical goodwill credit. [ieh-fy2026, Item 1; ieh-q2fy27, Notes 3,10; ieh-proxy26] # Management and allocation ## Ownership and incentives | Person / holder | Disclosed position | Limits | | --- | --- | --- | | Dave Offerman, CEO/chair | 652,127 beneficial shares, including 250,000 vested options; 402,127 actual shares | March Form 4 confirms 402,127 shares; proxy's 24% includes options and a different denominator | | Gail Offerman | 499,606 shares | Company-knowledge figure; not independently current holdings verification | | Zeff Capital | 232,862 shares | Proxy relies on January 2022 Schedule 13G | | Intelligent Fanatics | 104,147 shares | Proxy cites August 2025 Schedule 13G; below 5% | | Subrata Purkayastha, CFO | 35,000 options in FY-end table; 35,000 additional October options at $29.60 | New agreement adds an immediately vested award; not an open-market purchase | | Six non-executive directors | FY-end table mainly 10,000–15,000 options each; Glenn 18,379 beneficial shares | Forms 4 show later sales/exercises; proxy is not a fresh ownership census | CEO and Gail's disclosed actual shares sum to 901,733, about 36.25% of September issued shares. Do not add the reported 24.0% and 20.2% beneficial percentages: their denominators differ. No separate super-vote class is disclosed. The CEO's actual stake is worth $12.57m at $31.25; 12% compounding for five years would make it $22.15m before tax, ignoring options. This is an incentive illustration, not a forecast return. [ieh-proxy26, pp.28–30; ieh-edgar, March Form 4] CEO FY2026 salary/bonus were $516,319/$50,000, total $566,319 versus FY2025 $683,245 including option value. CFO totals were $292,493/$321,700. Director cash fees were $17,500–$20,000 plus $22,100 option value each. The October CFO agreement raises base salary to $293,950, permits a bonus up to base salary and supplies 35,000 immediately vested options; it includes 24 months' severance in specified circumstances. The cash-bonus section says payments require profitability yet records $398,975 bonus expense in loss-making FY2026. Without a disclosed reconciliation, do not call incentives purely profit-contingent or assume a clawback. Annual payout targets and outcomes are insufficiently disclosed for a full metric-by-metric assessment. [ieh-proxy26, pp.22–27; ieh-event-2026-10-27] ## Track record and references The record includes the Allentown investment, restored reporting, a deep operating downturn, later order recovery and the 2026 cost shock. Restated FY2021 EBIT was $2.585m, not a pristine pre-restatement earnings peak. Operating losses in FY2022–24 and FY2026 prevent a claim of uninterrupted compounding. The original filing index and Forms 4 were inspected: the CEO sold 5,000 shares on 6/9 March at $16.50/$17.275, while Rosenfeld sold 4,240 on 13/14 August at $33/$32.50. Sales are facts, not proof that management knows intrinsic value. No verified tenure-long total-return series or independent personal references were obtained. Offerman's customer relationships and production coordination could affect conversion and gross margin. Evidence of a system is a long-lived technical workforce and two operating sites; there is no disclosed succession study showing that those relationships transfer unchanged. The base requires ordinary competent scheduling and pricing. The upside requires better wins and execution; the downside allows delayed conversion and weak cost recovery under an ordinary successor. No separate arbitrary key-person penalty is added. [ieh-fy2022–2026; ieh-proxy26; ieh-edgar, Forms 4] ## Capital allocation principles No acquisitions, dividends or repurchases occurred in the five-year cash record. Capital went into production and operating liquidity. The Allentown build and vertical-integration plan connect spending to a strategy, but losses following the build prevent treating new capacity as an already-proven high-return cohort. Zero-based allocation and willingness to exit poor growth are unproven; no disclosed divestment supplies that evidence. Equity awards have a cost even when noncash. The model charges future compensation through EBIT and recognises outstanding options separately. No demonstrated buyback discipline or acquisition price record justifies a capital-allocator premium. A future acquisition could diversify demand, but the price, target economics and integration capacity must be underwritten when known. Retained cash is valued at zero interest until exit, reflecting a deliberately conservative alternative to assuming immediate dividends; destructive acquisitions remain a tail. The evidence supports a continuing owner-managed manufacturer, not an established per-share capital-allocation system. [ieh-fy2022–2026; ieh-exhibit-2026-04-22] ## Integrity and governance The SEC dismissed the delinquency proceeding on 14 January 2026 after reporting compliance was restored. That resolves that proceeding; it is not a current controls clean bill. September's 10-Q still declares ineffective IT general controls and review controls. August's mailbox intrusion made potentially export-controlled information accessible; the company reported no evidence of exfiltration and expected no material operating effect, while investigation continued. Those issuer statements are neither proven harm nor independent clearance. [ieh-exhibit-2026-01-15; ieh-event-2026-08-06; ieh-q2fy27, Item 4] Family influence, immediately vested awards, severance and the unexplained bonus wording create agency concerns over resources and time horizon. Concentrated ownership may favour continuity over outside holders' preferred payout or risk. The downside and low confidence reflect these mechanisms; no fabricated fine is charged. The next controls disclosure, option register, cash use and AGM voting result are the checks. The 12 November meeting is past; no result was visible in the filing snapshot, so no vote outcome is claimed. # Cash allocation and productive capacity ## Sources and uses of cash | Fiscal year | CFO | Cash capex | D&A | SBC in CFO | Financing cash | Closing cash | | --- | ---: | ---: | ---: | ---: | ---: | ---: | | 2022 | 1.414 | 2.647 | 0.837 | 0.383 | 0.000 | 12.675 | | 2023 | −3.785 | 0.546 | 1.035 | 0.000 | 0.000 | 8.345 | | 2024 | −1.914 | 0.347 | 0.872 | 0.343 | 0.057 | 6.140 | | 2025 | 4.884 | 0.532 | 0.745 | 0.267 | 0.049 | 10.540 | | 2026 | −0.631 | 0.953 | 0.772 | 0.265 | 0.692 | 9.648 | Opening FY2022 cash $13.907542m + five-year CFO −$0.032334m − capex $5.024904m + financing $0.797394m = FY2026 cash $9.647698m. Each annual statement reconciles independently. Financing includes option exercise proceeds and the FY2026 equipment loan; it is not owner earning power. No dividends, buybacks or business purchases appear in the reviewed five-year cash statements. FY2022 income includes $2.104m PPP forgiveness that the CFO statement reverses; do not call it recurring cash. [ieh-fy2022–2026, cash-flow statements; calculations.json] H1 FY2027 CFO $0.511634m − capex $0.151716m − debt repayment $0.033966m + option proceeds $0.030050m = $0.356002m increase in total cash. Opening $9.647698m becomes $10.003700m, including $1m newly restricted collateral. Unrestricted cash actually falls to $9.003700m. Restricting cash is not an operating outflow, and the collateralised revolver is not extra unencumbered liquidity. [ieh-q2fy27, pp.1–4, Notes 2,6,7] ## Investment rates per sales change Across the five fiscal years FY2022–26, sales fell $5.297595m from the FY2021 opening base. Net PP&E cash investment was $0.764459m (capex $5.024904m less D&A $4.260445m), giving a signed fixed-capital/sales-change ratio of −14.4%. Operating WC released $3.963461m, giving +74.8% on the same negative sales denominator. These signed diagnostics describe contraction, not attractive growth productivity. Acquisition investment was zero. Incremental intangible investment cannot be separated from ordinary SG&A in the disclosures; no blanket SG&A add-back is permitted. Named arithmetic is in `valuation/calculations.json`. The current ramp tells a different story: net operating WC rose $2.258693m from March to September while sales accelerated. Receivables and raw/WIP inventory consume cash before shipment/collection; $0.912505m customer advances partly finance it. Forecast capital must fund those balances even if the historical contraction released them. A stable 32% WC/sales ratio is a current underwriting estimate, not the historical signed ratio recycled as a growth rule. ## Reinvestment and returns The FY2022 Allentown spending preceded FY2023–24 operating losses. That can reflect spare capacity, pandemic disruption and weak orders; it does not prove either a worthless factory or an attractive new project. Current recovery uses some existing paid capacity, while manufacturing labour and SG&A stay charged. No separate quantified setup loss is disclosed well enough to add back. All future capex is paid; negative early free cash flow is allowed when opening excess cash funds it. Two lagged windows were examined: FY2021→FY2026 NOPAT against FY2021→FY2025 capital, and FY2023→FY2026 NOPAT against FY2022→FY2025 capital. Both capital changes are negative. A positive repeatable ROIIC cannot be inferred; the trough recovery is chiefly margin/utilisation on a smaller capital base. Five-year normalized aggregate earnings are negative, so a positive earnings-weighted historical reinvestment rate is undefined. The shared engine's forecast return and reinvestment diagnostics are shown separately in `10-capital-efficiency.md`, with recovery distinguished from new capital productivity. [ieh-fy2022–2026] ## Acquisition record No business acquisition cash is disclosed in FY2021–26. All reviewed investing cash is PP&E. Management's presentation describes M&A as a possible diversification lever, but no target, price, purchased revenue, funding or post-deal cash yield is provided. There is therefore no observed acquisition yield, deployment average or acquisition contribution in any scenario. The model does not treat the absence of a signed deal as proof that diversification has zero economic possibility; it leaves the option unpriced and requires fresh underwriting before any transaction enters owner cash. [ieh-exhibit-2026-04-22, strategic diversification; ieh-fy2022–2026] # IEH: real orders, demanding expectations At $31.25, the business can recover and the investment can still disappoint. Qualified designs, growing orders and available cash support a larger factory. The price already requires more cash or a richer valuation than the evidence-weighted model supplies. This is a watchlist case, not a favourable margin-of-safety conclusion. Price: delayed quote at 15:50:08 Eastern,13 November 2026. Estimates are USD before taxes, dealing costs and spread. | Horizon | Downside | Base | Upside | IRR of weighted cash flows | | --- | ---: | ---: | ---: | ---: | | Five years | −23.60% | −0.89% | 9.66% | −2.67% | | Ten years | −10.21% | 3.57% | 9.41% | 2.67% | Exits are13 November 2031/2036. Base terminal prices are $29.89/$44.38, including retained excess cash once. No dividends are forecast. At the judged12% required return, base-case entry prices are about $16.95 for five years and $14.28 for ten. These are conditional scenario prices, not precise intrinsic values or a recommendation to transact at those levels. The difference reflects longer cash retention and capital economics, not two independently proven values. [ieh-quote; valuation/model.json; dated engine audit] ## Pre-mortem - Understandable component economics; the harder hurdle is programme-level prices, capacity and orders, which are undisclosed. - Demand combines replenishment and design cycles; defence tailwinds are not proof of uninterrupted secular growth. - A value trap results if backlog fails to replenish, input repricing lags and newly hired capacity remains underused. - The factory is productive, not a liquidation cigar butt; that does not establish an appreciating asset at any entry price. - Agency costs include family influence, awards, severance and unproven discipline on retained cash and M&A. - Ordinary competent execution underpins base; stronger sales, margin and capacity execution belong in upside. - This is a recovery from operating losses; a record order book does not erase turnaround risk. - The historical Allentown build followed by losses challenges any claim that expansion always earns attractive returns. - Smiths supplies competing hyperboloid products; IEH does not own an unassailable technology monopoly. - Multiple applications reduce one-product risk, but one customer was29.9% of H1 sales and programme concentration can dominate diversification. These are testable mechanisms, not a checklist that proves safety. Loss of qualification, an extended unprofitable ramp or serious cyber/regulatory consequences can produce losses beyond the representative downside. [ieh-fy2022–2026; ieh-q2fy27; smiths-hypertac; ieh-event-2026-08-06] ## Exit multiple Use7.5x/8.5x/8.5x unlevered terminal-year NOPAT, with bank net debt and excess cash added separately. Rent and future compensation remain charged. The12% rate is a judged owner hurdle, not a measured market cost of equity. No reliable same-definition historical or current peer EV/NOPAT series was established, so no invented peer median or implied historical range justifies the choice. Historical losses make P/E comparisons particularly misleading. Smiths is a technology competitor; Amphenol is too different in scale and acquisition intensity to supply an automatic multiple. The independent debate rejected provisional8x/14x/16x. Extend the shared forecast beyond the selected exit, pay actual capex and WC, discount FCF at 12%, and end after a finite period with no growth and NOPAT/12% (sustaining capex=D&A proxy). Five/ten/twenty further years support base 8.38/8.41/8.44x; upside about 8.56x at the five-year exit and8.33x at the ten-year exit. Downside ten-year continuation supports7.34x, versus7.72x with five more years. Rounded7.5/8.5/8.5x makes that uncertainty explicit. No-growth capitalization is8.33x. Base/upside 8.5x contain just0.17x, about 2% of enterprise value, as a small rounding/persistence premium, supported by the explicit5–20-year continuation rather than a claimed permanent moat. Downside7.5x is below no-growth because continued low-return spending consumes cash. A14x base would put40.5% of value above the8.33x no-growth component; the forecast cannot earn that premium. Even lowering mature capex to2.7–3% supports only about 9.9–9.6x at 3% growth. No unsupported future ROIIC, lower discount rate or re-rating repairs the gap. [research/12-debate.md; updates/2026-11-13-valuation-audit.json] ## Bear case Existing orders lift initial rolling sales to $43m, then delivery deferrals and weaker replenishment pull sales to $35m. Gold, labour and underutilisation reduce EBIT to3% before cost action restores6%. Capex3% and WC38% keep the factory functioning and finance slower conversion. At year five, NOPAT is $1.62m, excess net cash about $10.07m and the7.5x operating value gives an $8.13 economic-share exit. This is a seller-with-time outcome, not a forced liquidation; severe programme loss or indefinite losses could be worse. Early cash deficits are funded, so the downside does not equate negative FCF with bankruptcy. ## Why the opportunity exists The security is a roughly $77.73m basic market-cap OTC manufacturer at the current quote and September issued count. At2.733526m economic shares, the option-adjusted purchase claim is about $85.42m before the cash/debt bridge. A small float and thin trading can explain limited participation, but neglect does not establish cheapness. The IR site explicitly labels Atrium coverage sponsored; no claim of zero research coverage is made. Current short interest, independent coverage count, retail split, index membership and a reliable average float-turnover series were not verified. March's reinstated OTCQX quotation and January's dismissed delinquency proceeding remove historical barriers; they are past events, not future catalysts. November results and order releases already made the recovery visible. There is no identified forced seller or hard catalyst that establishes an overlooked bargain. Earnings conversion and controls remediation could change confidence, but cash and price must still support the return. [ieh-ir; ieh-proxy26; ieh-event-2026-01-15/2026-02-20; ieh-quote] ## Variant view There is no evidenced favourable mispricing at $31.25. A mechanical reverse valuation, holding the base sales/capital/exit assumptions fixed and shifting the entire EBIT-margin path, requires about 24.21% terminal EBIT to earn12% over five years, versus the underwritten14%. That is not an observed consensus belief: price could instead imply lower required returns, faster sales, better investment productivity or a richer exit. It states precisely what the price requires under this model. The implied margin exceeds what current gross contribution less recurring overhead supports. The research disagrees with an uncomplicated “backlog growth means cheap earnings growth” argument. Orders are real, and operating upside is credible, but capital conversion, recurring awards and valuation duration absorb much of the apparent benefit. The independent advocates agree on those limits after cross-examination. No hard catalyst is needed for a watchlist: a lower quote, better demonstrated cash economics or a larger funded opportunity would warrant re-underwriting. ## What would prove us wrong Higher shipments accompanied by a sustained mid-teens EBIT margin, funded capex and collections would challenge the operating caution. A growing order book without shipment/collection would not. Q3 FY2027 has not been reported: revenue, after-compensation EBIT, AR, stock, advances and controls are frozen for future scoring. The issuer has not confirmed its release date; February 2027 is an estimated window. Programme contract conversion and procurement funding must be checked at the primes and agencies, not inferred from IEH commentary alone. Conversely, a cancelled programme, gross margin that fails to recover, receivables growing faster than sales, or repeated ITGC weakness would lower the paths or confidence. New acquisitions need a separately priced cash/funding bridge. Source/index dates and new awards matter: a current quote alone is not a refreshed business forecast. The prospective register specifies periods and scoring; future outcomes must be appended rather than replacing today's predictions. # Permanent-loss paths and disconfirming evidence | Risk | Cash mechanism | Observation / action | | --- | --- | --- | | Programme cancellation, redesign or lost allocation | Repeat orders fail while staff/tools remain; underused factory destroys margin | Separate shipments, net new orders and cancellations; do not call a long backlog recurring revenue | | Gold, imports, wages and union renewal | Fixed-price order book reprices more slowly than inputs; price increases lose bids | Gross margin, procurement terms and contract renewal; union agreement expires31 March 2027 | | Capacity execution | More staff and tools fail to raise yield/throughput; quality defects cost the programme | Staffing, capex, on-time shipment and scrap evidence; no verified plant ceiling/project return exists | | WC / funding | AR and WIP rise before cash; advances reverse | AR concentrations, inventory ageing/provision, customer deposits; stress38% WC and larger reserve. Annual net-cash solvency is not intra-year liquidity proof | | Accounting/control quality | Inventory and reported profit need correction; confidence and customer access suffer | September ITGC weakness remains unresolved. Use restated FY2020–21, not original profit peaks | | Cyber/export-control exposure | Accessible technical/customer information leads to notification, restrictions, penalties or lost business | August 8-K says no evidence of exfiltration, not no breach. No quantified liability/independent clearance available | | Minority governance / allocation | Awards, severance, retained cash or poor diversification absorb per-share value | October CFO grant included; monitor Form4s, new agreements, AGM vote and acquisitions. No presumed family ownership all-clear | | Valuation and liquidity | Even a good factory recovery produces inadequate shareholder return; OTC exits may be costly | Timestamped quote, full options sensitivity, terminal cash-flow persistence; no reliable current peer-multiple series | No scenario is a loss bound. Three coarse weights cannot identify the probability of a catastrophic legal/customer event. The model includes ordinary delays, input pressure and underutilisation; a permanently unfinanceable loss phase would need an explicit recovery or unavailable valuation, not a mechanical zero based solely on a leverage guardrail. Positive future cash also depends on sensible retention: a dollar of hoarded or badly acquired cash may be worth less than a dollar at exit. Evidence gaps are substantive: customer names behind concentration, price per connector, unit content by missile, tooling/qualified labour capacity, product margins, order cancellation rights, project returns, post-exit maintenance requirements and current ultimate holdings of old13G holders are not established. No fieldwork or member-only investor research was performed. The next quarter's issuer date is unconfirmed; results are expected in February 2027. [ieh-fy2026; ieh-q2fy27; ieh-event-2026-08-06/2026-10-27; ieh-proxy26; smiths-hypertac] # Definitions and assumption register Vintage: 13 November 2026. No investor premise is designated authoritative. Reported facts are in `financials/facts.json`; named arithmetic and evidence bindings are in `valuation/calculations.json`; all future business paths are independent judgments. | Input | Observation / calculation | Judgment, limitation and revision trigger | | --- | --- | --- | | Starting revenue | 29.417600 + 20.721530 − 13.385747 = 36.753383 | TTM through September; no peak-quarter annualisation plus a second ramp | | Starting EBIT | −1.631879 + 1.711298 − (−0.859686) = 0.939105 | Reported costs retained; no tax or SBC add-back | | Normal tax | H1 net benefit 0.197905 includes deferred release 0.485068; current charge 0.287163 | Use 25% normal cash-equivalent operating tax; stress 21–30%; revise for durable legal/tax evidence | | D&A | Matching TTM 0.811177 / sales = 2.2071% | Wear proxy, not proven maintenance. Charge all capex independently | | Operating WC | AR6.405635 + stock8.175984 + prepaid0.661096 − AP1.610004 − advances0.930976 − other accruals0.915677 =11.786058 | Base32%, down38%, up30% of sales each year. No taxes, cash or lease/debt balances; revise on collections and advance recurrence | | Cash and reserve | Unrestricted9.003700; restricted1.000000 | Reserve2.5m, roughly one month's production/overhead, leaves6.503700m free. Neither reserve nor restricted cash adds to exit excess cash; test reserve±1m | | Bank debt / interest | Equipment0.381958 at 5.51%; revolver undrawn | Fixed opening gross interest is conservatively retained; bank net debt counted once. No assumed revolver renewal after July 2027 | | Economic shares | Issued2.487454 + September options0.244224 + October options0.001848 =2.733526m | Treasury-equivalent claims at frozen $31.25; not lifetime full-exercise dilution. Full-exercise/proceeds sensitivity disclosed; no second credit of proceeds | | Future compensation | SBC is already inside reported EBIT | Forecast EBIT continues to pay recurring awards as cash replacement, zero extra net dilution. Existing awards are a separate opening claim. Extra1% net issuance is a sensitivity, not free compensation | | Revenue | Per-year absolute budgets48/57/63/67/70 m base, then3% nominal | Existing backlog counted once; named programme orders not additive. No disclosed units/ASP. Revision: shipments, new orders, cancellations and dated customer procurement | | Operating margin | H1 8.26%, Q2 9.78%; H1 incremental35.05% | Base10/12/13/14/14%, then14%; up to18% only with better repricing/absorption. Gross margin33% less SG&A16.79% and D&A2.21% explains mature base. Not a measured historical moat | | Capex | FY2024–26 cash capex2.30% of sales; FY2022 10.9% | Pay3/4/5% of revenue down/base/up for every modeled year. Extra capacity/vertical integration and sustaining wear included. Missing project budgets remain uncertain; recent light capex is no reason for a free ramp | | Exit | Finite shared-engine FCF continuation at 12% plus no-growth NOPAT/12% endpoint | Judged7.5/8.5/8.5x NOPAT. Provisional8/14/16x rejected. No premium peer multiple or hidden lower required return | | Payout / acquisitions | None in reviewed five-year cash statements | Retain surplus at zero return, exit realization; zero deals. Revise on announced allocation with price/funding | The half-year cash gap is itemized, not waved away: net income2.017880 + D&A0.415878 + SBC0.521500 + obsolescence0.150000 − deferred tax0.485068 + ROU amortisation0.251437 − AR1.686412 − inventory1.516262 + tax receivable0.319734 − prepaids0.220191 + AP0.183937 + advances0.912505 − lease liabilities0.271034 − other accruals0.082270 = CFO0.511634. After capex0.151716, reported FCF is0.359918; replacing noncash SBC costs0.521500 leaves−0.161582. Tax payment timing remains separate. Net stock investment is modeled on inventory after provision: do not deduct the historical write-down again as new cash. Operating NOPAT excludes interest income because opening cash is valued separately. Cash rent stays in costs, so ROU amortisation is not added as a free operating cash adjustment and lease debt is not deducted again at exit. No identified acquisition claim or quantified cyber liability exists to insert; unquantified tails remain in risks. Every operating forecast pays full capex and net WC and must remain net-cash funded. These definitions generate all dashboard and report figures through the existing engine. The September corporate tax receivable of $0.148260m is net of a $0.177618m uncertain-tax liability including interest/penalties (Note8). Neither the receivable nor its offset is valued in operating WC or excess cash; do not add the gross tax asset while omitting that liability. A separate stress pays the gross liability without a refund. Straight-line operating rent is a cash-rent proxy: H1 lease expense0.294524m exceeds ROU amortisation0.251437m by an estimated interest component0.043087m, and the liability cash reduction0.271034m leaves a0.019597m cash-versus-expense timing difference. No lease payment is charged twice. Current full-known-option exercise sensitivity, with strike proceeds credited once: five-year base−0.81% / upside8.92%, ten-year base3.29% / upside8.82%. The five-year base exit is slightly below entry, so full exercise can marginally improve the fixed-treasury approximation there; at higher exits it reduces return. This convention is not imposed on the downside where some options are underwater. # Opportunities, conversion and funding ## Opportunity ledger | Opportunity | Present evidence and why IEH can win | Reachable revenue budget / timing | Capital and limits | Evidence against | | --- | --- | --- | --- | --- | | Existing defence orders | Qualified designs; September total backlog53.751m; named Patriot/APKWS awards | Base first rolling year 48m total sales: about 30m existing book +18m repeat/new shipments. First two years 105m needs about 51m replenishment beyond opening book | Labour, tools, metals, AR and WIP before cash; 6–24-month delivery windows, cancellations possible | No firm programme-by-programme delivery schedule or capacity ceiling disclosed | | Repeat defence procurement | Army funding actions, prime multiyear production and BAE actual munitions funding | Base year 5 defence about 50m versus current H1 annualised30.2m; residual aerospace14m and other6m give total70m | Repeat allocation and capacity needed; do not multiply prime output growth by guessed IEH content | Funding/appropriation risk, supplier competition, procurement windows ending 2032–33 | | Medical repeat designs | $3.6m mobile-defibrillator orders | Existing order is inside backlog, not3.6m extra revenue. Base other6m by year 5 includes medical; upside needs repeat designs | Qualification, commissions, manufacturing stock and customer concentration | End customer unnamed; order duration and recurring cadence undisclosed | | Aerospace and space | Existing aerospace19.5%, space1.5% H1 mix | Base aerospace14m by year 5 against H1 annualised8.1m; space included in other, no separate large rebound | Same production resources, not independent free capacity | H1 space fell60%; delayed orders may recur or disappear | | Price/vertical integration | Gold and imported-input inflation; strategic purchasing, tools and price increases | Included in total sales/margin budgets, not a second growth line | Pay tools and inventory; no assumed margin recovery from falling gold alone | H1 gross margin exceeded Q2; pass-through lag and mix may persist | | M&A | Management interest only | No priced contribution in any path | No observed purchase yield or seller pipeline | A cheap unannounced acquisition cannot rescue valuation | The detailed quantities are estimates designed to be falsified, not contracts. Base $70m is roughly1.9× TTM and2.4× FY2026; upside $96m is2.6× TTM. The176-person March workforce and shared-site footprint mean either substantial hiring, productivity or both. At FY2026 sales/head, base $70m would require about 419 staff; at H1 annualised productivity per March employee, about 297. These are staffing bounds with a stale headcount denominator, not a forecast or verified site capacity. A budget that cannot be staffed/tooled must be cut. [ieh-fy2026, Item1; ieh-q2fy27; ieh-exhibit-2026-09-25/2026-11-13; army-pac3-apr26; bae-apkws-oct26; lmt-pac3-jul26] ## Outside view FY2021→26 sales CAGR was about−3.3%, while FY2023→26's15.4% CAGR starts in a trough. Neither is the forecast. Existing aggregate base-rate table matches under $100m sales,10–20% trailing growth, incidental/no deals and under 5% cash capex. Base three/five/ten-year revenue CAGRs are19.68/13.75/8.24%, around the80th/74th/63rd percentiles; upside 28.51/21.17/12.15%, around90th/86th/77th. Five-year cohort:4,782 overlapping outcomes,2,618 firms. No exception is needed, but this broad surviving-company cohort excludes failures without future observations and does not validate IEH's margin or programme allocation. No raw licensed data were accessed or transmitted for this study; only the repository's existing aggregate table was read locally. ## Projections, year by year Rows are twelve-month November anniversaries from13 November 2026; the September TTM baseline contains six weeks of timing lag. No unknown October/November cash or sales are invented. For fiscal checks, do not compare the first rolling $48m directly with FY2027: six months of FY2027 revenue20.722m is already reported, and Q3 fiscal predictions are frozen separately. | Year ending November | Down sales / EBIT% | Base sales / EBIT% | Up sales / EBIT% | | --- | ---: | ---: | ---: | | 2027 | 43 /7% | 48 /10% | 52 /12% | | 2028 | 38 /5% | 57 /12% | 66 /15% | | 2029 | 35 /3% | 63 /13% | 78 /17% | | 2030 | 35 /5% | 67 /14% | 88 /18% | | 2031 | 36 /6% | 70 /14% | 96 /18% | | 2032 | 36.72 /6% | 72.10 /14% | 100.80 /18% | | 2033 | 37.45 /6% | 74.26 /14% | 104.83 /18% | | 2034 | 38.20 /6% | 76.49 /14% | 109.03 /18% | | 2035 | 38.97 /6% | 78.79 /14% | 112.30 /18% | | 2036 | 39.75 /6% | 81.15 /14% | 115.66 /18% | Each year's growth and EBIT margin has its own reason and source IDs in `inputs.yearly`. All ten years explicitly retain the scalar policy: capex3/4/5% of sales, WC38/32/30%, normalized tax25%, D&A2.2071%, no payouts/deals and no additional dilution beyond charged compensation. Interest remains5.51% on opening equipment debt. The shared engine projects revenue → EBIT → tax → NOPAT + D&A − full capex − net WC − interest → cash. No cash-only CAGR or separate browser formula is used. April's old FY2027 sales proposition was $36m; H1's $20.72m and the newer order book render it a dated bound rather than the latest forecast. It would imply only $15.28m in the remaining half. There is no fresh numerical FY2028 margin/sales target in the November release to reproduce. The model therefore does not claim to run at a current management target; the current CEO's qualitative strong outlook is a hypothesis tested against the order and cash budgets. [ieh-exhibit-2026-04-22/2026-11-13] ## Scenario mechanisms The downside still ships existing orders initially, then falls to $35m as replenishment fades and repricing disappoints. Cost action restores6% EBIT after the trough; committed factory costs are not wished away. It is not a worst case: complete programme loss, a prolonged loss-making shutdown or cyber sanctions can be worse. WC38% and3% capex pay for slower stock/collection and basic productive capacity. The base converts the book, wins repeat work and gradually absorbs overhead, then settles at 3% nominal growth. It must earn14% EBIT at $70m, above its recent restated record; that is a meaningful execution requirement, not the conservative case merely because it is labelled base. Four per cent capex is paid throughout and WC remains32%. Opening free cash $6.504m funds the initial ramp; no new net borrowing or expiring-line renewal is required in annual rows. Intra-year liquidity is less certain. The upside is a larger, productive factory with stronger design wins, faster throughput and18% EBIT. Five per cent capex and30% WC mean more spending despite better conversion. Long-run growth slows rather than extending the initial order surge. Its exit multiple stays equal to the base: more growth has already consumed capital and earned cash in the explicit years. Competitor evidence and unmeasured persistence do not justify a second valuation reward. Exact funding balances and period-weighted returns appear in the dated engine output. # Initial underwriting evaluation 13 November 2026. This is an open-book study from current public filings, with independent advocates and a separate neutral judge. It is not a blind historical reconstruction, an author-result calibration or proof of investment skill. No forecast outcome has yet been scored. The strongest stage was reported cash reconciliation: annual opening/closing cash and the latest half-year bridge were rebuilt, deferred-tax credit and recurring awards separated, restricted collateral removed, and options/claims checked. The weakest was persistence: the first candidate combined a plausible operating ramp with 14x/16x NOPAT exits that its mature capital costs could not support. Both advocates challenged it; the judge extended the existing engine and rejected the premium before the first forecast was frozen. Existing stage 9 already requires the exit multiple to agree with terminal growth, reinvestment and return on new capital. No new universal rule, engine feature or prompt checklist is justified. Apply that question properly. Scope: a low recent cash-capex rate does not establish the cost of the next capacity increment; equally, negative funded ramp cash is not a reason to reject productive investment. Counterexample: a documented mature facility with excess capacity could justify lower marginal capital, but that needs evidence rather than an assumed fall in capex. The correction is company-specific initial model calibration, not a newly discovered shared financial definition. No existing company's inputs, manual overrides or shared mechanics changed. Capital charts, assumptions, thesis, current sensitivities and dated output all use the judged exits. Independent initial drafts retain their clearly labelled provisional numbers only to make the judgment inspectable; they are not competing live forecasts. The missing project-level capacity/return data, product profit splits and exact next reporting date remain visible. Gates establish structural completeness and runnable cash arithmetic, not complete knowledge of the company. Future fiscal predictions are separately frozen so September results cannot be presented as a successful forecast. The first discrimination will be December-quarter shipments, EBIT after compensation and collections, followed by full-year cash/capex and evidence of repeat procurement. # Capital productivity and reinvestment 13 November 2026. USD millions. These are the shared engine’s forecast diagnostics, not measured historical ROIIC. | Horizon | Case | Partial capital return | Period-weighted reinvestment | New growth capital | Matching NOPAT | Funding residual | | --- | --- | ---: | ---: | ---: | ---: | ---: | | 5 years | downside | 27.12% | 45.62% | 3.377 | 7.402 | 8.88e-16 | | 5 years | base | 41.32% | 54.97% | 16.082 | 29.258 | 3.55e-15 | | 5 years | upside | 44.36% | 58.92% | 27.627 | 46.890 | 3.55e-15 | | 10 years | downside | 17.17% | 39.47% | 6.316 | 16.002 | 3.55e-15 | | 10 years | base | 29.48% | 38.18% | 26.513 | 69.450 | 7.11e-15 | | 10 years | upside | 30.63% | 40.52% | 48.681 | 120.143 | 7.11e-15 | The numerator of partial return is exit NOPAT minus starting normalized NOPAT; the denominator is cumulative capex less D&A plus operating WC investment. Much of the earnings change is repricing and recovery from underutilisation, so this is not the return independently earned on a new production line. Recurring labour, rent and share compensation remain inside EBIT. Unidentified intangible growth expense is not capitalized. Period-weighted reinvestment divides total matching new investment by total matching NOPAT; it does not average annual percentages. All full capex is paid before excess cash accumulates. Annual funding remains net-cash, excluding the operating reserve and restricted collateral. Minimum net excess cash is about 3.47 m downside,5.27 m base and5.52 m upside; intra-year needs remain unverified. A useful countercheck is the mature marginal increment rather than the large recovery numerator. Base year 6 adds0.2205m NOPAT against1.9647m new net physical/WC capital, about 11.2%. That weak mature productivity invalidated the provisional14x terminal premium. Shared-engine finite continuation now supports the selected8.5x instead; see12-debate.md. Maintenance≈D&A at the final no-growth endpoint is a disclosed proxy, not a reported schedule. Historical peak/trough windows in04-capital-allocation.md both have negative lagged capital changes, so no positive measured historical ROIIC or historical earnings-weighted reinvestment rate is asserted. Values and funding below each selected horizon are also available in updates/2026-11-13-valuation-audit.json. Sources:ieh-fy2022–2026,ieh-q2fy27 and valuation/calculations.json. # Scenario weights Dated 13 November 2026. These are coarse subjective joint-path judgments, not measured event frequencies, a target-return fit or a claim of calibrated investment skill. | Horizon | Downside | Base | Upside | Reason | | --- | ---: | ---: | ---: | --- | | Five years | 30% | 50% | 20% | Existing book and customer procurement support conversion, but capacity, repricing, concentration and controls keep a substantial adverse path | | Ten years | 35% | 40% | 25% | More programme renewal, succession and allocation uncertainty widens both sides; the longer period also permits more design wins | The paths were defined from orders, costs and funding before assessing weights. The debate retained operating upside, rejected unsupported terminal premiums and kept the weights: no new operating evidence justified fitting different odds to the lower valuation. Both advocates accept demand, while unmeasured project returns and margin persistence remain genuine uncertainty. Sources support those mechanisms, not30/50/20 precision. [ieh-q2fy27; ieh-exhibit-2026-11-13; bae-apkws-oct26; lmt-pac3-jul26; research/12-debate.md] The downside is not a catastrophic-loss bound. Cyber consequences, extended shutdown, loss of qualification, destructive acquisitions and dilution can be worse. The weighted output is the shared engine's IRR of probability-weighted dated receipts, not the weighted average of scenario IRRs. New information that changes an operating path requires reassessment; price movement alone does not. Local scenario edits suspend these researched weights. Frozen prospective checks retain the original predictions so future scoring cannot quietly fit the past. # IEHC independent bull–bear review Dated 13 November 2026. Separate advocates submitted independent openings before cross-examination; a separate neutral judge reviewed both, their rebuttals and primary evidence. All opening numerical effects below use the explicitly provisional8/14/16x exits. Final outputs and current sensitivities use7.5/8.5/8.5x; do not confuse the two. No historical author calibration was supplied. ## Bull case ### Bull opening — 13 November 2026 The strongest bear case is that IEH remains a weakly controlled, working-capital-heavy factory whose current price already anticipates a successful recovery. Restated FY2021 EBIT was only $2.58m on $34.72m sales, four of the next five fiscal years lost operating money, and the latest half-year's $2.02m net income yielded only $0.51m operating cash. A 14% base exit margin is a prospective achievement, not recovered historical fact (ieh-fy2022, operations and restatement note; ieh-fy2023–26; ieh-q2fy27 pp.1–4). The positive case is a change in customer production requirements, not an extrapolation of the recent share-price rise. September backlog is $53.75m, versus $18.40m a year earlier, for delivery over 6–24 months. IEH's separately announced Patriot, APKWS and medical awards identify actual orders. BAE's October framework targets capacity up to 35,000 APKWS kits/year, while Lockheed's July multiyear PAC-3 action supports a larger production base. Neither establishes IEH's future unit content or guarantees appropriations. Together with rising IEH shipments, they justify a real upside path rather than treating every new order as a temporary restocking bulge (ieh-q2fy27 MD&A; ieh-exhibit-2026-09-02/09-25; bae-apkws-oct26; lmt-pac3-jul26). The engine has already begun showing operating leverage: H1 revenue rose $7.34m, EBIT improved $2.57m, an incremental EBIT margin of 35.0%. Q2 EBIT margin was 9.78%, versus Q1's 6.63%, despite continuing gold/freight costs and a full compensation charge. This supports the mechanism behind later margins, without making 18% normal. Holding the base sales and valuation paths fixed, a two-point improvement in annual margins raises five-year IRR from 7.40% to 10.60%; ten-year IRR moves from 7.00% to 8.63% (ieh-q2fy27 operations and MD&A; shared-engine sensitivity). Capacity spending is another conservative part of the base. A 4% capex rate for ten years exceeds the FY2024–26 measured 2.30% rate; the latest half-year spent just $0.152m cash PP&E. It funds genuine growth, but there is no public machine-capacity schedule proving that spending must remain so high after the ramp. A 3% rate raises five-/ten-year base IRR to 7.94%/7.43%. The effect is modest: lighter spending alone does not rescue the valuation (ieh-fy2024–26 cash statements; ieh-q2fy27 pp.4; shared engine). The upside needs orders and throughput, not a richer multiple. Applying the upside sales schedule ($52m, $66m, $78m, $88m, $96m) to otherwise unchanged base economics produces 13.11% five-year and 10.30% ten-year IRR. The required repeat orders exceed today's disclosed book and must be observed before promotion. The base's five-year sales CAGR is 13.75% (about the 74th percentile of its aggregate reference class); upside is 21.17% (about the 86th). Neither percentile validates factory capacity or supplier allocation. I do not defend increasing the 14x base multiple: at a 12% hurdle it already includes a substantial premium above 8.33x no-growth earnings. Qualified programme persistence can support excess returns, but public evidence cannot establish their required post-exit duration. The strongest bull position is to retain a meaningful upside, monitor order conversion, and avoid a high-conviction buy at $31.25 until margins or price offer more room. ## Bear case ### IEHC bear opening — 13 November 2026 The strongest bull case is a qualified component supplier entering a genuine production ramp. September backlog was $53.75m, new Patriot/APKWS orders have identifiable programmes behind them, and medical orders widen the customer need. Existing excess cash can fund negative near-term free cash flow; this is not a distressed-balance-sheet thesis. The concern is the price paid for execution and persistence. 1. **Even the base needs a material operating improvement.** Q2 EBIT was $1.0465m on $10.7001m sales, or 9.78%; restated FY2021 EBIT margin was 7.45%, and FY2022–26 included four operating-loss years. The base reaches $70m sales and 14% EBIT within five years, versus $36.75m TTM sales and 2.56% EBIT today. Gold, labour, commissions and a larger production organisation still have to be paid. Holding EBIT at 10% while retaining the base revenue, capital and 14x exit gives only **0.21% five-year / 3.13% ten-year IRR**, against 7.40% / 7.00% initially. A smaller two-point margin reduction gives 3.78% / 5.12%. These are sensitivities, not measured probabilities. [ieh-q2fy27, operations pp.2/18–20; ieh-fy2022 restated operations; ieh-fy2023–2026.] 2. **The exit premium conflicts with the mature funding economics unless something changes.** In base year six, NOPAT rises $0.2205m, while net capex plus working capital consumes $1.9647m. That is roughly 11.2% model-implied incremental productivity, below a 12% hurdle. Three-percent growth requiring about 26% reinvestment implies approximately 8.2x NOPAT in a perpetual value-driver cross-check, not 14x. The current 14x multiple has a 40.5% premium above 1/12%; it needs a post-exit improvement in capital needs or profitability that has not been established. At 10x, base IRRs are **1.65% / 4.61%**; at 8.33x they are **−1.19% / 3.45%**. [model.json base paths; shared-engine year-six capital bridge; ieh-q2fy27 capital spending and working-capital disclosures.] 3. **Orders are not yet distributable cash.** H1 net income of $2.0179m includes a $0.4851m deferred-tax release. CFO was $0.5116m, capex $0.1517m and SBC $0.5215m: economic cash after replacement compensation was negative $0.1616m. Receivables and inventories absorbed $3.2027m, partially funded by $0.9125m customer advances. This can be productive investment, but it does not establish a permanent 32% working-capital requirement. Raising that requirement to 38% lowers base returns to **6.66% / 6.69%**. [ieh-q2fy27, cash flow p.4, Notes 2–3/8.] 4. **Qualification is valuable without being monopoly power.** One customer supplied 29.9% of H1 sales; orders may be postponed/cancelled. Smiths sells competing hyperboloid technology. September IT controls remained ineffective after the inventory restatement history, and August’s compromised mailbox exposed potentially export-controlled information; the issuer reported no evidence of exfiltration, not independent clearance. These support restrained conviction and explicit downside/tail coverage, not invented damages. [ieh-q2fy27 Notes 10 and Item 4; ieh-fy2026 backlog; smiths-hypertac; ieh-event-2026-08-06.] The database outside view places base five-year revenue CAGR near its 74th percentile, below the gate; that does not validate the margin or exit premium. The best observation against this bear case would be several quarters of cash conversion, repeat orders and sustainable EBIT above 12%, with an evidenced capex/capacity bridge. At $31.25, execution success alone offers insufficient margin of safety. ## Rebuttals ### Bull reply ### Bull rebuttal — 13 November 2026 **Margins: accept the risk; retain a genuine operating upside.** The bear correctly shows that 14% EBIT exceeds the restated FY2021 record and current Q2's 9.78%. H1's 35.05% incremental EBIT margin is evidence of absorption, not a steady-state margin guarantee. Orders plus programme expansion warrant a higher-throughput scenario, but not promotion to base without repeated margin and cash-conversion observations. A 10% flat margin remains a useful adverse sensitivity. **Terminal premium: concede.** No disclosed factory capacity, project-return cohort or maintenance/growth spending split establishes the post-exit productivity required for 14x/16x NOPAT. I extended the existing shared operating forecast to thirty years, continuing the last mature revenue growth and EBIT assumptions and the actual scenario capex/WC rates. Its unlevered free cash flows were discounted with shared `presentValue` at 12%; growth then stops and terminal NOPAT is capitalised at 1/12%, assuming sustaining capex equals D&A. No higher ROIIC was inserted. | Scenario / model exit | Five-year continuation | Ten-year continuation | Twenty-year continuation | | --- | ---: | ---: | ---: | | Base / year 5 | 8.377x | 8.408x | 8.441x | | Base / year 10 | 8.380x | 8.410x | 8.442x | | Upside / year 5 | 8.564x | 8.564x | 8.563x | | Upside / year 10 | 8.332x | 8.331x | 8.330x | These modest differences include the timing of the actual annual cash flows; they do not rescue the premium. Keeping the full capex rate after growth stops lowers the base cross-check to about 7.44–8.17x. The correct response is to revise terminal capitalisation or substantiate lower mature capital requirements, not to invent a high-return tail. Diagnostic: `/tmp/iehc-research/bull-tail.mjs`; source inputs: model's annual paths, ieh-q2fy27 Notes 3/6/7 and statements, ieh-fy2022–26 capex record. This is an open-book consistency check, not a measured project return. **Cash conversion: accept the uncertainty, reject a blanket negative-FCF objection.** Receivables and inventories finance orders that can be attractive. The bear explicitly recognizes this. Higher working capital belongs in the downside and monitoring; it does not by itself establish bad growth. The 4% capex base remains a sensitivity, not a measured maintenance requirement. Lowering it to 3% adds only about 0.53 points to initial five-year IRR, insufficient to create a margin of safety. **Option claims: accept the stronger denominator check.** Issued shares plus all September/October options equal 3.002454m. Crediting their $8.404m aggregate exercise proceeds once gives initial-base five-/ten-year IRRs of 6.82%/6.52%, versus 7.40%/7.00% under frozen-price treasury equivalents. This is a relevant sensitivity because rising exit prices increase dilution relative to the entry-price approximation. Full exercise is not a valid assertion for a downside below strike; treat settlement coherently. Sources: ieh-q2fy27 option table and ieh-event-2026-10-27. No future SBC add-back is justified. **Controls and competition: concede the surviving attacks.** Qualification and reliable delivery have value, but Smiths disproves technological exclusivity, while ITGC weaknesses and potentially export-controlled mailbox contents remain unresolved. The evidence warrants restrained conviction and a meaningful downside; it does not warrant invented damages or an automatic order-loss estimate. The strongest bull case survives as a conditional operating outcome, not a high-conviction investment at $31.25. ### Bear reply ### Bear cross-examination and concessions — 13 November 2026 The bull's opening was read only after the independent bear opening was saved. **Backlog and replenishment.** Accept that this is more than a narrative: actual orders and customer production programmes support higher shipments. Withdraw any suggestion that the base should assume an immediate return to the FY2023 trough. The existing downside already grants first-year revenue of $43m before deferrals. But $53.75m of backlog for 6–24 months does not establish $70m recurring annual sales, nor the labour and capital productivity to achieve them. The bullish $96m sales sensitivity is a valid upside, not a reason to promote it without order replenishment and throughput evidence. Sources: ieh-q2fy27 MD&A; ieh-exhibit-2026-09-02/09-25; bae-apkws-oct26; lmt-pac3-jul26. **Operating leverage.** Accept the 35% H1 incremental EBIT margin as evidence that absorption and repricing work. It cannot be a steady-state margin: the comparison starts with a loss, gross margins changed, and commissions/production staffing rise as output expands. The Q1-to-Q2 margin improvement is encouraging but two quarters do not establish a mature 14–18% margin. A two-point margin sensitivity is appropriate in both directions. The bear does not argue that negative near-term FCF alone invalidates the ramp. Source: ieh-q2fy27 operations/MD&A. **Capex.** Concede that ten years at 4% sales is not an observed requirement; 3% raises the base five-year return to 7.94%, still below the hurdle. Conversely, the latest $0.152m half-year outlay cannot establish the cost of doubling throughput. Mature capex of 2.7–3% may be an explicit sensitivity after a funded ramp; it needs a capacity/maintenance bridge, not extrapolation from recent light spending. Sources: ieh-fy2022, FY2024–26 cash flow; ieh-q2fy27 pp.4/20. **Exit premium survives as the strongest unresolved attack.** At 12% required return, 3% long-run growth and the existing margin/working-capital paths, a value-driver cross-check gives approximately 8.47x current NOPAT with 4% mature capex, 9.56x at 3%, or 9.89x at 2.7%. These multiples include next-year growth when expressed against current earnings; the next-year-NOPAT equivalents are 8.23x/9.29x/9.60x. Even zero reinvestment costs would cap a 3% perpetuity at 11.44x current earnings. A 14x exit therefore needs faster post-exit growth or a lower market discount rate (about 9.36% with 2.7% capex), independently justified. Discounting the actual base years 6–10 unlevered FCF plus an 8.33x no-growth exit at year 10 gives 8.38x year-five NOPAT. These are mathematical consistency checks, not measured project returns. The public sources do not establish the missing persistence or discount-rate argument. The bull itself declines to defend a richer exit. **Judgment proposed.** Retain real upside and the funded first-year cash deficit. Lower the base terminal valuation or explicitly show a supported post-exit bridge before retaining 14x. Retain low conviction: sustainable margins, capacity cost and controls remain unresolved. No quantified cyber fine, automatic zero recovery, or extra tax/SBC charge is justified by the current evidence. ## Judgment The order recovery is supported. The initial exit premium is not. Retain the operating scenarios as explicit estimates, but change downside/base/upside terminal capitalisation from **8x/14x/16x to 7.5x/8.5x/8.5x unlevered NOPAT**. Higher sales and margins must earn the upside; they do not also receive an unsupported rerating. The bull and bear saved independent openings before receiving each other's work. Both written rebuttals were read. The judge checked the September filing, October employment disclosure and the shared engine, rather than treating agreement between advocates as validation. ### Contested claims | Claim | Evidence and counterargument | Judgment | | --- | --- | --- | | Actual orders and programme funding support a higher production level | September backlog $53.751m, deliveries over 6–24 months; named Patriot, APKWS and medical orders; customer production/funding actions. Cancellation, replenishment, IEH unit content and factory capacity remain uncertain. | **Adopt the bull mechanism, not the upside as base.** Retain $48m/$57m first two base rolling years and $70m by year five as explicit judgments. The downside still starts at $43m before deferrals. Do not add individual awards to backlog a second time. | | Operating leverage supports 14–18% mature EBIT | H1 incremental EBIT margin is 35.05%, Q2 EBIT is 9.78%, but the restated FY2021 margin was 7.45% and four of FY2022–26 lost money. The incremental comparison starts at a loss and is not mature unit economics. | **Adopt-modified.** Keep the causal margin paths, with 14% base and 18% upside explicitly prospective, not historically achieved earning power. The bear's 10% ceiling and ±2-point paths remain serious sensitivities and prospective tests. No stronger base forecast is justified. | | Lower capital spending or working capital should improve the base | FY2024–26 cash capex averaged 2.30% of sales, versus forecast 4% base. September operating WC was 32.1% of trailing sales. Neither observation establishes capacity cost for doubling output. | **Unresolved, current judgments retained.** Fund 4% capex/32% WC in base and 5%/30% in upside; downside uses 3%/38%. No blanket rejection of funded negative FCF, and no invented maintenance-versus-growth split. | | 14x/16x NOPAT is supported by a durable qualified franchise | The actual mature paths keep capex at 4%/5% sales, D&A at 2.207%, and growth at 3%. Their future cash does not justify those multiples at the chosen 12% required return. Both advocates conceded the premium. | **Reject the premium.** Use the finite continuation cross-check below, with all forecast capital costs paid. Qualification is not proof of a monopoly or a quantified post-exit excess-return duration. | | Controls, concentration and cyber risk invalidate the business | One customer was 29.9% of H1 sales, ITGC controls remained ineffective in September, and the August mailbox incident exposed potentially controlled technical information. The issuer said it had no evidence of exfiltration, not that the matter was independently cleared. | **Adopt the risks, reject fabricated losses.** Preserve meaningful downside and low conviction. No sourced amount establishes a fine, automatic customer departure or zero recovery. Smiths' competing hyperboloid products preclude a technology-exclusivity claim. | | Known option claims require more than basic shares | September 480,000 options at $15.35 weighted strike plus the 35,000 CFO grant at $29.60 are material. Full exercise plus $8.404m proceeds is an alternative to the entry-price treasury proxy, but cannot be imposed indiscriminately below strike. | **Adopt the sensitivity.** Keep the transparent 2.733526m fixed-price economic-share proxy, retain all future economic compensation in EBIT, and disclose full-exercise/proceeds and dilution sensitivities. No second proceeds credit or SBC add-back. | Primary locations: `ieh-q2fy27` operations/cash flow, Notes 2–3, 6–10, MD&A and Item 4; `ieh-fy2022` restated operations/Note 3; `ieh-fy2023`–`ieh-fy2026` cash statements; `ieh-event-2026-10-27` Item 5.02; `ieh-event-2026-08-06` Item 8.01; `ieh-exhibit-2026-09-02` and `ieh-exhibit-2026-09-25`; `bae-apkws-oct26`; `lmt-pac3-jul26`; `smiths-hypertac`. ### Exit capitalisation and duration The independent diagnostic extends the **shared operating forecast** through year 30, continuing each scenario's last stated growth and EBIT margin with the same capex, D&A and WC definitions. Discount future unlevered FCF at a **judged 12% required return**, then stop growth and capitalise terminal NOPAT at 1/12%. The no-growth endpoint assumes sustaining capital equals D&A; this remains a proxy, not an observed maintenance schedule. Cash already accumulated at the original exit is excluded from the operating multiple and added separately by the main model. Bank debt is also counted once. | Scenario / original exit | Five-year continuation | Ten-year continuation | Twenty-year continuation | | --- | ---: | ---: | ---: | | Downside / year 5 | 7.724x | 7.344x | 6.957x | | Downside / year 10 | 7.727x | 7.346x | 6.959x | | Base / year 5 | 8.377x | 8.408x | 8.441x | | Base / year 10 | 8.380x | 8.410x | 8.442x | | Upside / year 5 | 8.564x | 8.564x | 8.563x | | Upside / year 10 | 8.332x | 8.331x | 8.330x | **Choose 7.5x/8.5x/8.5x**, rounded valuation judgments around the ten-year continuation results. The 5–20-year table exposes duration sensitivity rather than asserting a measured remaining programme life. The downside's costly 2% expansion earns less cash per NOPAT unit, so more duration does not deserve a premium. Base/upside do not need an extra post-exit moat premium: 8.5x is only about 2% above 8.33x no-growth NOPAT at the selected required return. The upside five-versus-ten-year difference is roughly 0.23x; this is smaller than the evidence uncertainty and does not justify a new shared engine feature. A single disclosed 8.5x is adequate at both horizons, with the duration table retained. A 14x base with only 3% terminal growth cannot be defended merely by lowering capex to 2.7–3%: the perpetual current-NOPAT cross-check is still only about 9.9x–9.6x. Raising the exit back to 14x would require independently justified faster growth, better capital productivity or a lower required return. None is established. The 12% rate is an underwriting hurdle, not a verified market cost of equity. ### Same-price effect At **$31.25, 13 November 2026**, no revenue, margin, tax, capital, financing or share input changed in this judgment. | Horizon | Case | Before | After | | --- | --- | ---: | ---: | | 5 years | Downside | −23.05% | −23.60% | | 5 years | Base | 7.40% | −0.89% | | 5 years | Upside | 22.18% | 9.66% | | 5 years | Weighted cash-flow IRR | 6.03% | −2.67% | | 10 years | Downside | −9.94% | −10.21% | | 10 years | Base | 7.00% | 3.57% | | 10 years | Upside | 14.36% | 9.41% | | 10 years | Weighted cash-flow IRR | 6.44% | 2.67% | The joint operating-path weights were reassessed and retained: 30/50/20 at five years and 35/40/25 at ten. The exit correction supplies no new probability evidence and weights must not be fitted to restore an attractive return. Recompute all displayed sensitivities under the new capitalisation; the advocates' opening numbers belong to the initial model only. ## Conviction Low conviction in a favourable investment at $31.25. Actual demand and available cash rebut a distressed-factory thesis; the surviving attacks are unproven mature margins/capacity economics, concentration, controls and a valuation requiring more than currently underwritten cash. The revised upside itself is below the 12% hurdle. A lower price can change the decision without changing business odds. Freeze future reported-period tests on repeat orders, revenue conversion, EBIT after compensation, receivables/inventory/advances, cash after full investment, capex/capacity and ITGC remediation. Higher margins accompanied by credible cash and funded throughput would rebut the operating bear case. A backlog headline alone would not. No precise issuer date was verified for the next quarterly release; February 2027 remains an estimated reporting window. This is an **open-book consistency correction during initial underwriting**, not an investment-skill test or historical calibration. Existing stage 9 already requires reconciling the exit multiple with terminal reinvestment economics; apply that rule rather than claiming a new universal method. Other company assumptions were not examined by this judgment and no cross-company financial change is implied. One factual correction: the CFO agreement/grant was executed **23 October 2026**, not 24 October; Item 5.02 and the Form 4 agree. It does not change the amount or model return. ## Public Primary & Secondary Sources - `ieh-fy2022` (issuer filing): IEH FY2022 Form 10-K (restated FY2020–2021 comparatives) — https://www.sec.gov/Archives/edgar/data/50292/000121390023050974/f10k2022_iehcorp.htm (published 2023-06-22) - `ieh-fy2023` (issuer filing): IEH FY2023 Form 10-K — https://www.sec.gov/Archives/edgar/data/50292/000121390023081163/f10k2023_iehcorp.htm (published 2023-10-06) - `ieh-fy2024` (issuer filing): IEH FY2024 Form 10-K — https://www.sec.gov/Archives/edgar/data/50292/000121390024052905/ea0207042-10k_iehcorp.htm (published 2024-06-14) - `ieh-fy2025` (issuer filing): IEH FY2025 Form 10-K — https://www.sec.gov/Archives/edgar/data/50292/000121390025053885/ea0243237-10k_iehcorp.htm (published 2025-06-12) - `ieh-fy2026` (issuer filing): IEH FY2026 Form 10-K — https://www.sec.gov/Archives/edgar/data/50292/000121390026068122/ea0292779-10k_iehcorp.htm (published 2026-06-12) - `ieh-q2fy27` (issuer filing): IEH September 2026 Form 10-Q — https://www.sec.gov/Archives/edgar/data/50292/000121390026118861/ea0310481-10q_iehcorp.htm (published 2026-11-12) - `ieh-ir` (issuer filing): IEH investor relations — https://www.iehcorp.com/about/investors (published 2026-11-13) - `ieh-proxy26` (issuer filing): IEH definitive proxy, 30 October 2026 — https://www.sec.gov/Archives/edgar/data/50292/000207709626000297/ea0309212-02.htm (published 2026-10-30) - `ieh-edgar` (issuer filing): SEC complete recent filing index — https://data.sec.gov/submissions/CIK0000050292.json (published 2026-11-13) - `ieh-event-2026-11-13` (issuer filing): IEH 2026-11-13 Form 8-K — https://www.sec.gov/Archives/edgar/data/50292/000117494726000991/form8k-36531_iehc.htm (published 2026-11-13) - `ieh-event-2026-10-27` (issuer filing): IEH 2026-10-27 Form 8-K — https://www.sec.gov/Archives/edgar/data/50292/000117494726000937/form8k-36462_iehc.htm (published 2026-10-27) - `ieh-event-2026-09-25` (issuer filing): IEH 2026-09-25 Form 8-K — https://www.sec.gov/Archives/edgar/data/50292/000117494726000897/form8k-36375_ieh.htm (published 2026-09-25) - `ieh-event-2026-09-02` (issuer filing): IEH 2026-09-02 Form 8-K — https://www.sec.gov/Archives/edgar/data/50292/000117494726000837/form8k-36298_iehc.htm (published 2026-09-02) - `ieh-event-2026-08-07` (issuer filing): IEH 2026-08-07 Form 8-K — https://www.sec.gov/Archives/edgar/data/50292/000117494726000780/form8k-36203_iehc.htm (published 2026-08-07) - `ieh-event-2026-08-06` (issuer filing): IEH 2026-08-06 Form 8-K — https://www.sec.gov/Archives/edgar/data/50292/000117494726000761/form8k-36187_iehc.htm (published 2026-08-06) - `ieh-event-2026-06-12` (issuer filing): IEH 2026-06-12 Form 8-K — https://www.sec.gov/Archives/edgar/data/50292/000117494726000663/form8k-36028_iehc.htm (published 2026-06-12) - `ieh-event-2026-06-08` (issuer filing): IEH 2026-06-08 Form 8-K — https://www.sec.gov/Archives/edgar/data/50292/000117494726000648/form8k-36013_ieh.htm (published 2026-06-08) - `ieh-event-2026-05-18` (issuer filing): IEH 2026-05-18 Form 8-K — https://www.sec.gov/Archives/edgar/data/50292/000207709626000129/ea0291349-8k_iehcorp.htm (published 2026-05-18) - `ieh-event-2026-04-22` (issuer filing): IEH 2026-04-22 Form 8-K — https://www.sec.gov/Archives/edgar/data/50292/000117494726000500/form8k-35769_iehc.htm (published 2026-04-22) - `ieh-event-2026-03-26` (issuer filing): IEH 2026-03-26 Form 8-K — https://www.sec.gov/Archives/edgar/data/50292/000117494726000432/form8k-35627_iehc.htm (published 2026-03-26) - `ieh-event-2026-03-04` (issuer filing): IEH 2026-03-04 Form 8-K — https://www.sec.gov/Archives/edgar/data/50292/000117494726000279/form8k-35434_iehc.htm (published 2026-03-04) - `ieh-event-2026-02-20` (issuer filing): IEH 2026-02-20 Form 8-K — https://www.sec.gov/Archives/edgar/data/50292/000117494726000205/form8k-35310_iehc.htm (published 2026-02-20) - `ieh-event-2026-02-10` (issuer filing): IEH 2026-02-10 Form 8-K — https://www.sec.gov/Archives/edgar/data/50292/000117494726000154/form8k-35238_iech.htm (published 2026-02-10) - `ieh-event-2026-01-15` (issuer filing): IEH 2026-01-15 Form 8-K — https://www.sec.gov/Archives/edgar/data/50292/000117494726000054/form8k-35124_iehc.htm (published 2026-01-15) - `ieh-exhibit-2026-11-13` (issuer filing): IEH November results release (issued 12 November) — https://www.sec.gov/Archives/edgar/data/50292/000117494726000991/ex99-1.htm (published 2026-11-13) - `ieh-exhibit-2026-09-25` (issuer filing): IEH Medical $3.6m orders (issued 24 September) — https://www.sec.gov/Archives/edgar/data/50292/000117494726000897/ex99-1.htm (published 2026-09-25) - `ieh-exhibit-2026-09-02` (issuer filing): IEH APKWS $6.5m orders — https://www.sec.gov/Archives/edgar/data/50292/000117494726000837/ex99-1.htm (published 2026-09-02) - `ieh-exhibit-2026-05-18` (issuer filing): IEH Patriot $7.9m orders — https://www.sec.gov/Archives/edgar/data/50292/000207709626000129/ea029134901ex99-1.htm (published 2026-05-18) - `ieh-exhibit-2026-03-26` (issuer filing): IEH Patriot $5.1m order (issued 24 March) — https://www.sec.gov/Archives/edgar/data/50292/000117494726000432/ex99-1.htm (published 2026-03-26) - `ieh-exhibit-2026-04-22` (issuer filing): IEH April investor presentation — https://www.sec.gov/Archives/edgar/data/50292/000117494726000500/ex99-1.htm (published 2026-04-22) - `ieh-exhibit-2026-06-08` (issuer filing): IEH June investor presentation — https://www.sec.gov/Archives/edgar/data/50292/000117494726000648/ex99-1.htm (published 2026-06-08) - `ieh-exhibit-2026-01-15` (issuer filing): IEH SEC opinion dismissing delinquency proceeding — https://www.sec.gov/Archives/edgar/data/50292/000117494726000054/ex99-1.htm (published 2026-01-15) - `lmt-pac3-jul26` (customer release): Lockheed Martin PAC-3 MSE multiyear award — https://news.lockheedmartin.com/2026-07-29-Department-of-War-Awards-Lockheed-Martin-58-62B-for-Multiyear-PAC-3-MSE-Production-to-Strengthen-the-Arsenal-of-Freedom (published 2026-07-29) - `army-pac3-apr26` (government release): US Army PAC-3 MSE contract action — https://www.army.mil/article/291670/u_s_army_advances_accelerated_pac_3_mse_production_through_contract_action (published 2026-04-10) - `bae-apkws-oct26` (customer release): BAE APKWS production framework — https://www.baesystems.com/en-us/article/bae-systems-us-department-of-war-sign-long-term-agreement-to-increase-apkws-production (published 2026-10-19) - `bae-nov26` (customer release): BAE 12 November trading update — https://www.baesystems.com/en-us/article/market-update-nov-2026 (published 2026-11-12) - `smiths-hypertac` (competitor product evidence): Smiths Interconnect Hypertac technology — https://www.smithsinterconnect.com/our-company/who-we-are/our-technology-brands/hypertac/ (published 2026-11-13) - `ieh-quote` (market data): IEHC timestamped Yahoo Finance quote — https://finance.yahoo.com/quote/IEHC/ (published 2026-11-13) - `ieh-price-history` (market data): IEHC public price history — https://stockanalysis.com/quote/otc/IEHC/history/ (published 2026-11-13) - `amphenol-q3-2026` (primary): Amphenol Q3 2026 issuer results — https://investors.amphenol.com/news-and-events/news-details/2026/Amphenol-Reports-Record-Third-Quarter-2026-Results-and-Announces-Dividend-Increase-and-New-Stock-Repurchase-Program/default.aspx (published 2026-10-28) --- # Lumine Group (LMN.V) - Legal Name: Lumine Group Inc. - Exchange: TSX Venture (Canada) - Sector: Communications and media software - Currency: CAD - Reviewed: 2026-11-15 - Model Framework: Recurring software cash reinvestment - Financial Period: TTM September2026; September cash, gross bank principal and remaining acquisition claims; fixed October2026 exit vintage (anchored 2026-10-26) - Research Status: 8 of 11 research checks complete. Still to complete: Security and evidence base; Opportunities and funded scenarios; Variant view and catalysts - Static Report URL: https://longview-investing.vercel.app/research/lumine.html - Interactive Model URL: https://longview-investing.vercel.app/#/company/lumine - JSON API: https://longview-investing.vercel.app/api/companies/lumine ## Executive Summary Acquire mission-critical software, strengthen operations and redeploy after-financing cash. Recurring customer revenue supports the cash engine; acquisition price, integration and financing determine per-share compounding. ## Key Operational Watch Items - Recurring retention and organic growth versus acquisition growth - Working-capital normalization and post-acquisition cash margins - Debt refinancing, deferred consideration and deployment discipline ## Valuation Scenario Assumptions ### Downside Case An explicit stress, not issuer guidance: 2% annual baseline cash erosion, 50% deployment to preserve liquidity, purchases costing 75% more per unit of cash earnings and a 35% lower terminal multiple. This tests weaker retention, integration and financing sentiment; it is not a financing-default simulation. Purchase economics are financing-neutral with respect to legacy group debt; owner cash retains that debt cost. Terminal value uses finite postexit cash and capped annual purchases, with the researched year-specific multiple path; peer multiples are reference evidence only. - Key Drivers: Dividend / available owner cash: 0.0% | Annual share dilution: 0.0% | Available cash deployment fraction: 50.0% - Organic revenue growth: -4.0%, -4.0%, -1.0%, -1.0%, -1.0% - Owner-cash margin on organic revenue: 25.5%, 25.5%, 25.5%, 25.5%, 25.5% - Price paid per unit of acquired revenue: 2.00×, 2.00×, 2.00×, 2.00×, 2.00× - Owner-cash margin of acquired businesses: 17.0%, 17.0%, 17.0%, 20.0%, 20.0% - Acquired revenue growth after purchase: -2.0%, -2.0%, -2.0%, -2.0%, -2.0% - Deal budget (capital available for acquisitions): 23.00, 23.00, 105.57, 108.61, 111.36 - Exit / cash operating earnings: 8.58×, 8.57×, 8.57×, 8.56×, 8.55× ### Base Case Revenue build (3 November). Organic revenue grows 1% in year 1, 2% in years 2-5 and 1.5% after. Its owner-cash margin recovers from 26.6% to 27.8% as 2026 strengthening costs roll off. New deals buy revenue at 1.6x (2026 pricing) and then at the 1.55x record, and converge from 22% to the 30% group margin by year 4. The deal budget is 81.9 a year in years 1-2 (debt-repayment years) and then the 131.7 FY2021-FY2025 average (USD; inputs.yearly.deploymentCapacity); unplaced cash repays debt, then earns a deposit rate. Exit at 17.7x (Topicus trailing, converted). Purchase economics are financing-neutral; owner cash retains the legacy debt cost. Terminal value uses finite postexit cash and capped annual purchases, with the researched year-specific multiple path; peer multiples are reference evidence only. - Key Drivers: Dividend / available owner cash: 0.0% | Annual share dilution: 0.0% | Available cash deployment fraction: 100.0% - Organic revenue growth: 1.0%, 2.0%, 2.0%, 2.0%, 2.0% - Owner-cash margin on organic revenue: 26.2%, 27.0%, 27.8%, 27.8%, 27.8% - Price paid per unit of acquired revenue: 1.60×, 1.55×, 1.55×, 1.55×, 1.55× - Owner-cash margin of acquired businesses: 22.0%, 26.0%, 26.0%, 30.0%, 30.0% - Acquired revenue growth after purchase: 0.0%, 0.0%, 0.0%, 0.0%, 0.0% - Deal budget (capital available for acquisitions): 114.79, 114.79, 398.20, 469.93, 544.19 - Exit / cash operating earnings: 19.97×, 19.67×, 18.86×, 18.05×, 17.28× ### Upside Case An explicit favorable case, not issuer guidance: 3% baseline cash growth, purchases 15% cheaper per unit of cash earnings and a 20% higher exit multiple. Full reinvestment still requires funded purchases and the declared timing policy. Purchase economics are financing-neutral with respect to legacy group debt; owner cash retains that debt cost. Terminal value uses finite postexit cash and capped annual purchases, with the researched year-specific multiple path; peer multiples are reference evidence only. - Key Drivers: Dividend / available owner cash: 0.0% | Annual share dilution: 0.0% | Available cash deployment fraction: 100.0% - Organic revenue growth: 2.0%, 3.0%, 3.0%, 3.0%, 3.0% - Owner-cash margin on organic revenue: 27.5%, 29.0%, 29.0%, 29.0%, 29.0% - Price paid per unit of acquired revenue: 1.60×, 1.55×, 1.55×, 1.55×, 1.55× - Owner-cash margin of acquired businesses: 24.0%, 28.0%, 28.0%, 31.0%, 31.0% - Acquired revenue growth after purchase: 0.0%, 0.0%, 0.0%, 1.0%, 1.0% - Deal budget (capital available for acquisitions): 673.13, 730.07, 730.07, 730.07, 730.07 - Exit / cash operating earnings: 31.03×, 27.44×, 24.91×, 22.87×, 21.59× ## Research Notebook # Business model Lumine buys and operates software businesses serving communications and media customers. Money is USD millions unless stated. Source IDs refer to `sources/registry.json`; fact IDs to `financials/facts.json`; `rec*` and other step IDs to `valuation/calculations.json`. ## Evidence plan What we are buying: subordinate voting shares of Lumine Group Inc. on the TSX Venture Exchange (LMN), 256.62m economic shares including one super-voting share held by Constellation Software (CSU) that carries 50.1% of votes and one share's economics (`economicShares`, lumine-q2-2026-fs). CSU also owns about 157.6m subordinate shares, a 61.4% economic interest (`csuLumineShares`, csi-q2-2024-mda). Lumine reports in USD; the shares trade in CAD. The fiscal year is the calendar year. The Q3 2026 report date is not confirmed in issuer sources (lumine-events). | Question | Evidence used | What it cannot settle | | --- | --- | --- | | Group economics, cash, capital | FY2022-FY2025 MD&As, FY2025 audited statements, Q2 2026 interim statements and MD&A | Business-unit P&Ls: Lumine reports one segment | | Acquisition economics | Acquisition notes (FY2025 FS note 4, Q2 2026 FS notes 4 and 17), Motive announcement, Synchronoss 10-K and deal release | Post-close cash by deal; Motive's and WideOrbit's own revenue | | Demand and cycle | Revenue by type and organic growth by quarter since 2021; customer and peer filings (Verizon, T-Mobile, Comcast, Nokia, Ericsson, Ribbon, Telenor) | Contract renewal rates, churn, price increases | | Counterparties | Customer concentration notes, Synchronoss customer note, board page, CSU filings | The identity of Lumine's >10% customer; contract terms | | Management | MD&A related-party and key-management notes, board page, CSU MD&A | The management information circular was not obtained; individual pay and holdings are unverified | No fieldwork was done: no interviews with customers, former employees, sellers or competitors. Management statements are evidence of what management says. ## Value-creation mechanism Lumine buys established, mission-critical software that communications and media companies run their networks, devices, billing, advertising and broadcast operations on, usually when the seller no longer wants it: corporate carve-outs (Nokia's device management and service platform businesses as Motive, Arvato's Vidispine, Synamedia's video network assets in Quortex), founder exits (WideOrbit), or under-earning public companies (Synchronoss). It pays roughly 0.6-1.6x revenue, then "strengthens" the business: it removes overhead the business no longer needs inside a decentralized group, prices the installed base, and keeps funding maintenance and development so customers renew. The cash then funds the next purchase. **Most value is created in the strengthening step**, not the purchase multiple and not organic growth: group operating margin before amortization rose from 26.3% in FY2022 to 29.0%, 31.5% and 36.0% (FY2023-FY2025; `opIncome2022`-`opIncome2025`) while organic growth was −9%, 0%, −8% and +2% (`organic2022`-`organic2025`). Acquired revenue arrives at low margins and leaves at group margins. The measured FY2022-FY2025 achieved deal yield is about 20-26% after tax and before financing (stage 6), well above any plausible cost of capital, while organic growth contributes nearly nothing. The mechanism needs a supply of unloved assets at low prices; it does not need customers to buy more. ## Business-model profile Comparable firms are chosen by these traits, not by the industry label. **Growth vehicle: Acquisitions.** Revenue rose from 228.4 in FY2021 to 765.7 in FY2025, about 35% a year, while organic growth was −9%, 0%, −8% and +2% in FY2022–FY2025. Acquisition cash was 27.0% of revenue over FY2023–FY2025. Essentially all growth was bought, and value comes from strengthening the margins of acquired businesses, not from organic growth. The model buys revenue each year from a deal budget and grows the base at about zero. **Capital intensity: 0.4% of sales, asset-light.** Capital spending is about 0.4% of revenue, and tangible capital including leases is about 5 because customers fund working capital. The capital that drives growth is acquisition cash (27% of revenue over FY2023–FY2025), spent mostly on carve-outs of communications and media software. **Comparables.** A comparable firm buys software businesses, often carve-outs from larger vendors, serving a narrow customer set; earns mostly recurring maintenance revenue (70% in FY2025); raises margins after purchase; and needs almost no capital spending. Organic growth is near zero and growth is bought. Not the label: "Telecom software" ties Lumine to telecom capex cycles and organic SaaS peers; its value comes from buying and improving acquired businesses. Declared in `valuation/model.json` (`businessProfile`). `npm run check` measures deal and capital intensity from the reported facts, and the outside view grades the forecast only against firms that share them. ## Economic snapshot FY2025, reported statements, definitions stated (fact IDs in brackets; normalization below). | Measure | Value | Definition | | --- | ---: | --- | | Revenue (`revenue2025`) | 765.7 | of which maintenance and other recurring 537.7 (70%) | | Operating income before amortization (`opIncome2025`) | 275.7 (36.0%) | issuer non-IFRS; expensed R&D 101.5 staff is inside | | Delivery margin | about 75% | revenue less hardware 11.1, third-party 46.2, professional-services staff 59.5 and maintenance staff 75.5 | | Incremental margin | 31.7% FY2023, 38.9% FY2024, 67.1% FY2025, 13.6% H1 2026 | change in operating income / change in revenue | | Unlevered owner cash before working capital (`recEndCash`) | 245.9 (32.1%) | operating cash after tax, lease cash and capex, before bank interest and interest income | | FCFA2S (`fcfa2025`) | 217.0 | issuer: after interest, leases and capex, before deals | | Trade working capital | −9.5 | receivables 163.2 + unbilled 47.5 + inventory 0.6 − payables and accruals 123.8 − deferred revenue 97.0 | | PP&E / right-of-use assets | 8.3 / 5.8 | | | Tangible capital incl. leases | about 5 | trade working capital + PP&E + ROU | | Return on tangible capital (leases capitalized), after tax | not meaningful | about 239 of after-tax operating income on about 5 of capital | | Return on capital employed (leases expensed), after tax | not meaningful | negative trade working capital | | Return incl. net acquired intangibles and goodwill | 32.6% after tax | (275.7 − 36.7 current tax) / (728.4 + 5) | | Return incl. gross intangibles (before 414.1 accumulated amortization) | 20.8% after tax | / (1,142.5 + 5); book values partly carried at CSU's historical cost | | Receivable days (incl. unbilled) | 78 (100) | on revenue | | Deferred revenue | 46 days of revenue | maintenance billed in advance | | Capex / depreciation | 0.49x; capex + lease cash / depreciation 1.1x | depreciation includes right-of-use assets | | Net cash (debt), 31 Dec 2025 | +140.4 | cash 352.4 − bank debt 212.0; leases 6.8 | | Net debt after July 2026 closings | about 275 | debt 485.0 − cash after July payments 210.4 | Normalization before returns. Capex and lease cash together roughly equal depreciation, so no capex adjustment; there are no minorities. Recently acquired businesses are not in trailing earnings at full weight; the model adds them as funded cohorts (Synchronoss's missing months at the group cash margin; Quortex and Imagine at the purchase yield). Two items the model does not separately normalize: (1) the FY2024 working-capital outflow of 77.9 and the FY2025 outflow of 18.5 (the model treats ongoing working capital as zero; stage 6 treats the FY2024 build as part of the Motive purchase cost); (2) FY2025 current tax includes a 14.8 benefit from the US repeal of R&E capitalization, part of which is a one-time catch-up of previously capitalized costs (lumine-fy2025-mda p.10). The split is not disclosed; if half is non-recurring, normalized owner cash is overstated by about 7 (3%). This is recorded as an unquantified upward bias, not plugged. The real capital in this business is the purchase price, not the balance sheet. Book intangibles understate it because Lumine Holdings came in at CSU's historical book value in a common-control transaction. Stage 6 therefore measures returns on matched acquisition capital. ## Economic segmentation Lumine reports one operating segment. Re-cut first by revenue type, because each type has different economics (FY2025; direct costs allocated by the staff department table, lumine-fy2025-mda pp.4-6; estimates): | Stream | Revenue | Direct cost (est.) | Contribution (est.) | Share of contribution | Organic FY2025 / TTM Jun 2026 | Character | | --- | ---: | ---: | ---: | ---: | --- | --- | | Maintenance and other recurring | 537.7 | maintenance staff 75.5 + about 30 third-party | about 432 | about 75% | +2% / +3% | replacement-like; renews unless the customer migrates | | Licences | 72.6 | small | about 72 | about 13% | −7% / +5% | lumpy upgrades and new sales; −36% in FY2024 | | Professional services | 135.0 | PS staff 59.5 + about 16 third-party | about 60 | about 10% | +5% / −2% | implementation work tied to licence and upgrade cycles | | Hardware and other | 20.3 | 11.1 | about 9 | about 2% | −1% / −21% | resale at 40-45% margin | Contribution totals about 573; R&D 101.5, sales and marketing 49.0, G&A 67.4 and other opex about 80 bring it to 275.7. R&D and most sales and G&A exist to keep the recurring base renewing, so effectively all operating profit comes from maintenance and recurring revenue. Licence and service swings move reported growth but not the value. Re-cut second by acquired cohort, because the cohorts have different demand drivers (FY2025 revenue estimated from the issuer's pro forma adjustments: 241.4 of FY2022 pre-acquisition revenue for the 2023 cohort, 228.2 of FY2023 for the 2024 cohort; fact-level detail in the MD&As): | Cohort | Est. FY2025 revenue | Demand driver | Trend | | --- | ---: | --- | --- | | Pre-2023 portfolio (Telarix interconnect, broadband and mobile operator software) | about 230-240 | operator network and billing operations | flat to slightly down; one 2022 deal lost customers who had given notice before purchase | | 2023 cohort, mostly WideOrbit (TV and radio advertising sales and traffic software), WizTivi | about 240-270 | linear TV and radio advertising volumes and station counts | structural decline in linear advertising; licences fell sharply in FY2024 | | 2024 cohort, Motive (device and service management from Nokia) and Axyom | about 200-240 | operators' device and home-gateway management | carries the single >10% customer (71.5 in nine months of 2024, 106.3 in FY2025) | | 2025 deals (Vidispine, Datafusion) | about 31 | media asset management | small | | Since the period: Synchronoss (Feb 2026), Quortex and Imagine (Jul 2026) | about 170 (Synchronoss annualized Q2) plus undisclosed | carrier-branded consumer cloud storage; broadcast playout and video delivery | Synchronoss top five customers 97.6% of its revenue | Conclusion. The profit is in recurring maintenance on installed software, spread across cohorts whose end markets range from slowly declining (linear TV/radio advertising) to flat (operator back-office). No cohort is a growth engine. Concentration has risen: one undisclosed customer was 13.9% of FY2025 revenue, and Synchronoss adds Verizon and AT&T exposure. Counting Synchronoss at its annualized Q2 revenue, at least 40% of run-rate revenue (about 425 of about 950, before the July deals, whose revenue is undisclosed) now comes from businesses bought since April 2024. ## Unit economics The unit through which Lumine adds value is the acquired business. The comparison that matters is before and after Lumine's ownership. Three disclosed units: | | Synchronoss (Feb 2026) | Motive and Axyom (Apr 2024) | Datafusion (Jul 2025) | | --- | ---: | ---: | ---: | | Price | 309.3 consideration; 277.5 all-in net of 34.3 cash plus 2.6 costs (`synchronossNetAllInCost`) | up to EUR185m incl. EUR35m contingent; FY2024 group deal cash 140.6 plus 77.9 working-capital build | 21.35 incl. 7.7 holdbacks and contingent | | Revenue before purchase | 173.6 (FY2024, `sncrRevenue2024`) | about 195 (2023, from the issuer's pro forma adjustment, est.) | about 33 annualized (16.4 in the first six months) | | Price / revenue | 1.6x | about 1.0-1.1x incl. working capital | 0.65x | | Profit before purchase | income from operations 21.7 (12.5%); unlevered cash 28.5 (16.4%) = operating cash 28.3 + interest 15.1 − capex and capitalized software 14.9 | not disclosed (Nokia carve-out) | not disclosed | | Entry yield (unlevered cash / all-in price) | 10.3% | n/a | n/a | | After Lumine | at the group's 29.2% purchase cash margin (`purchaseCashMargin`) on 169 revenue: about 49, a 17.8% yield. Observed so far: net losses of 1.4 (Q2) and 3.9 (since close) after acquired-intangible amortization on 286 of intangibles | group unlevered cash rose 62.7 from FY2024 to FY2025 on 218.5 of FY2024 deal capital and working capital (`recRoiic2025`, 28.7%), mostly a full year of Motive | net income 1.7 in six months after amortization | What the strengthening step changes, using Synchronoss. Synchronoss spent 29.8% of revenue on SG&A (51.7 on 173.6) and 24.7% on R&D (42.8) as a listed company. Lumine's whole group spends 6.4% on sales and marketing staff, 8.8% on G&A staff and 13.3% on R&D staff. Reaching the group margin needs about 21 of annual savings (12% of revenue), mostly listed-company and duplicated overhead (board, audit, listing, investor relations, separate management layers) and sales cost on a base whose top five customers are 97.6% of revenue and renew on three-to-five-year contracts. That is plausible but unproven; it is the single largest assumption in the funded cohorts. A representative small deal, built from Datafusion and group ratios (illustrative): price 21 for 33 of revenue. At purchase it earns perhaps 10-15% unlevered cash margin (about 4); if Lumine lifts it to 25-30% over two to three years (about 9), the yield on price is 19% at entry-year cash and about 40% once strengthened, before working capital. Capital beyond the price is close to zero because maintenance is billed in advance, except in carve-outs that arrive without receivables. Failure cases. (1) A 2022 acquisition was bought with "known material attrition from customer notifications" received before purchase, which produced −9% organic growth in FY2022 (lumine-fy2022-mda p.3): Lumine priced in a shrinking base, so this was a known decline rather than a mistake, but it shows the model's tolerance for declining assets. (2) After WideOrbit's first year, FY2024 organic growth was −8%, with licences −36% and hardware −46% (lumine-fy2024-mda p.5): the linear advertising customer base bought fewer upgrades. Maintenance held at +2%, which is the point of the model. (3) Synchronoss so far: reported losses after amortization, strengthening costs that lifted H1 2026 expenses to 70% of revenue against 66% a year earlier, and an incremental group margin of only 13.6% in H1 2026. ## Independent checks 1. **Unit model against group accounts.** Units bought at 0.6-1.6x revenue that reach a 25-30% unlevered cash margin earn about 20-40% on price at maturity. The group's achieved FY2022-FY2025 deal yield measured from consolidated cash is 19.9-25.7% (stage 6, `recYieldLow`, `recYieldHigh`). The two are consistent once integration time and the FY2024 working-capital build are charged. 2. **Margin path against deal timing.** Each large deal year was followed by a margin step: 29.0% (FY2023, WideOrbit), 31.5% (FY2024, Motive at nine months), 36.0% (FY2025, a full year of strengthened Motive and no large deal). H1 2026, with Synchronoss and strengthening costs, fell to 30.0%. The pattern of a dip followed by recovery is what the mechanism predicts; FY2026-FY2027 margins test it again. 3. **Adverse allocation.** Assign the whole FY2022-FY2025 margin gain to the pre-2023 portfolio and ignore its organic decline: the deal yield still comes out at 19.9%, above CSU's group aggregate of 15.1% (`aggAchievedYield`) and above a 9-10% cost of equity. The conclusion that Lumine's purchases created value survives. Whether the rate is repeatable at 2026 prices is open: Synchronoss's entry yield was 10.3% and its price carried a 70% premium to the market (csi-synchronoss-release-2025), though entry yields of the record's deals were also below their achieved yields. The model prices future deals at the 19.9% low bound (3 November label reconciliation). 4. **Peer cross-check.** Synchronoss as a standalone listed company earned 16.4% unlevered cash on revenue; Ribbon and the network vendors (ribbon-q3-2026, nokia-q3-2026, ericsson-q3-2026) carry hardware and growth spending Lumine doesn't. Lumine's 32% unlevered cash margin sits at the CSU-family level for mature vertical software. Where the consensus may disagree. The 2025-2026 sell-off (from about C$55 in July 2025 to about C$19-26 in 2026, lumine-price-history-2026-11-03) is attributed in public commentary to AI disruption of vertical software and to slower deal flow. This study's reading is narrower: the risk to value is not that customers stop renewing (maintenance organic growth stayed +1% to +3%) but that the 2026 deals were bought at entry yields well below Lumine's record, with concentrated customers. The observation that would settle it is Synchronoss's post-close cash margin and maintenance organic growth in the Q3 and Q4 2026 reports. Sources: lumine-fy2022-mda, lumine-fy2023-mda, lumine-fy2024-mda, lumine-fy2025-mda, lumine-fy2025-fs, lumine-q2-2026-fs, lumine-q2-2026-mda, lumine-motive-announcement-2023, sncr-fy2024-10k, csi-synchronoss-release-2025, lumine-playbook. # History, demand and counterparties Money is USD millions unless stated. ## Reporting perimeter The listed company dates from February 2023: CSU transferred Lumine Holdings to the company at book value, the company bought WideOrbit for 504.6 (cash and debt repayment plus 10.2m special shares valued at 221.8), and CSU distributed the subordinate shares to its own shareholders (lumine-fy2023-mda pp.2-3). FY2021-FY2022 figures are the predecessor's combined carve-out accounts. In March 2024 the preferred and special shares converted into subordinate shares, and the accrued 5% dividends (about 87) were paid in shares. June 2026 statements show 256.620389m economic shares and no dilutive securities. Trailing revenue is FY2025 plus H1 2026 less H1 2025: 846.4. The same bridge gives 185.4 of reported FCFA2S. Synchronoss (closed 13 February 2026) is in trailing results for part of a year; Quortex and Imagine (1 July) are not. Their purchase cash is deducted from the June cash balance, which already includes the debt drawn to prefund them. The latest financial results located are Q2, released 4 August 2026; Q3 accounts were not linked in issuer sources on 2 November (lumine-filings-public-table). ## Demand drivers and cycle End user, payer and decision-maker. The end users are network engineers, operations staff, advertising sales and traffic teams at broadcasters, and consumers using carrier-branded apps (Synchronoss). The payer is the operator or broadcaster's IT or network budget. The decision-maker on renewal is usually an operations or IT executive for whom the system is already embedded; the decision-maker on replacement is a CTO-level programme that rarely touches a working back-office system unless a platform migration (5G core, IP/cloud playout, a merger) forces it. Demand split, judged from revenue type (FY2025, lumine-fy2025-mda p.4): | Part | Share of revenue | Driver | Evidence | | --- | ---: | --- | --- | | Replacement-like (maintenance, support, hosted, subscriptions) | 70% (76% in H1 2026 with Synchronoss) | installed base continues to run the software | organic +2% FY2024, +2% FY2025, +3% TTM June 2026 (reported currency); no quarter below −1% since Q1 2023 (it fell as low as −11% in Q3 2022) | | Upgrade and new-sale (licences, professional services) | 27% | operator and broadcaster project budgets; platform migrations | licences −36% FY2024, −7% FY2025; services −18% FY2024, +5% FY2025 | | Resale (hardware) | 3% | project-attached | −46% FY2024 | | Regulation-driven | small, not separable | lawful intercept, emergency messaging, broadcast compliance | no disclosure | The cycle measured with data, not adjectives: total organic growth by year was −9% (FY2022, one deal's known attrition), 0% (FY2023), −8% (FY2024), +2% (FY2025) and +1% TTM June 2026 (`organic2022`-`organicTtmJun2026`). The swings are almost entirely licences, services and hardware; maintenance stayed between −1% and +8% by quarter. Customers' own budgets explain this split. Operators are cutting physical capex while service revenue grows slowly: T-Mobile reported lower capital spending in Q3 2026 (tmobile-q3-2026), Verizon and Comcast report flat-to-low connectivity growth (verizon-q3-2026, comcast-q3-2026), and Telenor's Nordic operations grow low single digits with disciplined capex (telenor-q3-2026-release). Network-vendor software segments (Nokia Core Software, nokia-q3-2026) are growing modestly. That environment squeezes project and licence spending but leaves running systems in place. Market misperception, if any. The 2026 market narrative is that AI agents and cloud platforms will displace vertical software. The series that tests it for Lumine is maintenance and recurring organic growth: +1% FX-adjusted in Q2 2026 and +3% reported over the trailing year (lumine-q2-2026-mda p.6). It hasn't turned negative. A second, opposite misperception is possible: that Lumine is a growth company. It isn't; outside acquisitions, revenue has been flat for four years. The WideOrbit cohort faces a structural decline in linear TV and radio advertising, which no data here quantifies. ## Competition Competition happens product by product, at the operator or broadcaster account, not at the group level. Lumine's units compete with the large network and IT vendors (Nokia, Ericsson, Amdocs, Oracle, Netcracker) in operator software, with broadcast technology vendors (Grass Valley's AMPP and Altitude playout, grass-valley-ampp-2026) in channel origination, and with hyperscaler and in-house tools in cloud storage and video processing. Lumine has no meaningful market share at the industry level; each unit is a niche incumbent. The advantage in one sentence: Lumine is often the buyer of last resort for products their original owners no longer want to fund, and it can earn a high return on them because the installed customers keep paying for support while Lumine cuts the cost of serving them. The evidence it is durable is the record of sellers (Nokia, Bertelsmann's Arvato, Synamedia, a public board) choosing Lumine, the measured 20-26% deal yield, and maintenance revenue that kept growing slightly while licence sales fell by a third. Who could hurt the company. (1) Other permanent owners of vertical software competing for the same carve-outs: CSU's own operating groups, Topicus, private equity and other software consolidators; prices for listed targets (Synchronoss, a 70% premium) suggest more competition for larger deals. (2) Platform migrations: a customer moving to a new core network, a cloud broadcast platform or a merger partner's systems can end a renewal. Grass Valley's AMPP win at Channel 4 is an example of a competitor's platform replacing incumbent playout (no Lumine contract is identified). (3) The largest customers themselves, which can insource. Synchronoss's 10-K notes the cost and subscriber disruption Verizon or AT&T would face in replacing it (sncr-fy2024-10k), which is why they haven't, but that is the seller's own claim. The number of competitors is not falling in Lumine's niches in any observable way; what changes is which products get abandoned by large vendors, which is Lumine's deal supply. ## Counterparty incentives | Counterparty | What they want | What Lumine gives them | Outside options | What would change the arrangement | | --- | --- | --- | --- | --- | | CSU (61.4% economic, 50.1% votes via the super-voting share) | per-share compounding of the whole CSU family; capital placed at high returns without CSU management time | a dedicated communications-and-media acquirer, management fees of 3.0 a year (`mgmtFees2025`), an exit for deals outside other groups' verticals | CSU's Volaris and other groups can buy the same assets; CSU can sell or distribute shares | CSU needing cash, a change in CSU leadership philosophy (Mark Miller, CSU President and COO, chairs Lumine), or a conflict over which group gets a deal | | Corporate sellers (Nokia, Arvato, Synamedia) | a clean exit from non-core products without stranding their own customers or brand | speed, certainty, a commitment to continuity (Lumine's 2023 Motive release stresses partnering with Nokia for a seamless transition) | shutdown (costly, angers customers), PE buyers | sellers learning to price carve-outs higher once Lumine's results are visible | | Large customers (the undisclosed >10% customer; Verizon and AT&T through Synchronoss) | reliable systems at falling cost; leverage at renewal | continuity of embedded software, no migration risk | insource, migrate to a platform vendor, renegotiate | a platform migration, merger, or a renewal at lower price. Synchronoss's top five were 97.6% of its revenue; Lumine's single customer was 106.3 in FY2025 (13.9%) after 71.5 in nine months of FY2024, nil before Motive | | WideOrbit's rollover holders (founder Eric Mathewson sits on the board) | value for shares received at a US$21.74 face value | about 10.1m special shares that converted at 3.43 subordinate shares each, about 35m shares (about 13.5%) at conversion (derived; later sales not tracked) | sell into a thin market | a large sale could press on the price; no evidence of one was located | | Subsidiary lenders (WO Loan, Motive Loan, Synchronoss Loans, Lumine Facility) | covenant-compliant, ring-fenced cash flows | collateral of each subsidiary; no parent guarantee on the subsidiary loans | refuse refinancing | covenant stress at one subsidiary; maturities in 2028-2030 | Dependence runs both ways. The large customer relationships are a risk to revenue but also the reason the assets are cheap and sticky: operators rarely rip out working device-management or storage platforms, and the seller (often the vendor that sold to them) wants Lumine to keep them served. The identity of the >10% customer is not disclosed; its appearance with the April 2024 Motive carve-out (and a 44.3 receivable at December 2024, lumine-fy2024-mda p.12) suggests it is tied to that business, either a major operator or Nokia as a reseller. This is an inference, not a disclosed fact, and it matters: a reseller has different renewal incentives from an end user. CSU as controlling holder: it has never sold Lumine shares in the record located, holds a permanent-owner culture, and benefits from Lumine's per-share value roughly in proportion to its 61% stake, so its incentives largely match minority holders'. The agency risk is allocation of deals across CSU's groups, which outsiders cannot observe. Sources: lumine-fy2022-mda, lumine-fy2023-mda, lumine-fy2024-mda, lumine-fy2025-mda, lumine-q2-2026-mda, sncr-fy2024-10k, csi-q2-2024-mda, csi-synchronoss-release-2025, lumine-motive-announcement-2023, lumine-board-page, tmobile-q3-2026, verizon-q3-2026, comcast-q3-2026, telenor-q3-2026-release, nokia-q3-2026, grass-valley-ampp-2026. # Management as capital allocators Money is USD millions unless stated. The 2026 management information circular was not obtained in this pass, so individual pay, holdings and bonus terms are marked unverified. That is the main gap in this stage. ## Ownership and incentives Who decides where the cash goes: David Nyland (CEO and director; previously CEO of NBS Technologies and Blueprint Software Systems; according to the issuer, he and his team built the portfolio since 2013), and a board chaired by Mark Miller, CSU's President and COO. CSU controls the vote through one super-voting share (50.1% of votes, one share's economics) and owns about 157.6m subordinate shares, 61.4% of the economics (`csuLumineShares`, csi-q2-2024-mda). | Holder / director | Role | Economic interest | Purchases and sales | Evidence | | --- | --- | --- | --- | --- | | Constellation Software | controlling holder | about 157.6m shares, 61.4% (C$4.1bn at C$26.15) | no sale located since the 2023 spin-out | csi-q2-2024-mda | | WideOrbit rollover holders incl. Eric Mathewson (director, WideOrbit founder) | seller turned holder | about 35m shares at the March 2024 conversion (about 13.5%), derived from 10.1m special shares × 3.43; current holding unverified | not tracked | lumine-fy2023-mda p.22; lumine-board-page | | David Nyland | CEO, director | a third-party aggregator reports about 0.07% (about 0.17m shares, about C$4m); unverified | unverified | aggregator only | | Mark Miller, Brian Beattie, Laurie Schultz | CSU/Volaris executives and CSU director | personal holdings unverified | unverified | lumine-board-page | | Lucie Laplante (audit chair), Paul Cowling | independent directors | unverified | unverified | lumine-board-release-2023 | Pay. Key management personnel (executives and the board together) received 6.955 in FY2025 and 6.302 in FY2024, all salaries, bonus and benefits (`keyMgmtComp2025`, lumine-fy2025-mda p.14). There are no stock options, no share-based payment line and no dilution from compensation: shares outstanding were unchanged from the March 2024 conversion to June 2026. In the CSU family, executives' bonuses are tied to return on invested capital and growth, and a share of the after-tax bonus must buy shares in the market, held in escrow for years (CSU's own practice). Whether Lumine's plan copies it was not verified here. What a stake is worth if the stock compounds. On the aggregator's figure, Nyland's stake at 12% a year would gain about C$3m over five years, smaller than five years of a CSU-family executive's cash pay. On that reading his interests are dominated by pay and by CSU's evaluation of him, not by his own shares. The dominant owner is CSU: a 12% compound on its C$4.1bn stake is about C$3.1bn over five years, against 3.0 a year of management fees. CSU's interest is per-share value. Implied targets. No numerical hurdle rate is disclosed; the MD&A states the objective "to invest all of our FCFA2S in acquisitions which meet our hurdle rate" (lumine-fy2025-mda p.1). No incentive target can be backed out without the circular. ## Track record and references Record under this team, measured: - Total shareholder return: listed at about C$14-19 in April 2023 (lumine-price-history-2026-11-03), peaked near C$55 in July 2025, C$26.15 at the 26 October anchor. From the first full month of trading, roughly +40% over 3.5 years, about 10% a year, with no dividends. Most of the 2023-2025 rise reversed in 2026. - Returns on capital: the achieved FY2022-FY2025 deal yield is 19.9-25.7% after tax, before financing (stage 6), against CSU's group aggregate of 15.1%. FCFA2S per diluted share rose from 0.15 (FY2022 pro forma) to 0.36 (FY2023), 0.33 (FY2024) and 0.85 (FY2025). - Market share: not measurable; Lumine's units are niche incumbents. - Decisions that failed or look weak: the 2022 deal with known attrition (organic −9%, priced in by design); WideOrbit's licence decline in FY2024; FY2025 deployment of only 16.8 net while cash rose to 352 and the share price was high, which was cautious rather than wrong; Synchronoss in 2026 at a 10.3% entry yield and a 70% market premium, still unproven. Independent references. None were collected: no former employees, sellers or customers were interviewed. Public evidence from counterparties is limited to the seller's continuity quote in the Motive release, two board observers drawn from operators (Proximus and Spring Fibre, lumine-board-release-2023) and Synchronoss's board accepting the offer. That is zero independent references, stated plainly. Bench depth. Each acquired business keeps its own management (Motive, WideOrbit, Synchronoss and Imagine continue under their brands) and Lumine runs a decentralized model inherited from Volaris. The Volaris and CSU executives on the board are an outside bench. Key-person risk at Nyland exists but is reduced by CSU's ability to appoint. ## Capital allocation principles Governing objective: invest all FCFA2S in acquisitions above a hurdle rate, with no dividend. The behaviour shows long-term value per share as the goal, with one caveat (principle 5). 1. **Zero-based allocation.** Capital goes only to acquisitions or debt repayment; there are no internal projects capitalized. When no deals cleared the hurdle in FY2025, Lumine repaid 125.7 of debt and held cash rather than forcing deals. Evidence it moves capital away from incumbents: there are no divestitures in the record, so this half is untested. 2. **Funding strategies, not projects.** Each large deal is financed on its own subsidiary's cash flows without a parent guarantee (WO Loan for WideOrbit, Motive Loan, Synchronoss Loans), so a deal's financing capacity is set by the acquired business's cash, not by a central budget. 3. **Capital is accessible but has a cost.** The spin-out funded WideOrbit partly with 221.8 of special shares and CSU's 181.5 of preferred shares, both carrying a 5% dividend paid in about 87 of shares in March 2024 and convertible at fixed ratios. That issuance cost existing owners, priced at a time the stock was near C$15-20. There is no stock compensation disguising financing. 4. **Zero tolerance for bad growth.** No exit or shutdown is disclosed. A declining WideOrbit licence business is kept because maintenance cash continues; that is consistent with harvesting rather than growing. Untested. 5. **Price versus value.** Lumine did not buy back shares when the stock fell about 60% from July 2025 to April 2026, and did not issue shares when it was high. Instead it spent about 520 in seven months of 2026 on acquisitions at entry yields far below its own record. If the stock's own owner-cash yield (about 6% at C$26 on run-rate cash) was below the deal yields, that choice is defensible; it was below Lumine's record yields but not obviously below Synchronoss's 10% entry yield. The CSU family's stated preference for acquisitions over buybacks is a policy, not a price-versus-value test. ## Integrity and governance Concrete evidence: the FCFA2S definition and the organic-growth definition are stated and reconciled every quarter; prior-period restatements were described as immaterial. No adjustment to bonus metrics was observable (no circular). Related-party transactions: management fees to the CSU parent (3.0 in FY2025, 2.8 in FY2024), reimbursement of parent expenses, and the 2023 common-control transfer of Lumine Holdings at book value with preferred shares to CSU. Board: seven directors, of whom four are CSU-affiliated (Miller, Beattie, Schultz, Nyland), one is a major selling holder (Mathewson) and two are independent (Laplante, Cowling, both from Canadian telecom and cable). Voting control (50.1%) sits with CSU regardless of economic ownership. Agency costs (Mauboussin and Callahan pp.47-48): - **Control of resources.** Pay that grows with the size of the empire would favour more deals. Pay terms are unverified; the counterweight is CSU's 61% economic stake and its return-based culture. Moderate risk, untested. - **Risk tolerance.** A concentrated holder (CSU) and a CEO without a large stake: CSU is diversified across its groups, so it may accept riskier, larger deals (Synchronoss) than a single-company owner would. The ring-fenced subsidiary debt limits the damage. - **Time horizon.** CSU is a permanent holder; the horizon matches a long-term owner. - **Controlling holder.** Minority holders depend on CSU not allocating the best deals to wholly owned groups, and on it not changing the arrangement (fees, a take-private at a low price). No evidence of either was found. Buybacks do not offset share compensation because neither exists. The valuation treats the agency cost by charging the management fee as an operating cost, applying no control premium, and pricing future deals at the low bound of Lumine's record rather than its midpoint, partly because deal allocation inside the family is unobservable. Sources: lumine-fy2023-mda, lumine-fy2025-mda, lumine-q2-2026-fs, lumine-q2-2026-mda, csi-q2-2024-mda, lumine-board-page, lumine-board-release-2023, lumine-motive-announcement-2023, csi-synchronoss-release-2025, lumine-price-history-2026-11-03. # Capital allocation Money is USD millions unless stated. Measurements are named steps in `valuation/calculations.json` (`rec*`, research audit steps not bound to the model). ## Sources and uses of cash From the statements (FY2021-FY2022 predecessor carve-out accounts, lumine-fy2022-mda p.9; FY2023 lumine-fy2023-mda p.13; FY2024-FY2025 lumine-fy2025-fs; H1 2026 lumine-q2-2026-fs). | | FY2021 | FY2022 | FY2023 | FY2024 | FY2025 | Five years | H1 2026 | | --- | ---: | ---: | ---: | ---: | ---: | ---: | ---: | | Operating cash before working capital (after tax) | n/a | 61.5 | 120.7 | 192.8 | 254.9 | | 111.7 | | Working capital | n/a | −26.9 | −12.5 | −77.9 | −18.5 | | −23.4 | | Operating cash flow | 86.0 | 34.6 | 108.2 | 115.0 | 236.5 | 580.3 | 88.3 | | Lease cash (principal and interest) | | −3.0 | −5.9 | −6.6 | −5.0 | | −4.1 | | Capex net / depreciation incl. ROU | n/a | −0.8 / 5.3 | −1.1 / 7.9 | −2.9 / 9.1 | −4.0 / 8.1 | | −1.8 / 4.4 | | Acquisitions, net of cash acquired | −13.5 | −114.6 | −317.0 | −140.6 | −16.8 | −602.5 | −277.1 | | Non-cash consideration (special shares to WideOrbit holders) | | | (221.8) | | | (221.8) | | | Dividends / buybacks | 0 | 0 | 0 | 0 (preferred dividends of about 87 paid in shares) | 0 | 0 cash | 0 | | Equity issued for cash (preferred to CSU, capital contributions) | | | 204.0 | | | 204.0 | | | Net transfers with CSU (intercompany advances and repayments) | −79.6 | +104.9 | −33.1 | | | | | | Bank debt drawn less repaid | 0 | +19.7 | +135.0 | +127.7 | −70.7 | +211.7 | +273.2 | | Bank interest and fees | | −0.5 | −12.3 | −20.9 | −14.6 | | −8.8 | | Other (interest received, restricted cash, FX) | | | | | | | | | Net change in cash | −8.0 | +40.0 | +79.4 | +64.5 | +141.5 | +317.4 | +69.2 | | Closing cash | 27.0 | 67.0 | 146.5 | 211.0 | 352.4 | | 421.7 | Rows reconcile to each year's reported net change once the "other" items (interest received, restricted cash, post-acquisition settlements and FX, each under 10 a year) are included; they are left blank rather than plugged. FY2021-FY2022 intercompany flows with CSU reflect the carve-out before the spin-out. What the table shows about priorities. Over FY2022-FY2025 the business generated about 630 of operating cash before working capital and spent 589 net of cash acquired on acquisitions plus 222 of special shares; working capital absorbed 136, mostly in the Motive year. External money (CSU preferred and contributions 204, net bank debt 212) funded the gap, and cash built to 352 at the end of FY2025 before the 2026 deals. No cash was ever returned to shareholders. In H1 2026 Lumine drew 380 of new debt and repaid 107 to buy Synchronoss and prefund the July deals. ## Investment rates per sales change Window FY2022 to FY2025: revenue rose 510.0 (`recSalesChange`). Rates per dollar of sales change: | Rate | Value | Step | Reading | | --- | ---: | --- | --- | | Incremental M&A rate | 1.37 | `recMnaRate` (696.2 incl. the WideOrbit special shares / 510.0) | Lumine buys revenue at about 1.4x, consistent with deals at 0.6-1.6x revenue plus holdbacks | | Internally generated intangibles (expensed R&D) | 0.12 | `recRdRate` (R&D staff +31.1 / revenue +266.0, FY2023-FY2025) | R&D grows with acquired teams; it is maintenance of products, not discretionary growth | | Fixed capital rate | 0.00 | `recFixedRate` (capex and lease cash 25.5 vs depreciation 25.1) | no fixed capital intensity | | Working capital rate | 0.21 investment | `recWcRate` (outflow 108.8 / 510.0) | unusual for software; almost all from FY2024, when the Motive carve-out arrived without receivables | Inflection points. The working-capital rate turned from a source to a use in FY2024 (Motive carve-out; largest-customer receivable 44.3 from nil). The M&A rate collapsed in FY2025 (16.8 net), then jumped in 2026 (about 520 of cash in seven months). The same CEO was in charge throughout; CSU-appointed chairs oversaw both. ## Reinvestment and returns Owner earning power at the start and end of the window, measured as unlevered owner cash before working capital (operating cash after tax, lease cash and capex, before bank interest): - FY2022: 57.7 (`recStartCash`), on the predecessor perimeter under CSU. - FY2025: 245.9 (`recEndCash`), a 32.1% margin. - After financing, the current normalized TTM September 2026 owner cash is 243.8 (`normalizedOwnerCash`). Reinvestment rate. FY2023-FY2025 unlevered cash before working capital totalled about 543 (113.7, 183.3 and 245.9). Growth investment was 696 of acquisitions including non-cash consideration plus 109 of working capital, about 805: a 148% reinvestment rate, funded by CSU preferred cash, special shares and bank debt. Incremental return, lagged. The cash change of 188.2 against matched capital of 823.0 (half of the FY2022 and FY2025 deal years, all of FY2023-FY2024 including the WideOrbit special shares and the FY2024 working-capital build) gives 22.9% (`recIncrementalReturn`). A one-year lagged rolling figure, FY2025 cash change 62.7 on FY2024 capital of 218.5, gives 28.7% (`recRoiic2025`), inflated by a full year of Motive and its cost reductions. A three-year rolling series isn't possible yet: the listed history is three years and one deal (WideOrbit) dominates it. Operating leverage versus capital efficiency. The FY2025 incremental margin of 67% (ΔNOPAT ÷ Δsales at the operating line) was mostly operating leverage from cutting Motive's costs, a one-time gain per deal. Capital efficiency is the durable part: the business needs no fixed capital and, outside carve-outs, no working capital. Growth identity. Growth in owner cash ≈ (investment ÷ starting earnings) × return on new capital: (823 ÷ 57.7) × 22.9% ≈ 3.3x, matching the observed rise from 57.7 to 245.9 (+3.3x of the base). It holds by construction because the inputs share definitions; the useful check is that the matched capital and cash definitions don't double count. The investment came from outside capital, so per share the result is smaller: FCFA2S per diluted share went from 0.15 (FY2022 pro forma) to 0.85 (FY2025), a 5.6x rise, helped by the low pre-spin base. The mix: organic investment is expensed (R&D) and contributed about nothing net (organic −6% cumulative); acquisitions earned 20-26%; payouts were zero. ## Acquisition record | Date | Deal | Price | Revenue at purchase | Later results | | --- | --- | ---: | --- | --- | | 2022 | Several, incl. a business with known pre-purchase attrition; Telarix-related term loan October 2022 | 113.2 cash incl. holdbacks | undisclosed | organic −9% in FY2022 from the attrition, as expected at purchase | | Feb 2023 | WideOrbit (TV and radio advertising software) | 504.6: about 282.8 cash and debt repayment plus 221.8 of special shares | the 2023 cohort had 241.4 of FY2022 revenue (pro forma; mostly WideOrbit) | group margin 29.0% in FY2023; licences −36% in FY2024 | | Nov 2023 | WizTivi | undisclosed; €10m loan | undisclosed | loan repaid by 2026 | | Apr 2024 | Motive (Nokia device management and service management platform) and Axyom Core | up to €185m incl. €35m contingent; FY2024 deal cash 140.6 plus 77.9 working-capital build | 2024 cohort had about 195 of FY2023 revenue (est. from the 228.2 pro forma adjustment) | FY2025 group unlevered cash +62.7 | | Apr 2025 | Vidispine (from Arvato/Bertelsmann) | 9.0 incl. holdbacks; bargain-purchase gain 2.2 | 14.9 contributed in nine months | net income 1.0 in nine months after amortization | | Jul 2025 | Datafusion | 21.35 incl. 7.7 holdbacks and contingent | 16.4 contributed in six months | net income 1.7 in six months | | Feb 2026 | Synchronoss | 309.3; 277.5 all-in net of cash | 173.6 (FY2024); 42.3 in Q2 2026 | net losses 3.9 since close after amortization; cash margin unobserved | | Jul 2026 | Quortex (Synamedia video network assets) | 94.652 updated total consideration | Q3 21.511 | Q3 net income 1.120 includes 4.904 bargain gain | | Jul 2026 | Imagine Communications | 136.832 updated total consideration | Q3 35.611 | Q3 net loss 6.826; cash margin unobserved | Aggregate. FY2022-FY2025 matched capital of 823.0 bought businesses that, with their strengthening, added 163.8-191.8 of unlevered owner cash by FY2025 (`recAcquiredCashLow`, `recAcquiredCashHigh`). The achieved deal yield is **19.9% (low) to 25.7% (high); 23.3% midpoint including carve-out working capital** (`recYieldLow`, `recYieldMid`, `recYieldHigh`). Bounds, as stage 6 requires: the high bound credits all margin gains to deals and charges the reported −6% cumulative organic decline to the legacy base; the low bound gives the legacy base the FY2025 margin (32.1%) and no decline. The base business's start-year margin was 22.6%. Organic growth from FY2024 includes WideOrbit after its first year, which moves some of WideOrbit's decline into the "legacy" base and flatters the high bound; the low bound removes that. Comparison and deployment pace. CSU's measured group yield over 2020-2025 is 15.1% (`aggAchievedYield`), the model's base purchase price until the 3 November label reconciliation and now its downside anchor. Lumine's own record is higher, but it rests on two large carve-out and founder deals bought between February 2023 and April 2024, and the 2026 deals were priced differently: Synchronoss's entry yield was 10.3% and its offer a 70% premium to market. Deployment was lumpy: 539 in FY2023 including shares, 141 in FY2024, 17 in FY2025, about 520 in seven months of 2026. The old January2024–July2026 average of about 265 a year is a diagnostic. Current budgets instead use matched internal funding ratios, with explicit first-two-year limits and a fixed upside supply ceiling; see the September update below. Conclusion for current deal economics. The historical low-bound 19.9% supports a limit to test, not a single copied yield. The revenue forecast pays 1.6×/1.55× sales, with acquired cash margins improving over time; mature base yield is 19.35%. Downside matures at 10% and upside at 20%. Group strengthening and the limited deal record constrain these judgments. Historical calibration and frozen checks stay in their original records; current annual paths and capital sums are in the model and `research/10-capital-efficiency.md`. Then `research/10-capital-efficiency.md` contains the shared-engine forward diagnostics. ## September 2026 cash and capital update Source: Q3 statements pp.2, 7, 12–15, 18–20, 23–24; MD&A p.11, published 4 November. Nine-month CFO 155.173, financing 189.333, investing −491.507 and FX −3.040 reconcile opening cash 352.441 to closing 202.400. Acquisition cash was 527.751, acquired cash 43.480 and older settlement payments 5.699. Q3 bank repayments were 59.375, leaving gross principal 425.625. The model uses gross principal, rather than 422.526 book debt net of deferred fees. Remaining holdbacks 12.268 plus separate contingent consideration 2.313 replace earlier June/July claims. July acquisition cash is already in September cash. Actual target revenue now shortens the missing-month bridge; Quortex's 4.904 bargain-purchase gain is not recurring operating cash, and Imagine's Q3 loss does not establish a group cash margin. Matched internally funded deployment over January2022–September2026 is 0.6845 of reported matching FCFA2S (`internalDeployRatioSep26`), compared with 0.6318 in the old July window. Historical all-in and internal ratios are kept separately. The base carries the two existing repayment-period budgets, then uses the updated ratio to state annual dollar budgets. Deal supply and actual available cash still constrain deployment; a new accounting window is not a larger seller opportunity set. Nine-month current tax 45.808 less cash paid 36.913 leaves an 8.895 timing diagnostic. It is not automatically a permanent tax benefit. Product development (R&D staff 115.2) remains charged; no unsupported growth-intangible capitalization is added. The model assumes refinancing of the remaining principal and does not model a default or new leverage. Principal due in 2028–2030 remains material. [Current forecast, funding and return attribution](../updates/2026-11-04-daily-review.md). # Thesis and valuation Assessment:5 November 2026. Communications and media software is costly for customers to replace, but recurring revenue still faces customer bargaining and technology substitution. Lumine buys businesses, strengthens them and reinvests owner cash. Acquisition-led revenue growth, recurring organic growth and target margin recovery are different mechanisms. The Q3 report improves the starting cash/debt bridge without proving future target supply or mature acquired cash margins. No original investor forecast is invented. The live terminal valuation is independently reconstructed from finite future owner cash and capital costs. Sources support its operating anchors, not the chosen required returns or advantage durations. ## Valuation summary Quote CAD 26.36, market time 2026-11-04T20:59:59.000Z; valuation date 5 November 2026. Fixed original exits:26 October 2031/2036. Current reported anchors, normalized earning power and forward estimates remain separate. | Horizon | Bear | Base | Bull | Weighted cash-flow IRR | | --- | ---: | ---: | ---: | ---: | | 5 years | -3.18% | 18.40% | 32.91% | 17.57% | | 10 years | 1.68% | 14.42% | 20.61% | 13.60% | The reported September bridge and existing staged organic/target margins remain. Every scenario now shares the fixed CAD730.073m annual seller-supply ceiling; cash does not establish more sellers. Retained cash remains separate and earns the specified capped debt-saving/deposit return. Subsequent annual target purchases are funded without new borrowing. The base’s later budgets stop at the same seller limit as upside. ## Pre-mortem The main failure is mistaking a plausible mechanism for an evidenced quantity. Integration costs and customer attrition could persist, reported cash could include timing benefits, debt may refinance at worse terms and acquisition supply could be insufficient at the stated prices. A controlling shareholder and subsidiary restrictions can separate consolidated cash from cash accessible to common holders. The adverse scenario represents a specific failure path, not a maximum loss or a frequency estimate. The independent advocates’ strongest surviving attacks are recorded in 12-debate.md. ## Exit multiple Each forecast year has its own terminal owner-cash multiple. The adverse/base/favorable tails run 5/10/15 additional years at 11%/9.5%/8.5% required owner returns. They continue the scenario’s operating margins and modest organic/acquired growth, charge each purchase once, cap annual purchase capital by seller supply and owner cash, then assume no further growth or acquisitions. Current retained cash is valued separately once; no second bank-debt deduction is made from after-financing equity cash. Base multiples are 17.277× at the five-year exit and 14.250× at the ten-year exit: a fixed dollar opportunity is worth less per earnings dollar as the portfolio grows. Duration and discount rates are low-confidence judgments. The annual revenue/cash/capital ledger and 0/5/10/15-year duration sensitivities are inspectable in the model-change record; local sensitivities hold this researched tail fixed. ## Bear case Customer attrition, lower target cash margins, expensive purchases and a short profitable opportunity period jointly reduce owner value. The current low-growth/low-margin path and five-year finite tail express this mechanism. Tax timing is not made into a permanent benefit, no new facility is inferred from covenant compliance, and uncapped earnouts/access to subsidiary cash remain unresolved. Zero common recovery and financing failure can still occur outside the representative paths. ## Variant view The potential opportunity is buying a recurring software cash base plus disciplined acquisitions at an attractive price. Q3’s stronger cash bridge helps, but the apparent discount depends on target integration, accessible funding and sustained attractive purchases. The debate rejects an unlimited central deal supply and a peer-table terminal shortcut. A18.40% five-year base remains conditional; the longer-horizon base falls to 14.42% as finite seller supply limits compounding relative to a larger portfolio. Neither total revenue growth nor an external investor’s return table establishes that mispricing. ## Why the opportunity exists The following liquidity/holder observations retain their original 3 November measurement window; they are context, not a new float or short-interest survey. Evidence, not adjectives: - Market capitalization about C$6.7bn (256.6m shares × C$26.15; US$4.8bn). Free float is small: CSU holds 61.4%, and WideOrbit's rollover holders received about 13.5% at conversion. Excluding CSU, about 99m shares; excluding both, about 64m. - Daily trading over the three months to 3 November: median about 127,000 shares, about C$2.9m (lumine-price-history-2026-11-03). Days to turn over the float: about 780 trading days excluding CSU, about 510 excluding CSU and the WideOrbit holders (three years and two years). - Listed on the TSX Venture Exchange, not the TSX main board, so not eligible for the S&P/TSX Composite. Sell-side coverage: at least one broker hosts management (TD Cowen conference, 24 November, lumine-events); coverage depth and short interest were not verified. - Recent events that drove sellers: the 2025-2026 sell-off in software on fears of AI disruption, from about C$54 (July 2025) to C$19-20 (April 2026); Synchronoss bought at a 70% premium with concentrated customers; H1 2026 FCFA2S down 29% (75.8 against 107.4, working capital and strengthening costs); no buyback or dividend to support the price. Neglect or justified discount? Partly neglect (venture exchange, thin float, a controlled company), partly justified (modest organic growth, demanding integration, concentration). The illiquidity is real: an outside investor cannot build or exit a large position quickly. ## What would prove us wrong | Observation | Where and when | Refutes | | --- | --- | --- | | Maintenance and recurring organic growth below −2% for two quarters | Q3 and Q4 2026 MD&A organic tables (Q3 published 4 November; next report date unconfirmed) | the intact recurring base; moves weight to the downside | | Synchronoss cash margin: group operating margin back above 33% by H1 2027, or still near 30% | FY2026 and H1 2027 MD&A operating income and FCFA2S | the funded-cohort convergence estimate | | The >10% customer's revenue | FY2026 MD&A customer concentration note | the stability of the Motive cohort | | Deal pace and pricing: acquisitions over the next 12 months below 150 or entry prices above 1.5x revenue | quarterly acquisition notes | the ratio-based deal budget and staged purchased margins | | A special dividend or buyback | issuer releases | the model's retain-and-repay assumption (Lumine's FY2025 behaviour); a dividend would not lower value but would end the interest saving | | CSU sells or proposes to buy out minorities | CSU and Lumine filings | the controlling-holder alignment | The deal-pace and owner-cash checks are frozen in `valuation/prospective-checks.json` (3 November, both model vintages; the calendar-2028 owner-cash check tests the deal-yield choice); the organic-growth, Synchronoss-margin and customer observations are review triggers here and in `research/11-scenario-probabilities.md`, not frozen predictions, because the model makes no numerical forecast of them. Permanent-loss paths and unverified inputs are in `research/06-risks.md`. The [independent debate](12-debate.md) links each surviving attack to the relevant reporting event. Earlier frozen predictions remain unchanged; new debate-vintage forecasts below are stored separately rather than backfitted to realized outcomes. [Same-price/date returns, sensitivities and claim judgments](../updates/2026-11-05-debate-review.json); [annual model-change arithmetic](../updates/2026-11-05-debate-model-change.json). # Risks and disconfirming evidence Assessment: 5 November 2026. The independent debate sizes the risks below against current scenarios. These are business and financing mechanisms, not forecasts of exact failures or a worst-case loss bound. ## Surviving attacks Apply the same 730.073 CADm annual seller limit to every scenario; cash is funding, not target supply. No25%arbitrary haircut or new borrowing. Capped later purchases leave unplaced cash visible. Evidence: lumine-fy2022-mda through lumine-fy2025-mda; lumine-q3-2026-fs pp.12–15; named supplyCeilingCad/internalDeployRatioSep26. Replace live peer binding with yearly terminal multiple paths. Central5/10-year multiples are 17.277x/14.250x under the fixed-dollar tail. Preserve peer arithmetic as comparison and source calibration; bridge/sensitivity/model sheet read actual dated multiples. Evidence: named peerExitMultiple (preserved reference); updates/2026-11-05-debate-model-change.json finite annual revenue/cash/capital ledger. No additional normalization, immediate target cash or arbitrary margin cut. Keep current staged recovery and explicit customer/refinancing/earnout/tax risks. Those survive the debate, constrain conviction and are already represented in adverse paths; they require future evidence. Evidence: lumine-q3-2026-fs pp.7,12–15,18–25; lumine-q3-2026-mda pp.9–17. ## Evidence gaps and conviction Moderate in the recurring-software acquisition mechanism, low in precise target margins, accessible funding and postexit duration. Q3 accounts strengthen the starting bridge but do not prove future target supply or ten-year returns. Reassess after coherent capped scenarios and independent terminal definitions; retain existing coarse horizon-specific weights because stronger Q3 cash and persistent integration/customer/refinancing risks do not establish revised numerical frequencies. No odds fitted to returns. ## What would change the view Next reported quarter and FY2026 accounts (date not confirmed): acquired cash margins, organic recurring sales, cash taxes/debt, target deal prices and accessible liquidity; 24November technology conference is not an earnings report. [Full bull case, bear case and rebuttals](12-debate.md); [same-price/date financial effects](../updates/2026-11-05-debate-review.json). Existing frozen predictions are preserved, with the new debate vintage dated separately. # Thesis-to-model register Current assessment:5 November 2026. The source calibration is preserved separately. Reported facts, normalization estimates and new forecast judgments keep their own evidence labels. ## Starting financial bridge | Model input | Value | Classification | Economic definition and evidence | | --- | ---: | --- | --- | | equityCashEarnings | 341.902 | normalized estimate | September trailing reported FCFA2S, signed working-capital normalization, 6% on current gross principal, identified transaction expenses and cash interest removed. Cash taxes, leases, sustaining capex and development stay charged. Current-tax timing is disclosed separately; target profit recovery is not reported cash. Sources: lumine-q2-2026-fs, lumine-q2-2026-mda, lumine-fy2025-fs, lumine-fy2025-mda, boc-usdcad-2026-10-26, lumine-q3-2026-fs, lumine-q3-2026-mda | | shares | 256.620 | reported | 256.620389m shares = 256.620388m subordinate voting shares plus the single super-voting share, which has the same economic rights. June 2026 note 10. Preferred/special securities converted in March 2024; no extra outstanding preferred claim is deducted. Sources: lumine-q3-2026-fs, lumine-q3-2026-mda | | openingCash | 283.866 | pro forma estimate | Use reported September cash 202.400. July acquisitions, acquired cash and Q3 debt repayments are already included. Restricted cash is separately classified in other assets; do not add it or redraw facilities. Remaining acquisition claims are charged separately. Sources: lumine-q3-2026-fs | | revenue | 1331.879 | reported trailing revenue | FY2025 +9M2026−9M2025 reported consolidated revenue, translated consistently. Funded missing annual sales are additional named revenue contributions; no already consolidated quarter is added twice. Sources: lumine-fy2025-fs, lumine-q2-2026-fs, boc-usdcad-2026-10-26, lumine-q3-2026-fs | ## Scenario assumptions The reported September bridge and existing staged organic/target margins remain. Every scenario now shares the fixed CAD730.073m annual seller-supply ceiling; cash does not establish more sellers. Retained cash remains separate and earns the specified capped debt-saving/deposit return. Subsequent annual target purchases are funded without new borrowing. The base’s later budgets stop at the same seller limit as upside. Each forecast year has its own terminal owner-cash multiple. The adverse/base/favorable tails run 5/10/15 additional years at 11%/9.5%/8.5% required owner returns. They continue the scenario’s operating margins and modest organic/acquired growth, charge each purchase once, cap annual purchase capital by seller supply and owner cash, then assume no further growth or acquisitions. Current retained cash is valued separately once; no second bank-debt deduction is made from after-financing equity cash. Base multiples are 17.277× at the five-year exit and 14.250× at the ten-year exit: a fixed dollar opportunity is worth less per earnings dollar as the portfolio grows. Duration and discount rates are low-confidence judgments. The annual revenue/cash/capital ledger and 0/5/10/15-year duration sensitivities are inspectable in the model-change record; local sensitivities hold this researched tail fixed. ### base Revenue build (3 November). Organic revenue grows 1% in year 1, 2% in years 2-5 and 1.5% after. Its owner-cash margin recovers from 26.6% to 27.8% as 2026 strengthening costs roll off. New deals buy revenue at 1.6x (2026 pricing) and then at the 1.55x record, and converge from 22% to the 30% group margin by year 4. The deal budget is 81.9 a year in years 1-2 (debt-repayment years) and then the 131.7 FY2021-FY2025 average (USD; inputs.yearly.deploymentCapacity); unplaced cash repays debt, then earns a deposit rate. Exit at 17.7x (Topicus trailing, converted). Purchase economics are financing-neutral; owner cash retains the legacy debt cost. Terminal value uses finite postexit cash and capped annual purchases, with the researched year-specific multiple path; peer multiples are reference evidence only. | Driver | Value/path | Reason and sources | | --- | --- | --- | | payoutRatio | 0.000 | Zero reflects no declared dividend and management’s stated goal to reinvest FCFA2S. Operating FCF is not a shareholder receipt. Apply observed 1% organic growth to after-investment baseline cash, fund new deals with cash already on hand and future owner cash, and assume no additional borrowing. Sources: lumine-q2-2026-fs, lumine-q2-2026-mda | | dilutionRate | 0.000 | Zero additional dilution: issued shares were unchanged in FY2025 and H1 2026; no preferred conversion is still outstanding. Future issuance or compensation policy can change this assumption. Apply observed 1% organic growth to after-investment baseline cash, fund new deals with cash already on hand and future owner cash, and assume no additional borrowing. Sources: lumine-q2-2026-fs, lumine-q2-2026-mda | | reinvestmentRate | 1.000 | Base/upside 100% follows the stated FCFA2S reinvestment objective. The 50% stress retains cash for liquidity; retained cash adds to terminal equity once. Apply observed 1% organic growth to after-investment baseline cash, fund new deals with cash already on hand and future owner cash, and assume no additional borrowing. Sources: lumine-q2-2026-fs, lumine-q2-2026-mda | | organicGrowth, years 1–1 | 0.010 | TTM June 2026 organic growth was +1% reported (0% FX-adjusted). Maintenance and recurring revenue grew +1% FX-adjusted in Q2 2026, while professional services fell 13% and licences 1%. Synchronoss is not yet in the organic base, and the >10% customer (USD106.3m in FY2025) shows no decline. Year 1 repeats the latest observed rate. Sources: lumine-q2-2026-mda, lumine-q2-2026-fs, lumine-fy2025-mda, lumine-fy2025-fs, lumine-q3-2026-mda | | organicGrowth, years 2–5 | 0.020 | FY2025 organic growth was +2% reported. Maintenance and recurring (76% of H1 2026 revenue) grew +2-3% reported in FY2024, FY2025 and TTM June 2026, and licence and services volumes have no further trend once the FY2024 slump (licences −36%) is in the base. From February 2027 Synchronoss joins the organic base: its standalone revenue grew 5.7% in FY2024 on cloud subscriber growth, against WideOrbit's linear-advertising decline. 0.76 × 2.5% plus flat project revenue gives about 2%. Sources: lumine-fy2025-mda, lumine-fy2025-fs, lumine-fy2024-mda, sncr-fy2024-10k, lumine-q3-2026-mda | | organicGrowth, years 6–10 | 0.015 | Fades by half a point. Linear TV and radio advertising (the WideOrbit cohort, about a third of FY2025 revenue) keeps shrinking, and platform migrations (cloud playout, new core networks) replace some products at renewal. Lumine's record has had no year above +2% since 2021. Sources: lumine-fy2022-mda, lumine-fy2023-mda, lumine-fy2024-mda, lumine-fy2025-mda, lumine-fy2025-fs, lumine-q3-2026-mda | | organicMargin, years 1–1 | 0.262 | September normalized trailing cash margin is about 25.7%; recover about half a point in the first year while the newly acquired lower-margin businesses are strengthening. No immediate return to the former 27% assumption. Sources: lumine-q3-2026-fs, lumine-q3-2026-mda, lumine-fy2025-mda | | organicMargin, years 2–2 | 0.270 | Staged integration and removal of identified transaction friction; retain operating/product costs. A forecast judgment, not issuer guidance or an achieved target cash margin. Sources: lumine-q3-2026-fs, lumine-q3-2026-mda, lumine-fy2025-mda | | organicMargin, years 3–10 | 0.278 | Retain the prior long-run base normalization below the historical financing-neutral margin after debt cost; do not extrapolate quarterly recovery indefinitely. Sources: lumine-q3-2026-fs, lumine-q3-2026-mda, lumine-fy2025-mda | | priceToSales, years 1–1 | 1.600 | The 2026 deals were priced like Synchronoss: 277.5 all-in for 173.6 of revenue, 1.60x (sncrPriceToSales), a public take-private at a 70% premium. The July deals' revenue is undisclosed. Sources: lumine-q2-2026-mda, lumine-q2-2026-fs, sncr-fy2024-10k, csi-synchronoss-release-2025 | | priceToSales, years 2–10 | 1.550 | Lumine's FY2022-FY2025 record: 823.0 of matched capital, including the Motive working-capital build, bought 525.7 of FY2025 revenue, 1.57x (recPriceToSales). By deal: WideOrbit 2.09x on its cohort's prior-year revenue, Motive about 1.0-1.1x including working capital, Datafusion 0.65x, Vidispine about 0.45x, Synchronoss 1.60x. Sources: lumine-fy2022-mda, lumine-fy2023-mda, lumine-fy2024-mda, lumine-fy2025-mda, lumine-fy2025-fs, lumine-motive-announcement-2023 | | acquiredMargin, years 1–1 | 0.220 | Deals in their first contributing year earn below group margin. Synchronoss earned a 16.4% unlevered cash margin standalone (sncrStandaloneMargin) and was still loss-making after amortization in H1 2026. Strengthening costs lifted H1 2026 expenses to 70% of revenue. Applies to all acquired cohorts in that calendar year. Sources: lumine-q2-2026-mda, lumine-q2-2026-fs, sncr-fy2024-10k, lumine-q3-2026-mda | | acquiredMargin, years 2–3 | 0.260 | Partway to group margin. Group unlevered cash rose 62.7 in the first full year after Motive (recRoiic2025), and operating margin stepped from 31.5% to 36.0% in FY2025, so most of the strengthening comes in the second and third year. Sources: lumine-fy2024-mda, lumine-fy2025-mda, lumine-fy2025-fs, lumine-q3-2026-mda | | acquiredMargin, years 4–10 | 0.300 | Acquired businesses reach about the group's financing-neutral margin: FY2022-FY2025 deals earned 31.2% of their revenue by FY2025 even at the low bound (recAcquiredMarginLow), against the 29.2-31.7% group range. With a 1.55x price this is a 19.4% yield, in line with Lumine's 19.9% low-bound record yield (lumineRecordMultiple). Sources: lumine-fy2022-mda, lumine-fy2023-mda, lumine-fy2024-mda, lumine-fy2025-mda, lumine-fy2025-fs, lumine-q3-2026-mda | | acquiredGrowth, years 1–10 | 0.000 | Acquired revenue held flat. Lumine's reported organic growth, which includes acquired businesses after their first year, averaged −1.25% a year over FY2023-TTM June 2026 (0%, −8%, +2%, +1%), weighed down by WideOrbit. The record deal yield already embeds early post-deal trends. Sources: lumine-fy2023-mda, lumine-fy2024-mda, lumine-fy2025-mda, lumine-fy2025-fs, lumine-q2-2026-mda, lumine-q2-2026-fs | | deploymentCapacity, years 1–2 | 114.795 | Debt repayment competes for cash. H1 2026 drew 380 of bank debt for Synchronoss and the July deals, which are already funded cohorts and not counted again. After its earlier debt-funded deals Lumine deployed 140.6 (FY2024) and 23.1 (FY2025, while repaying 125.7), an average of USD81.9m (deployAfterDebtAvg). 5 November debate: the same fixed seller-supply ceiling applies in every scenario; excess cash remains retained. Sources: lumine-fy2022-mda, lumine-fy2023-mda, lumine-fy2024-mda, lumine-fy2025-mda, lumine-q2-2026-fs, lumine-q2-2026-mda | | deploymentCapacity, years 3–3 | 398.195 | 4 November Q3 update: 0.6845 × model year-3 owner cash (581.712 CAD m). Stated budget 398.195 CAD m, solved from the shared forecast without using an IRR target. The ratio is internally funded purchases on a matched January2022–September2026 cash/financing window. Actual purchases also remain limited by cash after claims. 5 November debate: the same fixed seller-supply ceiling applies in every scenario; excess cash remains retained. Sources: lumine-fy2022-mda, lumine-fy2023-mda, lumine-fy2024-mda, lumine-fy2025-mda, lumine-q2-2026-fs, lumine-q2-2026-mda, lumine-q3-2026-fs, lumine-q3-2026-mda | | deploymentCapacity, years 4–4 | 469.931 | 4 November Q3 update: 0.6845 × model year-4 owner cash (686.510 CAD m). Stated budget 469.931 CAD m, solved from the shared forecast without using an IRR target. The ratio is internally funded purchases on a matched January2022–September2026 cash/financing window. Actual purchases also remain limited by cash after claims. 5 November debate: the same fixed seller-supply ceiling applies in every scenario; excess cash remains retained. Sources: lumine-fy2022-mda, lumine-fy2023-mda, lumine-fy2024-mda, lumine-fy2025-mda, lumine-q2-2026-fs, lumine-q2-2026-mda, lumine-q3-2026-fs, lumine-q3-2026-mda | | deploymentCapacity, years 5–5 | 544.187 | 4 November Q3 update: 0.6845 × model year-5 owner cash (794.988 CAD m). Stated budget 544.187 CAD m, solved from the shared forecast without using an IRR target. The ratio is internally funded purchases on a matched January2022–September2026 cash/financing window. Actual purchases also remain limited by cash after claims. 5 November debate: the same fixed seller-supply ceiling applies in every scenario; excess cash remains retained. Sources: lumine-fy2022-mda, lumine-fy2023-mda, lumine-fy2024-mda, lumine-fy2025-mda, lumine-q2-2026-fs, lumine-q2-2026-mda, lumine-q3-2026-fs, lumine-q3-2026-mda | | deploymentCapacity, years 6–6 | 627.889 | 4 November Q3 update: 0.6845 × model year-6 owner cash (917.266 CAD m). Stated budget 627.889 CAD m, solved from the shared forecast without using an IRR target. The ratio is internally funded purchases on a matched January2022–September2026 cash/financing window. Actual purchases also remain limited by cash after claims. 5 November debate: the same fixed seller-supply ceiling applies in every scenario; excess cash remains retained. Sources: lumine-fy2022-mda, lumine-fy2023-mda, lumine-fy2024-mda, lumine-fy2025-mda, lumine-q2-2026-fs, lumine-q2-2026-mda, lumine-q3-2026-fs, lumine-q3-2026-mda | | deploymentCapacity, years 7–7 | 723.847 | 4 November Q3 update: 0.6845 × model year-7 owner cash (1057.449 CAD m). Stated budget 723.847 CAD m, solved from the shared forecast without using an IRR target. The ratio is internally funded purchases on a matched January2022–September2026 cash/financing window. Actual purchases also remain limited by cash after claims. 5 November debate: the same fixed seller-supply ceiling applies in every scenario; excess cash remains retained. Sources: lumine-fy2022-mda, lumine-fy2023-mda, lumine-fy2024-mda, lumine-fy2025-mda, lumine-q2-2026-fs, lumine-q2-2026-mda, lumine-q3-2026-fs, lumine-q3-2026-mda | | deploymentCapacity, years 8–8 | 730.073 | 4 November Q3 update: 0.6845 × model year-8 owner cash (1218.147 CAD m). Stated budget 833.848 CAD m, solved from the shared forecast without using an IRR target. The ratio is internally funded purchases on a matched January2022–September2026 cash/financing window. Actual purchases also remain limited by cash after claims. 5 November debate: the same fixed seller-supply ceiling applies in every scenario; excess cash remains retained. Sources: lumine-fy2022-mda, lumine-fy2023-mda, lumine-fy2024-mda, lumine-fy2025-mda, lumine-q2-2026-fs, lumine-q2-2026-mda, lumine-q3-2026-fs, lumine-q3-2026-mda | | deploymentCapacity, years 9–9 | 730.073 | 4 November Q3 update: 0.6845 × model year-9 owner cash (1402.347 CAD m). Stated budget 959.938 CAD m, solved from the shared forecast without using an IRR target. The ratio is internally funded purchases on a matched January2022–September2026 cash/financing window. Actual purchases also remain limited by cash after claims. 5 November debate: the same fixed seller-supply ceiling applies in every scenario; excess cash remains retained. Sources: lumine-fy2022-mda, lumine-fy2023-mda, lumine-fy2024-mda, lumine-fy2025-mda, lumine-q2-2026-fs, lumine-q2-2026-mda, lumine-q3-2026-fs, lumine-q3-2026-mda | | deploymentCapacity, years 10–10 | 730.073 | 4 November Q3 update: 0.6845 × model year-10 owner cash (1613.475 CAD m). Stated budget 1104.460 CAD m, solved from the shared forecast without using an IRR target. The ratio is internally funded purchases on a matched January2022–September2026 cash/financing window. Actual purchases also remain limited by cash after claims. 5 November debate: the same fixed seller-supply ceiling applies in every scenario; excess cash remains retained. Sources: lumine-fy2022-mda, lumine-fy2023-mda, lumine-fy2024-mda, lumine-fy2025-mda, lumine-q2-2026-fs, lumine-q2-2026-mda, lumine-q3-2026-fs, lumine-q3-2026-mda | | exitMultiple, years 1–1 | 20.036 | Finite10-year postexit acquisition/organic opportunity at 9.5%required return; fixed seller ceiling and explicit capital costs. Scale changes the multiple by year. The duration and required return are subjective judgments. Full annual ledger is in the 5 November debate model-change record. Sources: lumine-q3-2026-fs, lumine-q3-2026-mda, lumine-fy2025-fs | | exitMultiple, years 2–2 | 19.701 | Finite10-year postexit acquisition/organic opportunity at 9.5%required return; fixed seller ceiling and explicit capital costs. Scale changes the multiple by year. The duration and required return are subjective judgments. Full annual ledger is in the 5 November debate model-change record. Sources: lumine-q3-2026-fs, lumine-q3-2026-mda, lumine-fy2025-fs | | exitMultiple, years 3–3 | 18.859 | Finite10-year postexit acquisition/organic opportunity at 9.5%required return; fixed seller ceiling and explicit capital costs. Scale changes the multiple by year. The duration and required return are subjective judgments. Full annual ledger is in the 5 November debate model-change record. Sources: lumine-q3-2026-fs, lumine-q3-2026-mda, lumine-fy2025-fs | | exitMultiple, years 4–4 | 18.048 | Finite10-year postexit acquisition/organic opportunity at 9.5%required return; fixed seller ceiling and explicit capital costs. Scale changes the multiple by year. The duration and required return are subjective judgments. Full annual ledger is in the 5 November debate model-change record. Sources: lumine-q3-2026-fs, lumine-q3-2026-mda, lumine-fy2025-fs | | exitMultiple, years 5–5 | 17.277 | Finite10-year postexit acquisition/organic opportunity at 9.5%required return; fixed seller ceiling and explicit capital costs. Scale changes the multiple by year. The duration and required return are subjective judgments. Full annual ledger is in the 5 November debate model-change record. Sources: lumine-q3-2026-fs, lumine-q3-2026-mda, lumine-fy2025-fs | | exitMultiple, years 6–6 | 16.512 | Finite10-year postexit acquisition/organic opportunity at 9.5%required return; fixed seller ceiling and explicit capital costs. Scale changes the multiple by year. The duration and required return are subjective judgments. Full annual ledger is in the 5 November debate model-change record. Sources: lumine-q3-2026-fs, lumine-q3-2026-mda, lumine-fy2025-fs | | exitMultiple, years 7–7 | 15.747 | Finite10-year postexit acquisition/organic opportunity at 9.5%required return; fixed seller ceiling and explicit capital costs. Scale changes the multiple by year. The duration and required return are subjective judgments. Full annual ledger is in the 5 November debate model-change record. Sources: lumine-q3-2026-fs, lumine-q3-2026-mda, lumine-fy2025-fs | | exitMultiple, years 8–8 | 15.129 | Finite10-year postexit acquisition/organic opportunity at 9.5%required return; fixed seller ceiling and explicit capital costs. Scale changes the multiple by year. The duration and required return are subjective judgments. Full annual ledger is in the 5 November debate model-change record. Sources: lumine-q3-2026-fs, lumine-q3-2026-mda, lumine-fy2025-fs | | exitMultiple, years 9–9 | 14.643 | Finite10-year postexit acquisition/organic opportunity at 9.5%required return; fixed seller ceiling and explicit capital costs. Scale changes the multiple by year. The duration and required return are subjective judgments. Full annual ledger is in the 5 November debate model-change record. Sources: lumine-q3-2026-fs, lumine-q3-2026-mda, lumine-fy2025-fs | | exitMultiple, years 10–10 | 14.250 | Finite10-year postexit acquisition/organic opportunity at 9.5%required return; fixed seller ceiling and explicit capital costs. Scale changes the multiple by year. The duration and required return are subjective judgments. Full annual ledger is in the 5 November debate model-change record. Sources: lumine-q3-2026-fs, lumine-q3-2026-mda, lumine-fy2025-fs | ### downside An explicit stress, not issuer guidance: 2% annual baseline cash erosion, 50% deployment to preserve liquidity, purchases costing 75% more per unit of cash earnings and a 35% lower terminal multiple. This tests weaker retention, integration and financing sentiment; it is not a financing-default simulation. Purchase economics are financing-neutral with respect to legacy group debt; owner cash retains that debt cost. Terminal value uses finite postexit cash and capped annual purchases, with the researched year-specific multiple path; peer multiples are reference evidence only. | Driver | Value/path | Reason and sources | | --- | --- | --- | | payoutRatio | 0.000 | Zero reflects no declared dividend and management’s stated goal to reinvest FCFA2S. Operating FCF is not a shareholder receipt. An explicit stress, not issuer guidance: 2% annual baseline cash erosion, 50% deployment to preserve liquidity, purchases costing 75% more per unit of cash earnings and a 35% lower terminal multiple. This tests weaker retention, integration and financing sentiment; it is not a financing-default simulation. Sources: lumine-q2-2026-fs, lumine-q2-2026-mda | | dilutionRate | 0.000 | Zero additional dilution: issued shares were unchanged in FY2025 and H1 2026; no preferred conversion is still outstanding. Future issuance or compensation policy can change this assumption. An explicit stress, not issuer guidance: 2% annual baseline cash erosion, 50% deployment to preserve liquidity, purchases costing 75% more per unit of cash earnings and a 35% lower terminal multiple. This tests weaker retention, integration and financing sentiment; it is not a financing-default simulation. Sources: lumine-q2-2026-fs, lumine-q2-2026-mda | | reinvestmentRate | 0.500 | Base/upside 100% follows the stated FCFA2S reinvestment objective. The 50% stress retains cash for liquidity; retained cash adds to terminal equity once. An explicit stress, not issuer guidance: 2% annual baseline cash erosion, 50% deployment to preserve liquidity, purchases costing 75% more per unit of cash earnings and a 35% lower terminal multiple. This tests weaker retention, integration and financing sentiment; it is not a financing-default simulation. Sources: lumine-q2-2026-fs, lumine-q2-2026-mda | | organicGrowth, years 1–2 | -0.040 | Repeats half of FY2024's −8% organic decline, when licences fell 36%, services 18% and hardware 46%. It adds a partial loss or repricing at the >10% customer (13.9% of FY2025 revenue) and Synchronoss's carriers (top five 97.6% of its revenue) renewing at lower prices. Sources: lumine-fy2024-mda, lumine-fy2025-mda, lumine-fy2025-fs, sncr-fy2024-10k, lumine-q3-2026-mda | | organicGrowth, years 3–10 | -0.010 | After the project slump, maintenance erodes slowly as the linear-advertising cohort declines and some customers migrate. No year since 2023 has shown maintenance below −1%. Sources: lumine-fy2024-mda, lumine-fy2025-mda, lumine-fy2025-fs, lumine-q2-2026-mda, lumine-q2-2026-fs, lumine-q3-2026-mda | | organicMargin, years 1–10 | 0.255 | The H1 2026 dip persists, plus one point of pricing pressure from concentrated customers renewing (the >10% customer; Verizon and AT&T at Synchronoss). Sources: lumine-q2-2026-mda, lumine-q2-2026-fs, lumine-fy2025-mda, lumine-fy2025-fs, sncr-fy2024-10k | | priceToSales, years 1–10 | 2.000 | Deals priced like WideOrbit (2.09x, the highest in the record) or like public take-privates at premiums, as competition for comms and media software rises. Sources: lumine-fy2023-mda, csi-synchronoss-release-2025 | | acquiredMargin, years 1–3 | 0.170 | Synchronoss-like deals stay near their standalone margin (16.4%) because concentrated carrier customers capture the savings. Sources: sncr-fy2024-10k, lumine-q2-2026-mda, lumine-q2-2026-fs, lumine-q3-2026-mda | | acquiredMargin, years 4–10 | 0.200 | Partial strengthening only. At a 2.0x price this is a 10% yield, about the cost of capital and close to the earlier family-aggregate stress (8.6%). Sources: lumine-fy2025-mda, lumine-fy2025-fs, lumine-q3-2026-mda | | acquiredGrowth, years 1–10 | -0.020 | Acquired businesses decline like WideOrbit's licence base in FY2024. Sources: lumine-fy2024-mda | | deploymentCapacity, years 1–2 | 23.001 | A two-year dry spell at FY2021's 16.4 (the record's quietest year), longer than the record's longest quiet run of one year, while 2026-vintage prices stay high and the 2026 debt is repaid. 5 November debate: the same fixed seller-supply ceiling applies in every scenario; excess cash remains retained. Sources: lumine-fy2022-mda, lumine-fy2023-mda, lumine-fy2024-mda, lumine-fy2025-mda, lumine-q2-2026-fs, lumine-q2-2026-mda | | deploymentCapacity, years 3–3 | 105.567 | 4 November Q3 update: 0.2397 × model year-3 owner cash (440.327 CAD m). Stated budget 105.567 CAD m, solved from the shared forecast without using an IRR target. The ratio is the lower internally funded FY2022–FY2025 observation. Actual purchases also remain limited by cash after claims. 5 November debate: the same fixed seller-supply ceiling applies in every scenario; excess cash remains retained. Sources: lumine-fy2022-mda, lumine-fy2023-mda, lumine-fy2024-mda, lumine-fy2025-mda, lumine-q2-2026-fs, lumine-q2-2026-mda, lumine-q3-2026-fs, lumine-q3-2026-mda | | deploymentCapacity, years 4–4 | 108.613 | 4 November Q3 update: 0.2397 × model year-4 owner cash (453.031 CAD m). Stated budget 108.613 CAD m, solved from the shared forecast without using an IRR target. The ratio is the lower internally funded FY2022–FY2025 observation. Actual purchases also remain limited by cash after claims. 5 November debate: the same fixed seller-supply ceiling applies in every scenario; excess cash remains retained. Sources: lumine-fy2022-mda, lumine-fy2023-mda, lumine-fy2024-mda, lumine-fy2025-mda, lumine-q2-2026-fs, lumine-q2-2026-mda, lumine-q3-2026-fs, lumine-q3-2026-mda | | deploymentCapacity, years 5–5 | 111.357 | 4 November Q3 update: 0.2397 × model year-5 owner cash (464.479 CAD m). Stated budget 111.357 CAD m, solved from the shared forecast without using an IRR target. The ratio is the lower internally funded FY2022–FY2025 observation. Actual purchases also remain limited by cash after claims. 5 November debate: the same fixed seller-supply ceiling applies in every scenario; excess cash remains retained. Sources: lumine-fy2022-mda, lumine-fy2023-mda, lumine-fy2024-mda, lumine-fy2025-mda, lumine-q2-2026-fs, lumine-q2-2026-mda, lumine-q3-2026-fs, lumine-q3-2026-mda | | deploymentCapacity, years 6–6 | 114.148 | 4 November Q3 update: 0.2397 × model year-6 owner cash (476.119 CAD m). Stated budget 114.148 CAD m, solved from the shared forecast without using an IRR target. The ratio is the lower internally funded FY2022–FY2025 observation. Actual purchases also remain limited by cash after claims. 5 November debate: the same fixed seller-supply ceiling applies in every scenario; excess cash remains retained. Sources: lumine-fy2022-mda, lumine-fy2023-mda, lumine-fy2024-mda, lumine-fy2025-mda, lumine-q2-2026-fs, lumine-q2-2026-mda, lumine-q3-2026-fs, lumine-q3-2026-mda | | deploymentCapacity, years 7–7 | 116.986 | 4 November Q3 update: 0.2397 × model year-7 owner cash (487.957 CAD m). Stated budget 116.986 CAD m, solved from the shared forecast without using an IRR target. The ratio is the lower internally funded FY2022–FY2025 observation. Actual purchases also remain limited by cash after claims. 5 November debate: the same fixed seller-supply ceiling applies in every scenario; excess cash remains retained. Sources: lumine-fy2022-mda, lumine-fy2023-mda, lumine-fy2024-mda, lumine-fy2025-mda, lumine-q2-2026-fs, lumine-q2-2026-mda, lumine-q3-2026-fs, lumine-q3-2026-mda | | deploymentCapacity, years 8–8 | 119.872 | 4 November Q3 update: 0.2397 × model year-8 owner cash (499.997 CAD m). Stated budget 119.872 CAD m, solved from the shared forecast without using an IRR target. The ratio is the lower internally funded FY2022–FY2025 observation. Actual purchases also remain limited by cash after claims. 5 November debate: the same fixed seller-supply ceiling applies in every scenario; excess cash remains retained. Sources: lumine-fy2022-mda, lumine-fy2023-mda, lumine-fy2024-mda, lumine-fy2025-mda, lumine-q2-2026-fs, lumine-q2-2026-mda, lumine-q3-2026-fs, lumine-q3-2026-mda | | deploymentCapacity, years 9–9 | 122.809 | 4 November Q3 update: 0.2397 × model year-9 owner cash (512.244 CAD m). Stated budget 122.809 CAD m, solved from the shared forecast without using an IRR target. The ratio is the lower internally funded FY2022–FY2025 observation. Actual purchases also remain limited by cash after claims. 5 November debate: the same fixed seller-supply ceiling applies in every scenario; excess cash remains retained. Sources: lumine-fy2022-mda, lumine-fy2023-mda, lumine-fy2024-mda, lumine-fy2025-mda, lumine-q2-2026-fs, lumine-q2-2026-mda, lumine-q3-2026-fs, lumine-q3-2026-mda | | deploymentCapacity, years 10–10 | 125.796 | 4 November Q3 update: 0.2397 × model year-10 owner cash (524.703 CAD m). Stated budget 125.796 CAD m, solved from the shared forecast without using an IRR target. The ratio is the lower internally funded FY2022–FY2025 observation. Actual purchases also remain limited by cash after claims. 5 November debate: the same fixed seller-supply ceiling applies in every scenario; excess cash remains retained. Sources: lumine-fy2022-mda, lumine-fy2023-mda, lumine-fy2024-mda, lumine-fy2025-mda, lumine-q2-2026-fs, lumine-q2-2026-mda, lumine-q3-2026-fs, lumine-q3-2026-mda | | exitMultiple, years 1–1 | 8.577 | Finite5-year postexit acquisition/organic opportunity at 11.0%required return; fixed seller ceiling and explicit capital costs. Scale changes the multiple by year. The duration and required return are subjective judgments. Full annual ledger is in the 5 November debate model-change record. Sources: lumine-q3-2026-fs, lumine-q3-2026-mda, lumine-fy2025-fs | | exitMultiple, years 2–2 | 8.573 | Finite5-year postexit acquisition/organic opportunity at 11.0%required return; fixed seller ceiling and explicit capital costs. Scale changes the multiple by year. The duration and required return are subjective judgments. Full annual ledger is in the 5 November debate model-change record. Sources: lumine-q3-2026-fs, lumine-q3-2026-mda, lumine-fy2025-fs | | exitMultiple, years 3–3 | 8.566 | Finite5-year postexit acquisition/organic opportunity at 11.0%required return; fixed seller ceiling and explicit capital costs. Scale changes the multiple by year. The duration and required return are subjective judgments. Full annual ledger is in the 5 November debate model-change record. Sources: lumine-q3-2026-fs, lumine-q3-2026-mda, lumine-fy2025-fs | | exitMultiple, years 4–4 | 8.558 | Finite5-year postexit acquisition/organic opportunity at 11.0%required return; fixed seller ceiling and explicit capital costs. Scale changes the multiple by year. The duration and required return are subjective judgments. Full annual ledger is in the 5 November debate model-change record. Sources: lumine-q3-2026-fs, lumine-q3-2026-mda, lumine-fy2025-fs | | exitMultiple, years 5–5 | 8.551 | Finite5-year postexit acquisition/organic opportunity at 11.0%required return; fixed seller ceiling and explicit capital costs. Scale changes the multiple by year. The duration and required return are subjective judgments. Full annual ledger is in the 5 November debate model-change record. Sources: lumine-q3-2026-fs, lumine-q3-2026-mda, lumine-fy2025-fs | | exitMultiple, years 6–6 | 8.544 | Finite5-year postexit acquisition/organic opportunity at 11.0%required return; fixed seller ceiling and explicit capital costs. Scale changes the multiple by year. The duration and required return are subjective judgments. Full annual ledger is in the 5 November debate model-change record. Sources: lumine-q3-2026-fs, lumine-q3-2026-mda, lumine-fy2025-fs | | exitMultiple, years 7–7 | 8.537 | Finite5-year postexit acquisition/organic opportunity at 11.0%required return; fixed seller ceiling and explicit capital costs. Scale changes the multiple by year. The duration and required return are subjective judgments. Full annual ledger is in the 5 November debate model-change record. Sources: lumine-q3-2026-fs, lumine-q3-2026-mda, lumine-fy2025-fs | | exitMultiple, years 8–8 | 8.531 | Finite5-year postexit acquisition/organic opportunity at 11.0%required return; fixed seller ceiling and explicit capital costs. Scale changes the multiple by year. The duration and required return are subjective judgments. Full annual ledger is in the 5 November debate model-change record. Sources: lumine-q3-2026-fs, lumine-q3-2026-mda, lumine-fy2025-fs | | exitMultiple, years 9–9 | 8.525 | Finite5-year postexit acquisition/organic opportunity at 11.0%required return; fixed seller ceiling and explicit capital costs. Scale changes the multiple by year. The duration and required return are subjective judgments. Full annual ledger is in the 5 November debate model-change record. Sources: lumine-q3-2026-fs, lumine-q3-2026-mda, lumine-fy2025-fs | | exitMultiple, years 10–10 | 8.519 | Finite5-year postexit acquisition/organic opportunity at 11.0%required return; fixed seller ceiling and explicit capital costs. Scale changes the multiple by year. The duration and required return are subjective judgments. Full annual ledger is in the 5 November debate model-change record. Sources: lumine-q3-2026-fs, lumine-q3-2026-mda, lumine-fy2025-fs | ### upside An explicit favorable case, not issuer guidance: 3% baseline cash growth, purchases 15% cheaper per unit of cash earnings and a 20% higher exit multiple. Full reinvestment still requires funded purchases and the declared timing policy. Purchase economics are financing-neutral with respect to legacy group debt; owner cash retains that debt cost. Terminal value uses finite postexit cash and capped annual purchases, with the researched year-specific multiple path; peer multiples are reference evidence only. | Driver | Value/path | Reason and sources | | --- | --- | --- | | payoutRatio | 0.000 | Zero reflects no declared dividend and management’s stated goal to reinvest FCFA2S. Operating FCF is not a shareholder receipt. An explicit favorable case, not issuer guidance: 3% baseline cash growth, purchases 15% cheaper per unit of cash earnings and a 20% higher exit multiple. Full reinvestment still requires funded purchases and the declared timing policy. Sources: lumine-q2-2026-fs, lumine-q2-2026-mda | | dilutionRate | 0.000 | Zero additional dilution: issued shares were unchanged in FY2025 and H1 2026; no preferred conversion is still outstanding. Future issuance or compensation policy can change this assumption. An explicit favorable case, not issuer guidance: 3% baseline cash growth, purchases 15% cheaper per unit of cash earnings and a 20% higher exit multiple. Full reinvestment still requires funded purchases and the declared timing policy. Sources: lumine-q2-2026-fs, lumine-q2-2026-mda | | reinvestmentRate | 1.000 | Base/upside 100% follows the stated FCFA2S reinvestment objective. The 50% stress retains cash for liquidity; retained cash adds to terminal equity once. An explicit favorable case, not issuer guidance: 3% baseline cash growth, purchases 15% cheaper per unit of cash earnings and a 20% higher exit multiple. Full reinvestment still requires funded purchases and the declared timing policy. Sources: lumine-q2-2026-fs, lumine-q2-2026-mda | | organicGrowth, years 1–1 | 0.020 | Maintenance at the FY2025 rate (+2-3%), with licences and services recovering from the H1 2026 dip (services −13% in Q2). Sources: lumine-fy2025-mda, lumine-fy2025-fs, lumine-q2-2026-mda, lumine-q2-2026-fs, lumine-q3-2026-mda | | organicGrowth, years 2–10 | 0.030 | Maintenance +3% (the TTM June 2026 reported rate) plus Synchronoss's subscriber-driven growth (5.7% in FY2024) and an upgrade cycle in broadcast IP and cloud playout, where Imagine and Quortex compete. Sources: lumine-q2-2026-mda, lumine-q2-2026-fs, sncr-fy2024-10k, imagine-elta-2026, lumine-q3-2026-mda | | organicMargin, years 1–1 | 0.275 | Strengthening costs roll off faster than in the base. Sources: lumine-q2-2026-mda, lumine-q2-2026-fs | | organicMargin, years 2–10 | 0.290 | Returns to the FY2025 financing-neutral margin less the current interest drag: 31.7% − 2.6% ≈ 29%. Sources: lumine-fy2025-mda, lumine-fy2025-fs | | priceToSales, years 1–1 | 1.600 | At the upside's larger deal scale the deals are large ones, priced like Synchronoss (1.60x all-in to FY2024 revenue, sncrPriceToSales), the 2026 benchmark. Sources: sncr-fy2024-10k, lumine-q2-2026-fs | | priceToSales, years 2–10 | 1.550 | A budget near the USD520m supply ceiling means large deals, not small carve-outs. Lumine's large deals averaged about 1.58x revenue: WideOrbit 2.09x, Synchronoss 1.60x, Motive about 1.05x including working capital. That is the 1.57x record (recPriceToSales). Small deals at 0.45-0.65x (Vidispine, Datafusion) can't fill this budget, so the upside gets no price advantage over the base; its edge is faster strengthening (acquired margin). Sources: lumine-fy2022-mda, lumine-fy2023-mda, lumine-fy2024-mda, lumine-fy2025-mda, lumine-q2-2026-fs, lumine-q2-2026-mda | | acquiredMargin, years 1–1 | 0.240 | Faster strengthening of the 2026 deals. Sources: lumine-q2-2026-mda, lumine-q2-2026-fs, lumine-q3-2026-mda | | acquiredMargin, years 2–3 | 0.280 | As base, slightly faster. Sources: lumine-fy2025-mda, lumine-fy2025-fs, lumine-q3-2026-mda | | acquiredMargin, years 4–10 | 0.310 | The FY2025 record level for acquired revenue (31.2%, recAcquiredMarginLow) and the group's financing-neutral margin (31.7%). At 1.3x this is a 23.8% yield, near the record midpoint (23.3%). Sources: lumine-fy2025-mda, lumine-fy2025-fs, lumine-q3-2026-mda | | acquiredGrowth, years 1–3 | 0.000 | Flat during strengthening, as in the base. Sources: lumine-fy2025-mda, lumine-fy2025-fs | | acquiredGrowth, years 4–10 | 0.010 | Once strengthened, acquired businesses grow with maintenance (+1-3%). Sources: lumine-fy2025-mda, lumine-fy2025-fs, lumine-q2-2026-mda, lumine-q2-2026-fs | | deploymentCapacity, years 1–1 | 673.126 | 4 November Q3 update: 1.3963 × model year-1 owner cash (482.071 CAD m), bounded by the fixed 730.073 CAD m deal-supply ceiling. Stated budget 673.126 CAD m, solved from the shared forecast without using an IRR target. The ratio is the historical FY2022–FY2025 all-in ratio; it does not authorize new borrowing. Actual purchases also remain limited by cash after claims. 5 November debate: the same fixed seller-supply ceiling applies in every scenario; excess cash remains retained. Sources: lumine-fy2022-mda, lumine-fy2023-mda, lumine-fy2024-mda, lumine-fy2025-mda, lumine-q2-2026-fs, lumine-q2-2026-mda, lumine-q3-2026-fs, lumine-q3-2026-mda | | deploymentCapacity, years 2–2 | 730.073 | 4 November Q3 update: 1.3963 × model year-2 owner cash (680.890 CAD m), bounded by the fixed 730.073 CAD m deal-supply ceiling. Stated budget 730.073 CAD m, solved from the shared forecast without using an IRR target. The ratio is the historical FY2022–FY2025 all-in ratio; it does not authorize new borrowing. Actual purchases also remain limited by cash after claims. 5 November debate: the same fixed seller-supply ceiling applies in every scenario; excess cash remains retained. Sources: lumine-fy2022-mda, lumine-fy2023-mda, lumine-fy2024-mda, lumine-fy2025-mda, lumine-q2-2026-fs, lumine-q2-2026-mda, lumine-q3-2026-fs, lumine-q3-2026-mda | | deploymentCapacity, years 3–3 | 730.073 | 4 November Q3 update: 1.3963 × model year-3 owner cash (820.981 CAD m), bounded by the fixed 730.073 CAD m deal-supply ceiling. Stated budget 730.073 CAD m, solved from the shared forecast without using an IRR target. The ratio is the historical FY2022–FY2025 all-in ratio; it does not authorize new borrowing. Actual purchases also remain limited by cash after claims. 5 November debate: the same fixed seller-supply ceiling applies in every scenario; excess cash remains retained. Sources: lumine-fy2022-mda, lumine-fy2023-mda, lumine-fy2024-mda, lumine-fy2025-mda, lumine-q2-2026-fs, lumine-q2-2026-mda, lumine-q3-2026-fs, lumine-q3-2026-mda | | deploymentCapacity, years 4–4 | 730.073 | 4 November Q3 update: 1.3963 × model year-4 owner cash (1019.147 CAD m), bounded by the fixed 730.073 CAD m deal-supply ceiling. Stated budget 730.073 CAD m, solved from the shared forecast without using an IRR target. The ratio is the historical FY2022–FY2025 all-in ratio; it does not authorize new borrowing. Actual purchases also remain limited by cash after claims. 5 November debate: the same fixed seller-supply ceiling applies in every scenario; excess cash remains retained. Sources: lumine-fy2022-mda, lumine-fy2023-mda, lumine-fy2024-mda, lumine-fy2025-mda, lumine-q2-2026-fs, lumine-q2-2026-mda, lumine-q3-2026-fs, lumine-q3-2026-mda | | deploymentCapacity, years 5–5 | 730.073 | 4 November Q3 update: 1.3963 × model year-5 owner cash (1194.554 CAD m), bounded by the fixed 730.073 CAD m deal-supply ceiling. Stated budget 730.073 CAD m, solved from the shared forecast without using an IRR target. The ratio is the historical FY2022–FY2025 all-in ratio; it does not authorize new borrowing. Actual purchases also remain limited by cash after claims. 5 November debate: the same fixed seller-supply ceiling applies in every scenario; excess cash remains retained. Sources: lumine-fy2022-mda, lumine-fy2023-mda, lumine-fy2024-mda, lumine-fy2025-mda, lumine-q2-2026-fs, lumine-q2-2026-mda, lumine-q3-2026-fs, lumine-q3-2026-mda | | deploymentCapacity, years 6–6 | 730.073 | 4 November Q3 update: 1.3963 × model year-6 owner cash (1369.559 CAD m), bounded by the fixed 730.073 CAD m deal-supply ceiling. Stated budget 730.073 CAD m, solved from the shared forecast without using an IRR target. The ratio is the historical FY2022–FY2025 all-in ratio; it does not authorize new borrowing. Actual purchases also remain limited by cash after claims. 5 November debate: the same fixed seller-supply ceiling applies in every scenario; excess cash remains retained. Sources: lumine-fy2022-mda, lumine-fy2023-mda, lumine-fy2024-mda, lumine-fy2025-mda, lumine-q2-2026-fs, lumine-q2-2026-mda, lumine-q3-2026-fs, lumine-q3-2026-mda | | deploymentCapacity, years 7–7 | 730.073 | 4 November Q3 update: 1.3963 × model year-7 owner cash (1543.554 CAD m), bounded by the fixed 730.073 CAD m deal-supply ceiling. Stated budget 730.073 CAD m, solved from the shared forecast without using an IRR target. The ratio is the historical FY2022–FY2025 all-in ratio; it does not authorize new borrowing. Actual purchases also remain limited by cash after claims. 5 November debate: the same fixed seller-supply ceiling applies in every scenario; excess cash remains retained. Sources: lumine-fy2022-mda, lumine-fy2023-mda, lumine-fy2024-mda, lumine-fy2025-mda, lumine-q2-2026-fs, lumine-q2-2026-mda, lumine-q3-2026-fs, lumine-q3-2026-mda | | deploymentCapacity, years 8–8 | 730.073 | 4 November Q3 update: 1.3963 × model year-8 owner cash (1721.421 CAD m), bounded by the fixed 730.073 CAD m deal-supply ceiling. Stated budget 730.073 CAD m, solved from the shared forecast without using an IRR target. The ratio is the historical FY2022–FY2025 all-in ratio; it does not authorize new borrowing. Actual purchases also remain limited by cash after claims. 5 November debate: the same fixed seller-supply ceiling applies in every scenario; excess cash remains retained. Sources: lumine-fy2022-mda, lumine-fy2023-mda, lumine-fy2024-mda, lumine-fy2025-mda, lumine-q2-2026-fs, lumine-q2-2026-mda, lumine-q3-2026-fs, lumine-q3-2026-mda | | deploymentCapacity, years 9–9 | 730.073 | 4 November Q3 update: 1.3963 × model year-9 owner cash (1903.230 CAD m), bounded by the fixed 730.073 CAD m deal-supply ceiling. Stated budget 730.073 CAD m, solved from the shared forecast without using an IRR target. The ratio is the historical FY2022–FY2025 all-in ratio; it does not authorize new borrowing. Actual purchases also remain limited by cash after claims. 5 November debate: the same fixed seller-supply ceiling applies in every scenario; excess cash remains retained. Sources: lumine-fy2022-mda, lumine-fy2023-mda, lumine-fy2024-mda, lumine-fy2025-mda, lumine-q2-2026-fs, lumine-q2-2026-mda, lumine-q3-2026-fs, lumine-q3-2026-mda | | deploymentCapacity, years 10–10 | 730.073 | 4 November Q3 update: 1.3963 × model year-10 owner cash (2089.053 CAD m), bounded by the fixed 730.073 CAD m deal-supply ceiling. Stated budget 730.073 CAD m, solved from the shared forecast without using an IRR target. The ratio is the historical FY2022–FY2025 all-in ratio; it does not authorize new borrowing. Actual purchases also remain limited by cash after claims. 5 November debate: the same fixed seller-supply ceiling applies in every scenario; excess cash remains retained. Sources: lumine-fy2022-mda, lumine-fy2023-mda, lumine-fy2024-mda, lumine-fy2025-mda, lumine-q2-2026-fs, lumine-q2-2026-mda, lumine-q3-2026-fs, lumine-q3-2026-mda | | exitMultiple, years 1–1 | 31.125 | Finite15-year postexit acquisition/organic opportunity at 8.5%required return; fixed seller ceiling and explicit capital costs. Scale changes the multiple by year. The duration and required return are subjective judgments. Full annual ledger is in the 5 November debate model-change record. Sources: lumine-q3-2026-fs, lumine-q3-2026-mda, lumine-fy2025-fs | | exitMultiple, years 2–2 | 27.438 | Finite15-year postexit acquisition/organic opportunity at 8.5%required return; fixed seller ceiling and explicit capital costs. Scale changes the multiple by year. The duration and required return are subjective judgments. Full annual ledger is in the 5 November debate model-change record. Sources: lumine-q3-2026-fs, lumine-q3-2026-mda, lumine-fy2025-fs | | exitMultiple, years 3–3 | 24.911 | Finite15-year postexit acquisition/organic opportunity at 8.5%required return; fixed seller ceiling and explicit capital costs. Scale changes the multiple by year. The duration and required return are subjective judgments. Full annual ledger is in the 5 November debate model-change record. Sources: lumine-q3-2026-fs, lumine-q3-2026-mda, lumine-fy2025-fs | | exitMultiple, years 4–4 | 22.868 | Finite15-year postexit acquisition/organic opportunity at 8.5%required return; fixed seller ceiling and explicit capital costs. Scale changes the multiple by year. The duration and required return are subjective judgments. Full annual ledger is in the 5 November debate model-change record. Sources: lumine-q3-2026-fs, lumine-q3-2026-mda, lumine-fy2025-fs | | exitMultiple, years 5–5 | 21.591 | Finite15-year postexit acquisition/organic opportunity at 8.5%required return; fixed seller ceiling and explicit capital costs. Scale changes the multiple by year. The duration and required return are subjective judgments. Full annual ledger is in the 5 November debate model-change record. Sources: lumine-q3-2026-fs, lumine-q3-2026-mda, lumine-fy2025-fs | | exitMultiple, years 6–6 | 20.600 | Finite15-year postexit acquisition/organic opportunity at 8.5%required return; fixed seller ceiling and explicit capital costs. Scale changes the multiple by year. The duration and required return are subjective judgments. Full annual ledger is in the 5 November debate model-change record. Sources: lumine-q3-2026-fs, lumine-q3-2026-mda, lumine-fy2025-fs | | exitMultiple, years 7–7 | 19.809 | Finite15-year postexit acquisition/organic opportunity at 8.5%required return; fixed seller ceiling and explicit capital costs. Scale changes the multiple by year. The duration and required return are subjective judgments. Full annual ledger is in the 5 November debate model-change record. Sources: lumine-q3-2026-fs, lumine-q3-2026-mda, lumine-fy2025-fs | | exitMultiple, years 8–8 | 19.162 | Finite15-year postexit acquisition/organic opportunity at 8.5%required return; fixed seller ceiling and explicit capital costs. Scale changes the multiple by year. The duration and required return are subjective judgments. Full annual ledger is in the 5 November debate model-change record. Sources: lumine-q3-2026-fs, lumine-q3-2026-mda, lumine-fy2025-fs | | exitMultiple, years 9–9 | 18.624 | Finite15-year postexit acquisition/organic opportunity at 8.5%required return; fixed seller ceiling and explicit capital costs. Scale changes the multiple by year. The duration and required return are subjective judgments. Full annual ledger is in the 5 November debate model-change record. Sources: lumine-q3-2026-fs, lumine-q3-2026-mda, lumine-fy2025-fs | | exitMultiple, years 10–10 | 18.169 | Finite15-year postexit acquisition/organic opportunity at 8.5%required return; fixed seller ceiling and explicit capital costs. Scale changes the multiple by year. The duration and required return are subjective judgments. Full annual ledger is in the 5 November debate model-change record. Sources: lumine-q3-2026-fs, lumine-q3-2026-mda, lumine-fy2025-fs | ## Evidence that would change the inputs Next reported quarter and FY2026 accounts (date not confirmed): acquired cash margins, organic recurring sales, cash taxes/debt, target deal prices and accessible liquidity; 24 November technology conference is not an earnings report. Reassess after coherent capped scenarios and independent terminal definitions; retain existing coarse horizon-specific weights because stronger Q3 cash and persistent integration/customer/refinancing risks do not establish revised numerical frequencies. No odds fitted to returns. [Detailed claim-by-claim judgment](12-debate.md). All shared-engine calculations retain stable fact IDs and audit links; original source arithmetic is not a current forecast. # Underwriting plan 4 November 2026, September accounts. USD millions unless stated; model currency CAD, fixed translation 1.4025. Economic decisions below were recorded before comparing revised IRRs. [Dated evidence and attribution](../updates/2026-11-04-daily-review.md). ## Opportunity ledger | Opportunity | Current evidence | Reachable change and cost | Limits and contrary evidence | | --- | --- | --- | --- | | Recurring renewals and pricing | Q3 recurring organic growth 6%; total organic constant-currency growth 3%, trailing 1% | Modest base growth: 1% next year, 2% years 2–5, 1.5% later; continuing support/R&D remains charged | No portfolio retention or renewal-price series; customers can consolidate vendors or change technology. A strong recurring quarter is not a ten-year growth rate. | | Licences and services | Q3 organic FX-adjusted growth −14% and −10% | Base growth remains modest; upgrades require implementation staff and customer budgets | New platforms can replace existing products; no extrapolation of 55% reported acquisition-led growth. | | Synchronoss already owned | Q3 revenue 43.137, owned YTD 107.446 | Flat-quarter annualization leaves 65.102 missing sales, estimated cash at the financing-neutral group margin; cash recovery ramps over one year | Target cash margin not disclosed; concentrated carriers have renewal leverage. Q3 net income includes acquisition accounting and is not owner cash. | | July businesses already owned | Quortex/Imagine first full owned-quarter sales 21.511/35.611 | Three missing quarters add 171.366 revenue; cash converges separately over one year | Imagine lost 6.826; Quortex profit includes a 4.904 bargain gain. No observed mature cash margin. Do not buy these businesses twice. | | Future acquisitions | Measured through-cycle deal record and matched funding observations | Base first two annual budgets stay USD81.85m; later budgets 0.6845 × annual owner cash, stated as dollar paths. Downside starts with the two small-budget years and later uses 0.2397; upside uses 1.3963 but cannot exceed USD520.551m supply or available cash | Ratios are observations, not targets or seller commitments. The recent peak is only an upside supply ceiling. All-in prices, timing and strengthening costs matter. | | Unplaced cash | Q3 debt repayments 59.375; no common dividend | Retain cash; debt-saving return capped at gross principal 425.625, measured deposit yield above | No extra borrowing assumed. Subsidiary restrictions, liquidity needs, debt maturity and a future dividend/buyback can alter allocation. | | Customer budget risk | Amdocs 3 November report: lost/disputed T-Mobile work, discretionary cuts, weaker pro-forma outlook | Watch Lumine's maintenance, concentration and contract evidence | This is an independent counterparty warning, not proof Lumine lost a customer. Peer perimeter changes cannot become Lumine organic decline. | ## Scenario mechanisms Current5 November 2026 judgment: The reported September bridge and existing staged organic/target margins remain. Every scenario now shares the fixed CAD730.073m annual seller-supply ceiling; cash does not establish more sellers. Retained cash remains separate and earns the specified capped debt-saving/deposit return. Subsequent annual target purchases are funded without new borrowing. The base’s later budgets stop at the same seller limit as upside. Customer attrition, lower target cash margins, expensive purchases and a short profitable opportunity period jointly reduce owner value. The current low-growth/low-margin path and five-year finite tail express this mechanism. Tax timing is not made into a permanent benefit, no new facility is inferred from covenant compliance, and uncapped earnouts/access to subsidiary cash remain unresolved. Zero common recovery and financing failure can still occur outside the representative paths. The forward opportunity ledger determines the operating path before return comparisons. It is not a projection of past CAGR or management targets. Capital, timing and ownership claims remain matched to the revenue/cash perimeter. Reassess after coherent capped scenarios and independent terminal definitions; retain existing coarse horizon-specific weights because stronger Q3 cash and persistent integration/customer/refinancing risks do not establish revised numerical frequencies. No odds fitted to returns. ## Current absolute scale and terminal opportunities 5-year base: revenue 2669.164, owner cash 794.988, exit cash earnings 813.709 (CADm). 10-year base: revenue 5006.865, owner cash 1515.798, exit cash earnings 1530.533 (CADm). Each forecast year has its own terminal owner-cash multiple. The adverse/base/favorable tails run 5/10/15 additional years at 11%/9.5%/8.5% required owner returns. They continue the scenario’s operating margins and modest organic/acquired growth, charge each purchase once, cap annual purchase capital by seller supply and owner cash, then assume no further growth or acquisitions. Current retained cash is valued separately once; no second bank-debt deduction is made from after-financing equity cash. Base multiples are 17.277× at the five-year exit and 14.250× at the ten-year exit: a fixed dollar opportunity is worth less per earnings dollar as the portfolio grows. Duration and discount rates are low-confidence judgments. The annual revenue/cash/capital ledger and 0/5/10/15-year duration sensitivities are inspectable in the model-change record; local sensitivities hold this researched tail fixed. No return target determines this path. Source locators and disagreements are in 12-debate.md and 07-assumption-register.md. # Capital productivity and reinvestment Assessment:5 November 2026. Values below come directly from the same shared forecast used by the dashboard. Money is in CAD millions. | Horizon | Scenario | Capital-return definition | Return/proxy | Period-weighted reinvestment | Growth capital | Matching earnings | Funding residual | | --- | --- | --- | ---: | ---: | ---: | ---: | ---: | | 5 years | downside | Acquisition ROIIC · mature cash proxy | 9.39% | 17.07% | 371.539 | 2176.335 | 2.27e-13 | | 5 years | base | Acquisition ROIIC · mature cash proxy | 18.16% | 54.80% | 1641.902 | 2996.242 | 6.82e-13 | | 5 years | upside | Acquisition ROIIC · mature cash proxy | 18.28% | 84.67% | 3554.196 | 4197.642 | 4.55e-13 | | 10 years | downside | Acquisition ROIIC · mature cash proxy | 9.77% | 20.76% | 971.150 | 4677.354 | 5.68e-13 | | 10 years | base | Acquisition ROIIC · mature cash proxy | 18.98% | 57.24% | 5183.857 | 9055.914 | 9.09e-13 | | 10 years | upside | Acquisition ROIIC · mature cash proxy | 19.15% | 56.18% | 7204.559 | 12824.457 | 0.00e+0 | The acquisition return is purchased mature owner cash / new acquisition capital, not measured historical cohort ROIIC or a discounted project IRR. Previously funded businesses are not bought again. Ongoing product/development investment remains expensed and unquantified as growth capital, so acquisition reinvestment is not total economic reinvestment. Unused cash and debt-saving/deposit returns are separate from bought operating earnings. Reinvestment equals total new growth investment divided by total matching earnings over the period, not the average of annual percentages. Cash generation less distributions/old claims plus opening cash drawn and net borrowing reconciles to new investment. No extra borrowing or invented working-capital release is introduced. All six funding ledgers reconcile within 0.000001 model millions. Sources and definitions:04-capital-allocation.md, valuation/calculations.json and the shared ledger; docs/method/cash-flow-and-reinvestment.md. [Dated review](../updates/2026-11-05-debate-review.md). # Scenario probabilities Assessment:5 November 2026, after the independent debate. Low-confidence subjective joint-path weights; no claim of empirical calibration or source-author odds. | Horizon | Bear | Base | Bull | | --- | ---: | ---: | ---: | | 5 years | 30.00% | 50.00% | 20.00% | | 10 years | 35.00% | 40.00% | 25.00% | Reassess after coherent capped scenarios and independent terminal definitions; retain existing coarse horizon-specific weights because stronger Q3 cash and persistent integration/customer/refinancing risks do not establish revised numerical frequencies. No odds fitted to returns. The reported September bridge and existing staged organic/target margins remain. Every scenario now shares the fixed CAD730.073m annual seller-supply ceiling; cash does not establish more sellers. Retained cash remains separate and earns the specified capped debt-saving/deposit return. Subsequent annual target purchases are funded without new borrowing. The base’s later budgets stop at the same seller limit as upside. Customer attrition, lower target cash margins, expensive purchases and a short profitable opportunity period jointly reduce owner value. The current low-growth/low-margin path and five-year finite tail express this mechanism. Tax timing is not made into a permanent benefit, no new facility is inferred from covenant compliance, and uncapped earnouts/access to subsidiary cash remain unresolved. Zero common recovery and financing failure can still occur outside the representative paths. Moderate in the recurring-software acquisition mechanism, low in precise target margins, accessible funding and postexit duration. Q3 accounts strengthen the starting bridge but do not prove future target supply or ten-year returns. Review on Next reported quarter and FY2026 accounts (date not confirmed): acquired cash margins, organic recurring sales, cash taxes/debt, target deal prices and accessible liquidity; 24 November technology conference is not an earnings report. Weights sum to 100% independently for each horizon. These paths represent outcome regions and omit permanent-loss tails. Price alone does not change business-path odds; local scenario overrides suspend researched weights. Weighted IRR is calculated from probability-weighted dated shareholder receipts, not by averaging scenario IRRs. [Evidence and rebuttals](12-debate.md); method:docs/method/scenario-probabilities.md. # Bull–bear review — Lumine Group, 5 November 2026 Independent advocates prepared opening cases without seeing the other draft, then exchanged point-by-point cross-examinations. A separate judge who did not author the prior model assessed the competing claims. Source reconstructions retain their stated provenance; inaccessible private reports were not newly read. Quoted sensitivities test arguments and are not automatically adopted forecasts. Fixed quote CAD 26.36, market time 2026-11-04T20:59:59.000Z; valuation date 2026-11-05. Original forecast exits retained. | Horizon | Before bear/base/bull | After bear/base/bull | Weighted before → after | | --- | --- | --- | --- | | 5 years | 0.82% / 18.96% / 32.54% | -3.18% / 18.40% / 32.91% | 18.34% → 17.57% | | 10 years | 3.37% / 17.26% / 22.30% | 1.68% / 14.42% / 20.61% | 15.76% → 13.60% | ## Bull case As of 5 November 2026. No opposing advocate draft was read. Existing company research, fact/calculation ledgers and exact registered source locators were examined; original issuer sources additionally checked live: Lumine Q3, CSU Q2, Diploma Q3 transcript, NVIDIA CFO commentary, TerraVest Highland/KBK, WOSG FY26. Harvia Q3 PDF retrieval failed; its registered extraction remains the evidence limit. Private source interpretations below are the repository's documented reconstruction, not a new claim to have read an unavailable private original. The strongest bear case is a collection of legacy products earning cash while customer concentration, integration losses and financed acquisition growth hide deterioration. Cash and reported profit are not interchangeable, and the short acquisition record is dominated by a few deals. **LMN-B1 — The recurring base has not broken.** Independently checked the 4 November issuer release: Q3 organic growth 3% versus 55% total; FCFA2S53.9m, up 27%. The MD&A recurring organic 6% versus shrinking licences/services supports the existing modest upside organic path, not acquisition-led55% extrapolation. Sources `lumine-q3-2026-release`, Q3 headlines and revenue discussion; `lumine-q3-2026-mda`, pp.4–8. Test upside `organicGrowth` only; maintain financing and target margins. **LMN-B2 — Owned sales can mature without another purchase.** September reported cash 202.4 and gross debt 425.625 replace stale June closing estimates. Observed missing consolidated sales are 65.102 Synchronoss and 171.366 July businesses; estimated cash still ramps. Test upside `organicMargin` only, holding all owned-cohort ramps unchanged. Sources `lumine-q3-2026-fs`, pp.7,12–15,18–20; `lumine-q3-2026-mda`, pp.9–11. Imagine losses and Quortex bargain gain do not establish target cash. Base staged recovery already captures much of this bull mechanism; no second July purchase or full-year revenue double count is permitted. **LMN-B3 — Idle cash creates deployment optionality.** Test upside annual deployment budgets, retaining base purchase prices and margins; actual cash remains a constraint. The USD520.551m ceiling bounds seller supply; first-two-year base USD81.85m budgets intentionally model quiet years. Historical FY2022–25 all-in ratio 1.3963 is a possibility, not permission to borrow. Sources `lumine-fy2025-fs`, acquisition/cash-flow notes; `lumine-q3-2026-fs`, acquisitions pp.11–15 and liquidity/debt pp.18–20; named deployment calculations. This is the largest model upside, therefore the strongest uncertainty. **LMN-B4 — The exit premium needs proof beyond peers.** Test21.3× versus 17.73585×. Base17.74× requires~14 post-exit years at 10% growth, 19.35% marginal return and 9.5% required return. Specialised installed systems and carve-out sourcing are the claimed advantage; linear-media erosion and acquisition competition attack it. Peer cash-P/E definitions remain unresolved. This supports neither a higher central multiple nor high conviction. ### Isolated shared-engine effects Saved quote 26.36; market timestamp 2026-11-04T20:59:59.000Z. Fixed valuation date 2026-11-05; model 3f48908d71af; exits remain the original model vintage. These are sensitivities, not joint forecasts or probability revisions. Baseline base IRR 18.96% / 17.26% (5/10 years). | Claim | Exact input/path | 5-year base | 10-year base | Change, pp (5/10) | | --- | --- | ---: | ---: | --- | | LMN-B1 | `cases.base.inputs.yearly.organicGrowth` | 19.57% | 17.71% | +0.62 / +0.45 | | LMN-B2 | `cases.base.inputs.yearly.organicMargin` | 19.56% | 17.45% | +0.60 / +0.19 | | LMN-B3 | `cases.base.inputs.yearly.deploymentCapacity` | 26.01% | 19.10% | +7.05 / +1.84 | | LMN-B4 | `cases.base.inputs.exitMultiple` | 23.00% | 19.21% | +4.04 / +1.95 | ## Bear case ### Strongest opposing case The recurring base grew 6% organically in Q3, purchased sales now have an observed full owned quarter, and USD59.375m of bank principal was repaid. If the acquired businesses approach the historical group margin while modestly priced bolt-ons continue, low organic growth can still produce attractive per-share compounding. This is a real cash mechanism, not a revenue-growth story. ### Bear case The favorable return depends on three separate estimates that the new report does not establish: working-capital normalization, acquired-cash convergence, and a long reinvestment runway. Nine-month FCFA2S is USD129.7m, down 14%; trailing reported cash 196.7m becomes normalized 243.8m, then 309.3m after estimated mature missing-month contributions. The missing sales are observed; their group-like cash margins are not. Imagine's losses and Quortex's bargain gain make accounting profit an especially weak cash proxy. USD8.895m more current tax was accrued than paid in nine months. Treating all of the September cash as deployable and repeatedly refinancing 2028–2030 bank maturities adds a liquidity assumption; covenants met today do not resolve those later funding choices. [lumine-q3-2026-fs pp.12–15,18–20,23–24; lumine-q3-2026-mda pp.5–11; named calculations ownerCashUsd, missing-month cohort bridges in valuation/calculations.json.] The outside view is mixed. The low-bound historical deal yield near 20% comes from a short, concentrated cohort. Synchronoss's estimated standalone cash yield was 10.3%, and five carriers supplied 97.6% of sales. A group margin cannot be transferred to a customer-concentrated acquisition without showing who retains the restructuring savings. [sncr-fy2024-10k customer concentration and cash statements; lumine-q2-2026-fs note 4; research/04-capital-allocation.md.] The17.7x exit is a converted Topicus multiple from an investor table whose cash definition remains unresolved. The company's own value-driver check supports about 16x at 5% growth, 18% incremental return and 9.5% required return, while the no-runway value is about 11x. No evidenced count of post-2031 excess-return years supports 17.7x. A strong earnings quarter does not close that gap. Existing acquisition budgets also rise to aboutUSD787m in year 10, beyond the USD520.6m fixed seller-supply ceiling imposed only on the upside; an allocation ratio is not a seller-supply forecast. ### Contested claims - LMN-B1 —17.7x exit lacks a matched denominator and explicit advantage duration. Test16x fundamental cross-check. No precise probability is inferred. - LMN-B2 —Future purchased cohorts may retain only 20% cash margins. Test cap 20% versus 22→30%; stress anchored to existing downside and 16.4% standalone Synchronoss, not a new observed group outcome. - LMN-B3 —Recovery of organic margin above September normalized 25.7% is unproved. Test flat current normalized margin; continuing product/support costs cannot disappear as sales scale. - LMN-B4 —Cash availability does not establish sellers. Test25% lower annual acquisition budgets. This is a bounded sensitivity, not evidence that precisely 25% of deals disappear. The Q3 issuer release was reopened for this review; the other cited evidence is the source-linked repository reconstruction. No private customer contracts or target cash accounts were obtained. No automatic numerical change is warranted merely because a stress lowers returns; the judge must establish the burden for the current extrapolation. ### Shared-engine effects Fixed saved quote CAD26.36; market timestamp 2026-11-04T20:59:59.000Z; valuation date 5 November 2026; fixed original exit calendar. Base before: 18.96% / 17.26% (5/10years). These are one-change sensitivities to the base, except named bundles; existing scenario probabilities are not revised or claimed as validated. | Claim | Change | 5-year base IRR | 10-year base IRR | 5-year effect | 10-year effect | | --- | --- | ---: | ---: | ---: | ---: | | LMN-B1 | Base exit 16x | 16.77% | 16.19% | -2.18pp | -1.07pp | | LMN-B2 | New acquired cash margin capped at 20% | 15.74% | 14.03% | -3.21pp | -3.23pp | | LMN-B3 | Organic margin stays at September normalized cash margin | 17.94% | 16.94% | -1.01pp | -0.32pp | | LMN-B4 | Base annual acquisition budgets 25% lower | 17.23% | 15.54% | -1.72pp | -1.72pp | ## Rebuttals ### Bull response 5 November 2026. Written after independently frozen openings were exchanged. Every numbered bear claim is addressed below; original bull openings and sensitivities remain unchanged. 1. **Bear terminal 17.7× challenge: concede missing independent derivation.** Peer definitions remain unresolved; the conversion from Topicus is not a cash-economic terminal argument. Independently propose post-exit owner-cash growth 8%:2% organic after continued proportional software costs plus 6% acquisition-led. At19.35% incremental acquisition cash yield this needs~31% of cash reinvested in new deals, no new borrowing. Use9.5% required return for concentration/control/illiquidity and ten additional profitable-deal years, then no new excess-return deals. Product longevity supports considering a decade, not claiming it is measured. This is materially less demanding than 14years at 10%; it does not independently support 17.7×. Do not choose duration by solving for the old multiple or a target shareholder return. The bear's16× perpetual check also needs finite-duration testing. 2. **Bear20% purchased cash margin: concede target convergence uncertainty; rebut transplanting one deal to the whole future population.** `lumine-q3-2026-fs`, pp12–15 provides owned sales and acquisition losses/gains; no standalone acquired cash accounts establish 30%. Historical near 20% entry cash yields use multiple cohorts, but only a few years and large deals. Carrier concentration is real (`sncr-fy2024-10k`). My organic-margin sensitivity does not change funded-cohort cash ramps or relabel sales as cash. Treat20% as an adverse path unless transaction population evidence replaces the current future-margin path. 3. **Bear flat normalized organic margin: partly concede.** Q3 recurring organic 6% rebuts a broken legacy portfolio claim (`lumine-q3-2026-mda`, pp4–8), not proof of acquired-carrier retention or margin. Base26.2→27→27.8% requires cost absorption; current financed profitability and prior group margins support conditional partial recovery, not certainty. Software spending stays charged in all cases. 4. **Bear25% lower budgets: concede the funding/supply distinction; reject arbitrary 25%.** More importantly, the base's year 10~USD787m breaches the same USD520.6m supply ceiling used for upside. No source identifies enough additional targets at unchanged yields to justify treating upside supply as fixed and base unlimited. Cap base consistently or document a separately evidenced expansion path. Keep cash returns/claims and unused cash visible; no hidden dividend or borrowing assumption. 5. **Liquidity, tax and missing owner-cash attacks survive.** Do not repeat working-capital releases, convert accrued tax into a permanent saving, or infer free cash from covenant compliance. Conviction moderate in the mechanism, lower in exact margin/runway/terminal claims. The Q3 release changes cash facts; it does not settle those judgments. ### Bear response 5 November 2026; written after frozen independent openings were exchanged. Claim prefixes in the openings are local to each advocate; author identity distinguishes them. 1. **Bull recurring-base claim: partly conceded.** Q3 recurring organic 6% and total 3% are evidence against a currently broken legacy portfolio (`lumine-q3-2026-mda`, pp.4–8). They are not acquired-carrier retention, customer-level margin or a guarantee of future migrations. Retain modest organic paths; do not infer 55% organic expansion. 2. **Bull owned-sales maturity: facts conceded, margin inference unresolved.** The missing ownership months and reported sales are correctly counted once. My concern is the group margin applied to them. Imagine's losses, Quortex's bargain gain and Synchronoss concentration do not establish target cash (`lumine-q3-2026-fs`, pp.12–15). Flat September margin is an adverse test, not an observed permanent normalization. Historical group margin supports a conditional partial recovery; I do not claim 20% must replace every future deal centrally. 3. **Bull idle-cash optionality: survives conditionally.** Actual cash can buy future earnings, but the same seller universe must constrain every scenario. Both advocates concede that base year 10 purchases~USD787m cannot coexist unexplained with the upside's fixedUSD520.551m supply ceiling. Apply that cap to base too unless a separately evidenced larger supply path replaces it. A blanket 25% budget cut was a sensitivity; I concede it is not uniquely justified. Capping rather than arbitrary haircut is the supported correction. 5. **Strongest surviving bear attacks.** Working-capital normalization, tax timing, accessible cash, target cash profitability and refinancing remain unresolved. Covenant compliance does not prove deployability or future funding. Do not turn these into fabricated quantitative liabilities; reflect them in adverse joint paths and lower conviction. Sources remain Q3 FS pp.18–20,23–24, MD&A pp.9–11, and the exact bridge in calculations. ## Judgment Existing estimates receive no presumption merely because they are already in the model. Positive company evidence, the relevant outside view and the capital/funding bridge determine each verdict; neither arbitrary conservatism nor an unsupported favorable sensitivity replaces it. The evidence and reconstruction below explain numerical changes or explicit no-change conclusions claim by claim. | Contested claim | Verdict | Evidence/locator | Reason and model action | | --- | --- | --- | --- | | bull advocate: Recurring maintenance and owned acquisition sales support continued cash compounding | adopt-modified | lumine-q3-2026-mda pp.4–11; lumine-q3-2026-fs pp.12–15 | Keep the Q3 trailing-period cash/revenue/claims bridge, observed missing owned sales and separate estimated target-margin ramp.55% total growth includes acquisitions; do not promote it to organic growth or target net profit to owner cash. | | bear advocate: Central annual acquisition budgets eventually exceed the seller ceiling applied to upside | adopt | lumine-fy2022-mda through lumine-fy2025-mda; lumine-q3-2026-fs pp.12–15; named supplyCeilingCad/internalDeployRatioSep26 | Apply the same 730.073 CADm annual seller limit to every scenario; cash is funding, not target supply. No25%arbitrary haircut or new borrowing. Capped later purchases leave unplaced cash visible. | | bull advocate: A finite ongoing acquisition opportunity supports a premium to steady state | adopt-modified | lumine-playbook; lumine-fy2025-mda; lumine-q3-2026-fs/mda; measured acquired margins/price-to-sales in calculations | Value actual annual operating cash and acquisition capital through finite 5/10/15 postexit years at 11/9.5/8.5%required returns in adverse/central/favorable paths. All tail purchases obey seller supply and owner cash, then no-growth value. These duration/discount estimates are explicit judgments; no peer-table multiplier or return target. | | bear advocate: A single 17.7x peer-linked multiple cannot fit both five- and ten-year scale under fixed seller supply | adopt | named peerExitMultiple (preserved reference); updates/2026-11-05-debate-model-change.json finite annual revenue/cash/capital ledger | Replace live peer binding with yearly terminal multiple paths. Central5/10-year multiples are 17.277x/14.250x under the fixed-dollar tail. Preserve peer arithmetic as comparison and source calibration; bridge/sensitivity/model sheet read actual dated multiples. | | bear advocate: Owned sales do not establish mature cash margins, tax benefits or freely deployable subsidiary cash | unresolved | lumine-q3-2026-fs pp.7,12–15,18–25; lumine-q3-2026-mda pp.9–17 | No additional normalization, immediate target cash or arbitrary margin cut. Keep current staged recovery and explicit customer/refinancing/earnout/tax risks. Those survive the debate, constrain conviction and are already represented in adverse paths; they require future evidence. | Reassess after coherent capped scenarios and independent terminal definitions; retain existing coarse horizon-specific weights because stronger Q3 cash and persistent integration/customer/refinancing risks do not establish revised numerical frequencies. No odds fitted to returns. ## Conviction Moderate in the recurring-software acquisition mechanism, low in precise target margins, accessible funding and postexit duration. Q3 accounts strengthen the starting bridge but do not prove future target supply or ten-year returns. These are forward estimates with surviving attacks, not confirmed outcomes. Neither numerical calibration, a completed checklist nor passing software tests establishes investment skill. Reported facts stay separate from forward judgments, and adverse paths are not worst-case bounds. **What would change the view:** Next reported quarter and FY2026 accounts (date not confirmed): acquired cash margins, organic recurring sales, cash taxes/debt, target deal prices and accessible liquidity; 24 November technology conference is not an earnings report. [Exact inputs, isolated sensitivities, model hashes and same-price/date before/after returns](../updates/2026-11-05-debate-review.json). ## Public Primary & Secondary Sources - `lumine-q2-2026-fs` (primary): Lumine Q2 2026 financial statements — https://www.luminegroup.com/wp-content/uploads/2026/08/Lumine-Group-FS-2026-Q2.pdf (published 2026-08-04) - `lumine-q2-2026-mda` (primary): Lumine Q2 2026 MD&A — https://www.luminegroup.com/wp-content/uploads/2026/08/Lumine-Group-MDA-2026-Q2.pdf (published 2026-08-04) - `lumine-fy2025-fs` (primary): Lumine FY2025 audited financial statements — https://www.luminegroup.com/wp-content/uploads/2026/03/Lumine-Group-FS-2025-YE-Final.pdf (published 2026-03-04) - `lumine-fy2025-mda` (primary): Lumine FY2025 MD&A — https://www.luminegroup.com/wp-content/uploads/2026/03/Lumine-Group-MDA-2025-YE-Final.pdf (published 2026-03-04) - `lumine-playbook` (primary): Lumine acquisition criteria and operating approach — https://www.luminegroup.com/acquisition-criteria/ - `boc-usdcad-2026-10-26` (primary): Bank of Canada USD/CAD daily observation, 26 October 2026 — https://www.bankofcanada.ca/valet/observations/FXUSDCAD/json?start_date=2026-10-26&end_date=2026-10-26 (published 2026-10-26) - `lumine-price-2026-10-26` (market): LMN.V delayed quote, 26 October 2026 close — https://finance.yahoo.com/quote/LMN.V/ (published 2026-10-26) - `lumine-newsroom` (primary): Lumine issuer news releases — https://www.luminegroup.com/newsroom/ - `capital-method-roiic` (methodology): Mauboussin and Callahan: Return on Invested Capital — https://www.morganstanley.com/im/publication/insights/articles/article_returnoninvestedcapital.pdf (published 2022-10-06) - `capital-method-cash` (methodology): Dan Shuart, Eagle Point Capital: A Framework For Spotting Value Traps — https://eaglepointcapital.substack.com/p/spotting-value-traps-the-anti-compounders (published 2023-08-26) - `scenario-method` (methodology): Aswath Damodaran: Probabilistic Approaches in Valuation, chapter 33 — https://pages.stern.nyu.edu/~adamodar/pdfiles/val3ed/c33.pdf - `lumine-filings-index` (primary): Lumine statutory filings index — https://www.luminegroup.com/statutory-filings/ - `lumine-q3-2026-newsletter` (primary): Lumine Q3 2026 operating newsletter — https://insights.luminegroup.com/year-in-review/q3-2026/ (published 2026-10-20) - `lumine-events` (primary): Lumine investor events — https://www.luminegroup.com/investor-events/ - `ericsson-q3-2026` (primary): Ericsson Q3 2026 issuer results — https://www.ericsson.com/en/press-releases/2026/10/ericsson-reports-third-quarter-results-2026 (published 2026-10-15) - `nokia-q3-2026` (primary): Nokia Q3 2026 report filed on Form 6-K — https://www.sec.gov/Archives/edgar/data/924613/000110465926117972/tm2628115d1_6k.htm (published 2026-10-22) - `verizon-q3-2026` (primary): Verizon Q3 2026 issuer results — https://www.verizon.com/about/news/verizon-delivers-another-record-breaking-quarter (published 2026-10-26) - `lumine-performance-index` (primary): Lumine financial-performance index — https://www.luminegroup.com/financial-performance/ - `lumine-resources-index` (primary): Lumine resources and publication index — https://www.luminegroup.com/resources/ - `lumine-q2-2026-release` (primary): Lumine Q2 2026 issuer results release — https://www.luminegroup.com/statutory-filings/lumine-group-inc-announces-results-for-the-second-quarter-ended-june-30-2026/ (published 2026-08-04) - `tmobile-q3-2026` (primary): T-Mobile Q3 2026 issuer results — https://www.t-mobile.com/news/business/t-mobile-q3-2026-earnings (published 2026-10-28) - `lumine-iot-expo-2026` (primary): Lumine: IoT Tech Expo Europe 2026 observations — https://www.luminegroup.com/acquire/iot-tech-expo-europe-26-connected-software/ (published 2026-10-29) - `lumine-brighter-futures-2026` (primary): Lumine: Brighter Futures digital access — https://www.luminegroup.com/strengthen/brighter-futures-a-continued-commitment-to-digital-access/ (published 2026-10-29) - `comcast-q3-2026` (primary): Comcast Q3 2026 results — https://www.cmcsa.com/news-releases/news-release-details/comcast-reports-3rd-quarter-2026-results (published 2026-10-22) - `lumine-filings-public-table` (primary): Lumine public statutory-filings table — https://www.luminegroup.com/wp-admin/admin-ajax.php?action=ninja_tables_dt_public&table_id=6187 - `ribbon-q3-2026` (primary): Ribbon Q3 2026 results — https://investors.ribboncommunications.com/news-releases/news-release-details/ribbon-communications-inc-reports-third-quarter-2026-financial (published 2026-10-27) - `charter-q3-2026-calendar` (primary): Charter Q3 2026 earnings-call calendar — https://charter.gcs-web.com/ (published 2026-10-07) - `lumine-q3-2025-fs` (primary): Lumine Q3 2025 financial statements — https://www.luminegroup.com/wp-content/uploads/2025/11/Lumine-Group-FS-2025-Q3.pdf (published 2025-11-03) - `grass-valley-ampp-2026` (primary): Grass Valley and Encompass channel-origination platform — https://www.grassvalley.com/press-release/grass-valley-ampp-provides-an-all-encompassing-approach-to-advanced-channel-origination/ (published 2026-08-28) - `wideorbit-events` (primary): WideOrbit events and webinars — https://wideorbit.com/events/ - `imagine-elta-2026` (primary): Imagine Communications ELTA playout modernization — https://imaginecommunications.com/insights-and-resources/elta-tv-modernizes-playout-with-imagine-communications-for-coverage-of-record-breaking-global-football-tournament/ (published 2026-09-10) - `telenor-nordic-security-2026` (primary): Telenor Nordic digital-security survey release — https://www.telenor.com/media/newsroom/press-releases/only-half-of-nordic-businesses-say-they-are-prepared-to-maintain-operations-duri/ (published 2026-11-02) - `telenor-q3-2026-release` (primary): Telenor Q3 2026 results release — https://www.telenor.com/media/newsroom/press-releases/delivering-through-disciplined-execution/ (published 2026-10-28) - `telenor-bahnhof-approval-2026` (primary): Telenor regulatory announcement: Bahnhof acquisition approval — https://mfn.se/ob/a/telenor/tel-the-swedish-competition-authority-approves-telenor-sveriges-acquisition-of-a-controlling-stake-in-bahnhof-0e299a08 (published 2026-11-02) - `csi-fy2021-mda` (primary): CSU FY2021 management's discussion and analysis — https://www.csisoftware.com/wp-content/uploads/2026/04/CSI-MDA-Q4-2021-Final.pdf - `csi-fy2022` (primary): CSU FY2022 shareholder report — https://www.csisoftware.com/wp-content/uploads/2026/04/Q4-2022-Shareholder-Report.pdf - `csi-fy2023` (primary): CSU FY2023 shareholder report — https://www.csisoftware.com/wp-content/uploads/2026/04/Q4-2023-Shareholder-Report.pdf - `csi-fy2025` (primary): CSU FY2025 shareholder report and audited financial statements — https://www.csisoftware.com/wp-content/uploads/2026/04/Q4-2025-Shareholder-Report.pdf - `lumine-fy2023-mda` (primary): Lumine FY2023 management's discussion and analysis — https://www.luminegroup.com/wp-content/uploads/2025/02/Lumine-MDA-2023-Q4.pdf - `lumine-fy2024-mda` (primary): Lumine FY2024 management's discussion and analysis — https://www.luminegroup.com/wp-content/uploads/2025/03/Lumine-MDA-2024-Q4.pdf (published 2025-03-04) - `lumine-fy2022-mda` (primary): Lumine Group (Holdings) FY2022 management's discussion and analysis — https://www.luminegroup.com/wp-content/uploads/2025/02/Q4-2022-MDA.pdf (published 2023-03-01) - `lumine-motive-announcement-2023` (primary): Lumine: agreement to acquire Nokia's device management and service management platform businesses — https://www.luminegroup.com/newsroom/lumine-group-to-acquire-device-management-and-service-management-platform-businesses-from-nokia/ (published 2023-12-20) - `sncr-fy2024-10k` (primary): Synchronoss Technologies FY2024 Form 10-K — https://www.sec.gov/Archives/edgar/data/1131554/000113155425000029/sncr-20241231.htm (published 2025-03-11) - `csi-synchronoss-release-2025` (primary): Joint CSI/Lumine press release: Synchronoss to be acquired — https://www.csisoftware.com/wp-content/uploads/2026/04/Joint-CSILumine-Press-Release-December-4-2025.pdf (published 2025-12-04) - `csi-q2-2024-mda` (primary): CSU Q2 2024 management's discussion and analysis — https://www.csisoftware.com/wp-content/uploads/2026/04/CSI-MDA-Q2-2024-Final.pdf (published 2024-08-08) - `lumine-board-page` (primary): Lumine board of directors — https://www.luminegroup.com/board-of-directors - `lumine-board-release-2023` (primary): Lumine: appointment of independent directors and board observers — https://www.luminegroup.com/newsroom/lumine-group-inc-announces-appointment-of-independent-directors-and-addition-of-observers-to-board/ (published 2023-02-07) - `lumine-price-history-2026-11-03` (market): LMN.V daily and monthly price and volume history — https://finance.yahoo.com/quote/LMN.V/history/ (published 2026-11-03) - `lumine-q3-2026-portfolio-newsletter` (primary issuer marketing update): Lumine Q3 2026 portfolio newsletter — https://insights.luminegroup.com/year-in-review/q3-2026/ - `lumine-q3-2026-fs` (primary): Lumine Q3 2026 condensed consolidated financial statements — https://www.luminegroup.com/wp-content/uploads/2026/11/Lumine-Group-FS-2026-Q3.pdf (published 2026-11-04) - `lumine-q3-2026-mda` (primary): Lumine Q3 2026 MD&A — https://www.luminegroup.com/wp-content/uploads/2026/11/Lumine-Group-MDA-2026-Q3.pdf (published 2026-11-04) - `lumine-q3-2026-release` (primary): Lumine Q3 2026 results release — https://www.luminegroup.com/statutory-filings/lumine-group-inc-announces-results-for-the-third-quarter-ended-september-30-2026/ (published 2026-11-04) - `amdocs-fy2026-results-2026-11-03` (primary competitor/customer-industry report): Amdocs fiscal 2026 results and fiscal 2027 outlook — https://investors.amdocs.com/news-releases/news-release-details/fourth-quarter-results-fiscal-2026 (published 2026-11-03) - `harmonic-q3-2026-public-context` (primary public competitor report): Harmonic Q3 2026 results — continuing broadband versus divested video perimeter — https://investor.harmonicinc.com/news-releases/news-release-details/harmonic-announces-third-quarter-2026-results (published 2026-11-02) --- # NVIDIA (NVDA) - Legal Name: NVIDIA Corporation - Exchange: NASDAQ (United States) - Sector: Accelerated computing and AI infrastructure - Currency: USD - Reviewed: 2026-11-15 - Model Framework: AI platform operating cash and capital - Financial Period: Trailing four quarters to 25 Oct 2026 (Q3 FY2027 at guidance); July 2026 balances; August/October funding reserves (anchored 2026-10-27) - Research Status: 8 of 11 research checks complete. Still to complete: Security and evidence base; Opportunities and funded scenarios; Variant view and catalysts - Static Report URL: https://longview-investing.vercel.app/research/nvidia.html - Interactive Model URL: https://longview-investing.vercel.app/#/company/nvidia - JSON API: https://longview-investing.vercel.app/api/companies/nvidia ## Executive Summary GPU compute, networking and software serve the AI infrastructure buildout. Long-run returns depend on customer economics, platform durability and the funding needed to convert demand into cash. ## Key Operational Watch Items - Customer GPU utilization, paid AI demand and cash funding - Gross versus operating margin; supplier advances, inventory and extended terms - New chip/custom ASIC competition; software/platform switching costs - Ecosystem investments, partner capacity and contingent guarantees ## Valuation Scenario Assumptions ### Downside Case Digestion: revenue falls 15% below the trailing four quarters in year 1 (about 30% below the Q3 run rate, the base-rate range of past capex-cycle busts), then recovers at 3%. Margin reverts to 30% within three years, working capital 30%, capex 6%. Equity stakes recover 40%; 25% of guarantees and partner capacity paid in year 3. Exit 13x trough-normalized NOPAT. - Key Drivers: Initial growth from estimated run-rate sales: -15.0% | Revenue growth from year 3 onward: 3.0% | Year 5 GAAP operating margin: 30.0% | Capital spending / revenue: 6.0% | Operating working capital / revenue: 30.0% | Normalized operating tax rate: 23.0% | Exit enterprise value / unlevered NOPAT: 13.00× | Annual share dilution: 0.0% | Retained equity marks/cost recovered: 40.0% | Loss stress on partner capacity and guarantees: 25.0% - Revenue growth: -15.0%, 3.0%, 3.0%, 3.0%, 3.0% - Margin: 53.8%, 41.9%, 30.0%, 30.0%, 30.0% ### Base Case Customer-funded growth: year 1 +28% (hyperscalers at 20% of their cash-plus-financing budget, ACIE about 6 GW), then 5% as custom silicon takes share. Margin normalizes to 50% within three years. Capex 4%, working capital 22%. Exit 18x normalized NOPAT; stakes at 80%. - Key Drivers: Initial growth from estimated run-rate sales: 28.0% | Revenue growth from year 3 onward: 5.0% | Year 5 GAAP operating margin: 50.0% | Capital spending / revenue: 4.0% | Operating working capital / revenue: 22.0% | Normalized operating tax rate: 20.0% | Exit enterprise value / unlevered NOPAT: 18.00× | Annual share dilution: 0.0% | Retained equity marks/cost recovered: 80.0% | Loss stress on partner capacity and guarantees: 0.0% - Revenue growth: 28.0%, 5.0%, 5.0%, 5.0%, 5.0% - Margin: 60.4%, 55.2%, 50.0%, 50.0%, 50.0% ### Upside Case Near the demand ceiling: year 1 +58% (559bn against a 576bn ceiling), then 8% as NVIDIA holds share. Margin 58%, capex 3%, working capital 18%. Exit 22x, capped because margins stay near peak; stakes at 100%. - Key Drivers: Initial growth from estimated run-rate sales: 58.0% | Revenue growth from year 3 onward: 8.0% | Year 5 GAAP operating margin: 58.0% | Capital spending / revenue: 3.0% | Operating working capital / revenue: 18.0% | Normalized operating tax rate: 17.0% | Exit enterprise value / unlevered NOPAT: 22.00× | Annual share dilution: 0.0% | Retained equity marks/cost recovered: 100.0% | Loss stress on partner capacity and guarantees: 0.0% - Revenue growth: 58.0%, 8.0%, 8.0%, 8.0%, 8.0% - Margin: 63.1%, 60.5%, 58.0%, 58.0%, 58.0% ## Research Notebook # Business model NVIDIA sells accelerated computing systems: GPUs, CPUs, NVLink and network fabrics, reference rack designs and the CUDA software stack that ties them together. Data Center produced USD89.0bn of USD96.2bn Q2 FY2027 sales (`nvda-q2-fy2027-release`). The rest ("Edge": gaming, professional visualization, automotive) is a mature franchise. Manufacturing is outsourced to TSMC, memory makers, packaging houses and ODMs. External manufacturing does not remove capital needs: supply commitments, inventory, receivables, ecosystem equity stakes and customer credit support all use shareholder capital. Revised 3 Nov 2026 for the research pipeline. Fiscal years end on the last Sunday of January, so FY2026 is roughly calendar 2025. Money is in USD millions unless stated. Fact IDs are in `financials/facts.json`. ## Value-creation mechanism NVIDIA creates value by turning a fixed, front-loaded R&D budget (USD23.6bn trailing, `researchFY2026` etc.) into a new full AI computer system every year. It then secures enough of the world's scarce leading-edge wafers, advanced packaging and high-bandwidth memory to deliver that system before any alternative can. Customers pay for useful compute delivered soon, and judge the price against their cost per token and the revenue that idle capacity forgoes. The integration step creates most of the value: chip, interconnect, networking and CUDA libraries tuned together, sold into every cloud. Management sizes its revenue opportunity per gigawatt of AI capacity at about USD18bn for Hopper, USD25bn for Grace Blackwell and USD40bn for Vera Rubin (`revenuePerGwRubin`, `nvda-q2-fy2027-call` pp.4 and 10). Each generation captures more of the customer's capacity budget. Supply orchestration is the second value step. NVIDIA commits ahead of demand: USD279bn of supply and capacity commitments at July 2026, mainly memory, up from USD95.2bn in January (`supplyCommitments`, `supplyCommitmentsFY2026`). This makes NVIDIA the bottleneck owner in a capacity-constrained market, which lets a 75% gross margin coexist with rising component costs. The third step is newer, and its value is unproven: NVIDIA lends its balance sheet to customers without investment-grade credit. That takes the form of equity stakes in AI labs (nearly USD50bn, `aiLabInvestments`), guarantees (USD108.5bn maximum) and extended payment terms (DSO 60 days, `dsoQ2`). This step creates demand and also absorbs capital and risk. ## Business-model profile Comparable firms are chosen by these traits, not by the industry label. **Growth vehicle: Organic.** Revenue rose from 26,914 in FY2022 to 215,938 in FY2026, driven by data-center demand for its own platforms. Acquisition cash was 0.6% of revenue over FY2024–FY2026. Growth is organic: it comes from customers' capital spending on accelerated computing and is funded by the company's own operating investment. The model forecasts organic units and capacity, with no deal budget. **Capital intensity: 2.5% of sales, asset-light.** Capital spending is about 2.5% of revenue because manufacturing is outsourced to TSMC, memory makers, packaging houses and ODMs. The capital growth uses sits elsewhere: inventory, receivables, supply prepayments and purchase commitments (tangible operating capital of 85.7bn at July 2026), plus ecosystem equity stakes and customer credit support. **Comparables.** A comparable firm is a fabless platform supplier into a customer capital-spending cycle: revenue depends on a few large buyers' capex budgets, growth is organic and internally funded, and margins depend on a technology lead and a software ecosystem. Base rates must treat it as a very large supplier at the top of a spending boom. Not the label: "Semiconductors" mixes NVIDIA with capital-intensive foundries and commodity memory; it is a fabless systems and software platform exposed to its customers' capex cycle. Declared in `valuation/model.json` (`businessProfile`). `npm run check` measures deal and capital intensity from the reported facts, and the outside view grades the forecast only against firms that share them. ## Economic snapshot Definitions: - **Trailing (TTM).** FY2026 + H1 FY2027 − H1 FY2026. - **NOPAT.** GAAP operating income × (1 − 17%), using the FY2027 guided tax midpoint, with stock compensation charged. Investment gains and interest are excluded. - **Balances.** At 26 July 2026. | Measure | Value | Basis | | --- | ---: | --- | | TTM revenue | 302,970 | `revenueTtm` | | TTM GAAP operating income | 197,579 (65.2%) | `operatingProfitTtm` | | TTM NOPAT | 163,991 | 197,579 × 0.83 | | Q2 gross margin | 75.0% | FY2026 71.1%, FY2025 75.0% | | Incremental operating margin, FY2025→FY2026 | 57.3% | Δ operating income 48,934 / Δ sales 85,441 | | Receivable days | 60 | Up from 51 in January | | Inventory days on Q2 cost of sales | 120 | 31,575 / (24,079 × 4) × 365 | | Payable days | 57 | 15,059 / annualized cost of sales | Capital base at July 2026: | Component | Amount | | --- | ---: | | Working capital excluding cash (receivables 63,059 + inventory 31,575 + prepaids 3,409 − payables 15,059 − accrued liabilities 26,960) | 56,024 | | PP&E | 14,285 | | Right-of-use assets | 5,390 | | Other operating noncurrent assets (supply prepayments and similar, ex equity and warrants) | 9,965 | | **Tangible operating capital** | **85,664** | | Goodwill and intangibles (mostly Groq, 2026) | 24,123 | | Ecosystem equity stakes at carrying value | 98,897 | | Cash and debt securities | 56,586 | | Debt | 33,500 | | Lease liabilities | 5,494 | Returns on that capital: - Return on tangible capital, capitalizing leases: 191%. - Return on capital employed, expensing leases (excluding the ROU asset): 204%. Including goodwill and intangibles: 157%. - Including the ecosystem stakes as capital that now supports demand: 79%. These are ending balances rather than averages, which flatters nothing because capital is growing. Normalization: TTM capex is 7,474 against depreciation of 3,687. That is about twice D&A, which reflects growth capacity rather than under-maintenance, so no add-back is made. There are no minorities, and recent acquisitions are licenses and teams with no separately disclosed profit. The real normalization question is cyclical. The Q2 75% gross margin and 65% operating margin sit far above NVIDIA's own pre-AI record: - FY2022: 64.9% gross, 37.3% operating. - FY2023 downturn: 56.9% gross, 15.7% operating, including the Arm write-off and inventory provisions. Stage 7 and the scenarios treat today's margin as a shortage-era level, not a through-cycle norm. ## Economic segmentation Reported segments are evidence, not the answer. In FY2026, Compute & Networking earned 130,141 of 139,297 segment operating income, or 93%, on a 67% segment margin. Graphics earned 9,156, a 41% margin (`segmentComputeNetworkingOpFY2026`, `segmentGraphicsOpFY2026`). Cut by how value is created and who decides the purchase: | Economic unit | Size | Approximate profit share (estimate) | Demand driver | Character | | --- | --- | ---: | --- | --- | | Hyperscale AI systems (Microsoft, Google, Amazon, Meta, Oracle) | 48.7bn of Q2 Data Center, about 55% (`hyperscaleRevenueQ2`) | ~50% | The buyers' own capex budgets. Partly for rented capacity, where their cloud customers ask for CUDA; partly for internal workloads, where their own chips compete. | Growing, but this is where buyer power and substitutes are strongest. | | ACIE: neoclouds, sovereigns, enterprises and AI labs | About 40bn of Q2, +138% year on year | ~40% | AI labs' and start-ups' compute needs, funded by venture capital (over USD400bn in H1 2026, per management), debt and NVIDIA's own equity and credit. About a quarter of next year's business is with AI labs NVIDIA supports with its balance sheet (`aiLabFinancedShare`). | Fastest growing. Weaker credit and highly cyclical with capital markets. | | Networking (NVLink, InfiniBand, Spectrum-X) | 31,376 in FY2026, +142% (`dataCenterNetworkingFY2026`) | Inside the two units above | Attached to each compute deployment | The Mellanox franchise. It raises content per gigawatt and switching cost. | | Edge: gaming, workstations, automotive | 22,459 Graphics segment revenue in FY2026 | ~7% | PC and replacement cycles; memory prices now weigh on PCs | Mature and cyclical; a small share of value. | The profit split between hyperscale and ACIE is an estimate. NVIDIA does not disclose margins by customer type. I assume similar gross margins, because hyperscalers' volume discounts and ACIE's extended terms and financing roughly offset. The implication: about 90% of profit and nearly all of the value depend on one product cycle sold to two kinds of buyer: - Five hyperscalers, all of which build their own accelerators. - A set of capital-hungry AI labs and neoclouds whose purchases NVIDIA partly finances. A sector label ("semiconductors") hides this. The demand is a capital-spending cycle, and the cycle's financing is partly NVIDIA's own. ## Unit economics **Unit:** one gigawatt of Vera Rubin AI capacity. Customers add capacity in gigawatt-scale AI factories, and management prices its opportunity per gigawatt. Revenue comes from management's sizing. The other lines are this repository's estimates, built from guided ratios. | Line, per GW | Amount (USD bn) | Basis | | --- | ---: | --- | | NVIDIA revenue | 40.0 | Management sizing (`revenuePerGwRubin`) | | Cost of goods at the 72.5% FY2028 margin guide | 11.0 | Memory-heavy (`grossMarginGuideFY2028Low`, p.7) | | Gross profit | 29.0 | | | Allocated operating expense | 3.4 | GAAP opex at 8.5% of sales (Q3 guide 9.2bn on 108bn). Mostly fixed R&D; truly variable selling and support are under 1 point. | | Unit operating profit | 25.6 | 64% | | NOPAT at 17% | 21.2 | | | Working capital at 22% of sales | 8.8 | | | Capex at 1.5% of sales | 0.6 | | | **Tangible capital** | **9.4** | | | **Return on tangible capital** | **~225%** | | Adding NVIDIA's financing changes the picture. OpenAI's ~12 GW is supported by part of the ~USD50bn in lab equity, about USD4bn per GW (`nvda-q2-fy2027-call` p.6). With that included, capital becomes about USD13bn per GW and the return about 160%. The NVIDIA-side unit is therefore not capital-constrained: its value depends on price per gigawatt, not on capital intensity. The consensus question is the customer's unit, because their returns fund NVIDIA's price. Management puts a full gigawatt at about USD60bn of customer spend today (p.9). Servers are depreciated over five to six years at Amazon, Meta, Alphabet and Oracle (`amzn-q3-2026`, `meta-q3-2026`, `googl-q3-2026`, `orcl-q1-fy2027`). On that basis the customer carries roughly 8–10bn of annual depreciation per GW. Power at an assumed USD0.08/kWh adds about 0.7bn. Earning 10% pre-tax on USD60bn therefore needs about USD15–17bn of annual revenue per GW. If annual product cadence shortens the economic life to three years, the requirement rises to about USD22bn+. Customer revenue is responding: - Google Cloud +103% to 30.7bn in Q3. - AWS +39% to 46.0bn. But customers' capex now exceeds their operating cash flow: - Alphabet free cash flow −9.1bn in Q3. - Amazon trailing free cash flow −42.4bn. - Meta −2.1bn in Q3. - Oracle −5.4bn, while selling USD20bn of stock. The customer unit is not yet self-funding at the margin. **Failure case: H20 for China.** NVIDIA designed a product to fit export rules, then a regulator closed the market. The result was a 4.5bn inventory and purchase-obligation charge, about USD60m of licensed revenue, and no H200 revenue under the February 2026 license (`inventoryProvisionsFY2026`, `nvda-fy2026-10k`). The levers that work with customers (performance, supply, software) fail when the counterparty with the outside option is a government. An earlier failure was FY2023. Inventory doubled (2,605 to 5,159, `inventoryYeFY2023`), the gross margin fell 8 points and operating income fell 58% while revenue was flat. Shipped capacity outran end demand. ## Independent checks 1. **Segment accounts.** Compute & Networking earned a 67.3% segment margin in FY2026, before unallocated stock compensation and acquisition costs. Group GAAP operating margin was 60.4%. The per-GW model's 64% sits between them, consistent with allocated fixed R&D. 2. **Customer budgets.** The five hyperscalers' own 2026 guidance sums to about 810–830bn: - Microsoft about 170bn. - Alphabet 195–205bn. - Amazon about 220bn. - Meta 135–145bn. - Oracle FY2027 about 90–95bn. This figure comes from a call summary and is unverified. NVIDIA puts it at "nearly 800bn" (p.3). NVIDIA's hyperscale revenue annualizes to about 195bn, roughly 24% of that capex. That is consistent with 40bn of NVIDIA content in about 60bn per GW, once land, buildings, power and these customers' own TPUs, Trainium and MTIA are excluded. 3. **Competitors' audited revenue.** In the quarter to early August, Broadcom's AI semiconductor revenue was 16.7bn (+221%) and AMD's Data Center revenue was 6.7bn (+107%) (`avgo-q3-fy2026`, `amd-q2-2026`). NVIDIA's Data Center revenue was 89.0bn (+117%). NVIDIA still takes about 79% of these three firms' AI data-center revenue. Custom silicon is growing about twice as fast from a smaller base. 4. **Adverse allocation.** Charge all trailing R&D (23.6bn) and all stock compensation (7.2bn) to the hyperscale unit alone. Its operating margin still exceeds 55%. No unit is unprofitable today, so the question is durability of price, not current profitability. Where the consensus disagrees: one camp calls this a supply-constrained secular buildout; the other, circular financing. Three observations would settle it: - Receivable days and receivables growth against revenue. - The share of revenue from customers NVIDIA finances. - Whether hyperscaler free cash flow turns positive while capex keeps rising. Stage 11 freezes these as prospective checks. # History, demand and counterparties NVIDIA's GPU business grew from graphics into general accelerated computing. Its software ecosystem then made it the default data-center platform for training and inference. The financial question today is how much of the AI infrastructure buildout becomes lasting owner cash, after suppliers, employees and customers receive the capital they need. The record shows how fast the business moved: - Revenue went from 16,675 in FY2021 to 215,938 in FY2026. - GAAP operating income went from 4,532 to 130,387. - Trailing revenue at July 2026 was 302,970. Fiscal labels are not calendar years: FY2026 ended 25 January 2026 (`nvda-fy2026-10k`, `nvda-xbrl-companyfacts`). Prior H20 restrictions and inventory provisions distort year-on-year comparisons. The financing structure changed in 2026: - USD25bn of notes issued in June. - About USD99bn of equity stakes. - A USD105bn SB Energy guarantee (`nvda-q2-fy2027-10q`, notes 5–10; `nvda-sb-energy`). No external NVIDIA investment thesis or published shareholder IRR was supplied. Scenario schedules are this repository's estimates. ## Demand drivers and cycle **Who uses, who pays, who decides.** The end user is a business or consumer using an AI service. The payer is whoever owns the capacity: - A hyperscaler, for its own services or to rent out. - A neocloud. - A sovereign program. - An AI lab, often through a cloud contract. The decision-maker is the payer's infrastructure team, but cloud renters' software choices constrain it. Hyperscalers buy NVIDIA for rented capacity because their customers' code runs on CUDA. For internal workloads they can choose their own silicon. **Split of demand.** Using management's own categories and this repository's estimates, current demand is: | Kind of demand | Evidence | | --- | --- | | Mostly growth capacity | Neoclouds expected to go from about 3 GW installed at end-2025 to 8 GW by end-2026 (`nvda-q2-fy2027-call` p.3). | | Increasingly replacement | Annual product cadence; prior-generation Trainium and GPU capacity still rents at a premium (`amzn-q3-2026`), so replacement is driven by economics rather than wear. | | Partly regulation-driven | Sovereign "AI for data residency" programs grew 35% sequentially (p.5). | | A financing-driven part | AI labs supported by NVIDIA's balance sheet are about a quarter of next year's business (`aiLabFinancedShare`). | Little is discretionary consumer demand. **The cycle, measured.** NVIDIA's own record has two inventory cycles in the last eight years. The second, FY2023, had: - Flat revenue (26,914 to 26,974). - Inventory doubling (2,605 to 5,159). - Gross margin down from 64.9% to 56.9%. - Operating income down 58%. Today's customer side shows the ingredients of a late capital cycle (`googl-q3-2026`, `amzn-q3-2026`, `meta-q3-2026`, `orcl-q1-fy2027`). Capex roughly doubled year on year in Q3 2026: | Company | Q3 2026 capex | Q3 2025 capex | | --- | ---: | ---: | | Alphabet | 51.4bn | 24.0bn | | Amazon | 77.3bn | 35.1bn | | Microsoft | 39.4bn | 19.4bn | It now exceeds internal cash generation: - Alphabet Q3 free cash flow −9.1bn. - Amazon trailing free cash flow −42.4bn, plus 48.1bn of new long-term debt in Q3. - Oracle −5.4bn, plus a 20bn equity sale. Revenue is following the capex: - Google Cloud +103%. - AWS +39%, backlog 578bn. - Microsoft commercial RPO 677bn. - Oracle RPO 664bn. Much of that backlog is AI-lab commitments that are themselves externally financed. Microsoft lowered its calendar-2026 capex estimate from about 190bn to about 170bn, citing delayed third-party deliveries and lease accounting (`msft-fy2027-q1`). **Market misperception.** The common bull view is a "supply-constrained secular" buildout. The common bear view is "circular financing". The series that tests both is customer capex against customer operating cash flow and external financing, alongside NVIDIA's receivable days (45 → 60) and its financed share of revenue. The current data support strong 2027 demand: Alphabet and Amazon both guide 2027 capex higher. They also show that marginal demand is now funded by debt, equity and NVIDIA's own balance sheet rather than by hyperscaler surplus cash. That makes revenue more cyclical than an installed-base replacement model would suggest. ## Competition Competition happens in three places: 1. **Inside each hyperscaler's own fleet.** NVIDIA competes with: - Google's TPU: generation 8t/8i, nearly 100 customers, external system sales ramping in 2027, Anthropic as a large user. - Amazon's Trainium: commitments up more than 20x, to be sold outside AWS from late 2027. - Microsoft's Maia 200/300: Microsoft claims 40% better tokens per watt than leading GPU systems. - Meta's MTIA, built with Broadcom. Sources: `googl-q3-2026`, `amzn-q3-2026`, `msft-fy2027-q1`. Here the buyer is also the competitor, and the battleground is internal inference cost. 2. **Merchant accelerators for rented and lab capacity.** AMD's MI400/Helios has commitments from Anthropic (up to 2 GW), Meta (up to 6 GW), OpenAI, Oracle and Microsoft (`amd-q2-2026`). OpenAI's own Broadcom XPU is, on Broadcom's call, on track for 1.3 GW in 2027. That call comes from an unverified third-party transcript. 3. **Networking.** Broadcom and Ethernet switch vendors compete against NVLink, InfiniBand and Spectrum-X. **Share.** In the latest comparable quarter NVIDIA had about 79% of NVIDIA + AMD Data Center + Broadcom AI revenue (89.0 / 112.4bn). Broadcom grew 221% and AMD Data Center 107%, against NVIDIA's 117%. The number of serious competitors is rising, not falling: Broadcom alone reports six XPU customers. **The advantage in one sentence.** NVIDIA delivers the most useful compute soonest to any customer in any cloud, because it ships a fully integrated system every year, on supply it secured first, with the software most AI developers already use. The durability evidence is mixed: - For durability: share held through Hopper, Blackwell and Rubin; a 75% gross margin despite record memory costs; purchase orders for Vera Rubin from every major hyperscaler (p.4). - Against durability: the largest buyers, and NVIDIA's own investees OpenAI and Anthropic, are building alternatives (p.11). **The competitor that could hurt NVIDIA is Google.** It owns the whole stack: chip, compiler, cloud and frontier model. It is now selling TPU systems outside its own fleet, and Anthropic's commitments give it scale. It has not hurt NVIDIA yet for three reasons: - TPUs were internal-only until recently. - Their supply depends on the same TSMC and HBM capacity that NVIDIA reserves first. - Most outside developers' code and tooling target CUDA. ## Counterparty incentives **Hyperscalers** (Microsoft, Alphabet, Amazon, Meta, Oracle; about 55% of Data Center revenue). - What they want: lower cost per token, supply security and freedom from a single vendor's margin. - What NVIDIA gives them: the capacity their cloud customers ask for, delivered first. - Their outside options: proven in-house chips, AMD. - Rational behaviour: move internal inference to their own silicon while buying NVIDIA for rented capacity and frontier training. - What changes the deal: their own chips becoming good enough for external renters (Google's external TPU sales and Amazon's 2027 Trainium sales are the first moves). - Dependence runs both ways. NVIDIA's response is to grow ACIE, the customers without their own silicon, to "roughly half" of Data Center (p.5). That deliberately reduces buyer power. **AI labs** (OpenAI with about 12 GW of NVIDIA commitments, Anthropic, others). - What they want: the most compute at the lowest cost, plus supplier diversity. Both are building or buying alternatives: OpenAI's XPU, Anthropic's TPU, Trainium and AMD. - What NVIDIA gives them: equity (nearly 50bn), credit enhancement for about 2 GW at another lab, and guarantees for OpenAI's 4.25 GW Ohio site (p.6). - Their outside options: many for chips, few for credit. NVIDIA's balance sheet is therefore its bargaining chip. - Risk to NVIDIA: if lab revenue disappoints, NVIDIA loses sales, receivables, equity value and guarantee exposure at the same time. **Suppliers** (TSMC, SK hynix, Samsung, Micron, packaging firms). - What they want: long commitments that pay back their capacity, and pricing power while capacity is short. - What NVIDIA gives them: scale and up to 279bn of commitments. - Their outside options: Google/Broadcom, AMD, Amazon. - Current state: memory makers are taking value. Memory pricing "exceeded expectations", and NVIDIA lowered its margin expectations to 71–72% in Q4 and 72–73% in FY2028 (p.7). **The U.S. government** wants control over AI capability exports. China Data Center compute is out of the outlook. Regulation can remove a market regardless of customer demand, as the H20 charge showed. **ODMs and distributors** (the 22% and 14% direct customers) pass the systems through. Their power is small. # Management and judgment Jensen Huang co-founded NVIDIA in 1993. His resignation as president, CEO and director was accepted on 11 November and disclosed on 12 November. Colette Kress is interim CEO and continues as CFO; Harvey C. Jones chairs the board. The permanent successor is not yet named. See the filed [8-K, Item 5.02](https://www.sec.gov/Archives/edgar/data/1045810/000104581026000082/nvda-20261111.htm). The investment case depends on architectural execution, supply coordination and, increasingly, capital allocation across customers, financiers and suppliers. Management's product and demand statements are evidence to examine, not achieved cash flow. **13 November source review.** Three original Stevens Forms4 (filed2,8,22September) report4,236,740 shares sold31August–18September for approximatelyUSD946.63m at the disclosed weighted prices. All three10b5-1 flags are0. These are secondary sales from trusts, not issuer cash, dilution or evidence of motive. The latest form retains11,544,612 direct shares,15,017,750 Envy Trust shares and970,531 Third Millennium Trust shares. This corrects the limits of a media headline without implying misconduct. Reuters' Kress profile (AOL syndication, updated13November02:16UTC) describes attributed disagreements over financing; it is not a cancellation notice or a new authorized cash commitment. MHA's12November original statement distinguishes domestic false-declaration evidence from espionage: it reports an inaccurate-information warning/bar, insufficient evidence against Huang on the server declarations and an ongoing Megaspeed investigation. No all-clear or quantified NVIDIA loss follows. **8 November 2026 governance update.** The [board announced an independent investigation](https://nvidianews.nvidia.com/news/statement-from-nvidia-board-of-directors) into reports of foreign intelligence targeting Huang. Independent directors and outside counsel oversee it. The board retains confidence in current leadership and says it has no reason to believe company secrets were compromised; those are the board's positions while the inquiry is open. The [CEO's issuer-published response](https://nvidianews.nvidia.com/news/a-message-to-our-company) supports the inquiry. Those 8 November statements preceded the confirmed 11 November leadership decision; they no longer describe the current succession position. Monitor decision-making continuity, information security and regulatory/customer consequences; personal conduct alone does not quantify owner-cash impairment. On the August call, management gave a preliminary, supply-constrained outlook of about 70% FY2028 revenue growth (`fy2028GrowthOutlook`). That informs near-term cases only. It does not establish a decade of growth or customers' returns. Starting in FY2027, NVIDIA's non-GAAP measures retain stock compensation, and history was recast (`nvda-q2-fy2027-release`). The model uses GAAP operating expenses and never adds stock compensation back. ## Ownership and incentives Sources: `nvda-proxy-2026` pp.35–36 (ownership at 23 Mar 2026), the summary compensation table and CD&A pp.44–46; `nvda-huang-form4`. | Person | Role at the historical proxy date | Shares (m) | Value at USD217.94 | FY2026 reported pay | Notes | | --- | --- | ---: | ---: | ---: | --- | | Jen-Hsun Huang | Founder, CEO | 870.6 (3.58%) | ~USD190bn | USD36.3m | Salary 1.5m, PSUs 24.8m, cash incentive 6.0m, security and other 4.0m. 10b5-1 plan of 20 Mar 2025 for up to 6m shares (about 0.7% of holdings), expired 31 Dec 2025. No open-market sales in 2026, only tax withholding and gifts (438k shares to his foundation and a donor-advised fund in Sep 2026). In Dec 2025, 39.85m shares were distributed into an independently managed LLC that he no longer beneficially owns. | | Colette Kress | CFO | 4.4 | ~USD1.0bn | USD14.3m | 10b5-1 plans adopted and terminated in 2026 | | Jay Puri | EVP Worldwide Field Operations | 3.3 | ~USD0.7bn | n/a | 10b5-1 plan for 1m shares (Sep 2025) | | Debora Shoquist | EVP Operations | 1.6 | ~USD0.3bn | n/a | 10b5-1 plan for 0.31m shares | | Tim Teter | General Counsel | 2.7 | ~USD0.6bn | n/a | 10b5-1 plan (May 2026) | | All 14 directors and officers | | 957.3 (3.94%) | ~USD209bn | | Largest outside holders: BlackRock 7.43%, Vanguard 7.31% | **Historical founder alignment.** The proxy holdings and USD217.94 valuation below predate the transition. They are not a current beneficial-ownership update, an incentive schedule for Kress, or evidence that Huang retains executive authority. No new severance or award-forfeiture amount is disclosed in the transition filing. **What the stake meant under the former CEO.** If the stock compounds at 10% a year for five years, Huang's stake gains about USD116bn. Five years of pay at the FY2026 level is about USD180m, roughly 650 times smaller. Ownership, not pay, sets his interests. His sales have been small relative to his holdings and pre-scheduled. There is no evidence of exercising early to sell near a peak. **What pay rewards.** FY2026 pay had three parts: - **Cash plan:** paid on revenue. The base goal of USD130bn equalled FY2025's actual USD130.5bn, so zero growth earned a 100% payout. Actual revenue of USD215.9bn paid 200%. - **One-year PSUs:** paid on non-GAAP operating income. The base was USD71bn, below FY2025's level. The stretch goal was revised down from USD120bn to USD96bn after the H20 export controls. Actual was USD137.3bn (`cashPlanBaseRevenueFY2026`, `psuBaseOperatingIncomeFY2026`). - **Multi-year PSUs:** paid on three-year TSR against the S&P 500. FY2024–FY2026 finished at the 100th percentile (1,055% TSR). Backing out the targets: base operating income of USD71bn is about USD59bn of NOPAT at 17%. Against a market value of about USD5.3tn, that is roughly 90 times earnings, so the incentive goals say nothing about value at today's price. No metric charges management for capital used. Not return on capital, nor the roughly USD100bn of ecosystem stakes, nor the USD108.5bn of guarantees. Revenue-weighted pay rewards exactly the customer financing that inflates revenue. Pay is too small relative to ownership for this to dominate Huang's behaviour. It is a weaker check on the wider executive team. ## Track record and references **Record under the pre-transition team.** - **Returns on capital:** FY2022 operating margin 37.3%, FY2023 15.7%, FY2026 60.4%. GAAP operating NOPAT went from 4,413 in FY2023 to 110,676 in FY2026 (`operatingNopatFY2023`, `operatingNopatFY2026`). - **Market won:** about 79% of the three leading AI-accelerator vendors' data-center revenue (`research/02-history.md`). - **Total shareholder return:** 1,055% over FY2024–FY2026 (`nvda-proxy-2026`). **Decisions that worked.** - The Mellanox acquisition (completed FY2021): networking reached 31.4bn in FY2026, more than ten times FY2021 (`dataCenterNetworkingFY2026`). - Committing to CUDA and data-center systems a decade before the demand arrived. - Buying back USD10.0bn of stock in FY2023, during the downturn (`buybacksFY2023`). **Decisions that failed.** - **Arm:** terminated for regulatory reasons, with a 1.35bn write-off of the prepayment, though NVIDIA kept the 20-year license (`nvda-arm-termination`). - **FY2023 inventory glut:** inventory doubled and gross margin fell 8 points. - **H20 for China:** a 4.5bn charge. - **Cryptomining disclosure:** an SEC penalty over FY2018 disclosure of cryptomining effects (`sec-nvda-crypto-2022`). **References.** None were collected. This repository has done no interviews with former employees, customers or suppliers. Outside views read here are counterparties' own statements: - Broadcom's and AMD's claims about competing chips. - OpenAI's public claims about its own chip, which an analyst raised on NVIDIA's call (p.11). Those are competitors' claims, not references. The proxy-reported bench is long-tenured (Kress, Puri, Shoquist, Teter). Still, product architecture and the ecosystem-financing strategy are visibly Huang's, so key-person risk is real and unquantified. ## Capital allocation principles 1. **Zero-based allocation.** Partly met. R&D has followed data-center opportunity: Compute & Networking is 90% of revenue, and the networking, CPU and Groq inference technology all aim at raising content per gigawatt. Gaming was not starved but was not favored either. 2. **Funding strategies, not projects.** Met, and the strategy is explicit. Stakes in AI labs, guarantees for neocloud and lab sites, and six infrastructure-finance partnerships all aim to build a customer base that does not design its own chips (p.6). The strategy is coherent. It also makes NVIDIA a lender to its own demand. 3. **Capital is accessible but has a cost.** Mixed: - In favour: including stock compensation in non-GAAP from FY2027. - Against: the CFO justified the nearly USD50bn of lab investments as "a small fraction" of expected free cash flow (p.6), a test of affordability, not of return against a hurdle. - Against: buybacks run at about USD20bn a quarter regardless of price. 4. **Zero tolerance for bad growth.** Partly met. NVIDIA walked away from Arm when regulators blocked it, wrote off H20 rather than chase China, and exited cryptomining products. But guarantee and equity exposure has expanded quickly without a stated limit. 5. **Price versus value.** Weak evidence either way: - FY2023's buyback was bought cheaply. - FY2025–H1 FY2027 repurchases of about USD114bn (33.7bn + 40.1bn + 39.8bn) were made near record prices, with no stated valuation discipline. - The equity stakes are marked up (23.7bn of H1 gains) but not yet realized. **Governing objective before the transition.** Platform share and the AI buildout dominated the former team's public allocation decisions. The founder's historical stake does not establish the new interim team's authority or future discipline. Assess Kress's actual commitment decisions and technical-management structure rather than assign her either Huang's results or an assumed capital-allocation improvement. ## Integrity and governance **Governance structure.** - One share class with one vote per share; no dual-class control. - Nine of ten directors are independent; an eleventh joined in July 2026. - Shareholders passed a proposal to remove supermajority voting. - Say-on-pay passed with about 93.6% of for-plus-against votes (15.7bn for, 1.1bn against; `nvda-agm-2026-votes`). **Related-person transactions** (`nvda-proxy-2026` p.34): - Huang's two adult children are employees, paid about USD1.2m and USD1.3m. - A director's son is an employee. - The Huang foundation buys GPU time from CoreWeave to donate to researchers, and NVIDIA plans free engineering support for some recipients. CoreWeave is also an NVIDIA customer and investee. The amounts are small, but the overlap should be watched. **Integrity evidence.** - Bonus goals were revised automatically, and downward, when export rules changed. That was a formula, but it protected payouts. - The 2022 SEC settlement concerned under-disclosure of a cyclical demand source. That is relevant today, when financed AI-lab demand is about a quarter of next year's business. **Agency costs.** - **Control of resources.** The material agency cost. Pay keyed to revenue and operating income, together with a strategy of financing customers, can grow revenue while shifting risk onto the balance sheet: about USD99bn of stakes, USD25bn of further commitments and USD108.5bn of maximum guarantees. - **Risk tolerance.** The founder's concentrated stake is aligned with owners, though his wealth makes him tolerant of large bets. - **Time horizon.** One-year and three-year PSUs are shorter than this five-to-ten-year horizon. - **Buybacks.** They do not merely offset stock compensation. USD79.9bn was repurchased in FY2026–H1 FY2027, against USD10.3bn of stock compensation expense, and diluted shares fell from 24,804m to 24,285m. **How the valuation treats agency costs.** No separate discount. The model charges GAAP stock compensation, credits no buyback accretion, recovers only 40%/80%/100% of equity stakes, and pays a 25% loss on contingent exposure in the downside. ## 9 November legal follow-up The [dated review](../updates/2026-11-09-daily-review.md) records the filed-suit notice and sources distinguishing allegations, requests and findings. The complaint/docket was not directly reviewed. No new restriction, leadership decision or quantified company loss is established. Keep the existing operating paths and explicit unquantified tail; reassess on discriminating cash/contract or governance evidence. ## 12 November succession assessment The [dated review](../updates/2026-11-12-daily-review.md) reassesses the complete distribution. Founder-led architecture, talent coordination, supplier commitments and customer financing may be harder to coordinate during an interim period. Long-tenured operating leaders, installed CUDA and existing funded orders remain assets; a departure does not erase them. Annual cash paths and terminal multiples stay unchanged. A small, low-confidence transfer of five percentage points from upside to downside makes both horizons 45%/45%/10%. This is a judgment about joint execution outcomes, not a measured probability of misconduct or a cash-loss estimate. # Capital allocation and cash definitions This section was revised on 3 Nov 2026. Money is in USD millions, and fiscal years end in late January. The annual figures come from each 10-K's statements, extracted through `nvda-xbrl-companyfacts`. Each measurement is a named step in `valuation/calculations.json`. ## Sources and uses of cash | FY | Operating cash | Capex | Acquisitions and Groq | Equity stakes, net | Buybacks | RSU tax withholding | Dividends | Stock-plan proceeds | Debt, net | Other investing | Unreconciled | Change in cash and securities | | --- | ---: | ---: | ---: | ---: | ---: | ---: | ---: | ---: | ---: | ---: | ---: | ---: | | 2022 | 9,108 | −976 | −263 | 0 | 0 | −1,904 | −399 | 281 | 3,977 | −24 | −153 | 9,647 | | 2023 | 5,641 | −1,833 | −49 | −77 | −10,039 | −1,475 | −398 | 355 | 0 | 0 | −37 | −7,912 | | 2024 | 28,090 | −1,069 | −83 | −861 | −9,533 | −2,783 | −395 | 403 | −1,250 | −124 | 293 | 12,688 | | 2025 | 64,089 | −3,236 | −1,007 | −1,315 | −33,706 | −6,930 | −834 | 490 | −1,250 | 22 | 903 | 17,226 | | 2026 | 102,718 | −6,042 | −14,535 | −17,418 | −40,086 | −7,948 | −974 | 644 | 0 | 0 | 2,987 | 19,346 | | **Total** | **209,646** | **−13,156** | **−15,937** | **−19,671** | **−93,364** | **−21,040** | **−3,000** | **2,173** | **1,477** | **−126** | **3,993** | **50,995** | Cash and marketable securities rose from 11,561 to 62,556, which matches the total change (`cashSecuritiesYeFY2021`, `cashSecuritiesYeFY2026`). The unreconciled 3,993 is about 1.9% of operating cash. It consists mainly of changes in securities marks and FX, PP&E principal payments (101 in FY2026) and other small financing lines. The FY2026 residual is larger and has not been traced. I record it rather than plug it. Inside operating cash: - Stock compensation added back: 19,385 over five years. - Working capital absorbed: 30,317 (balance-sheet basis). - Operating cash before working capital and stock compensation: about 220,600. **Priorities revealed.** About half of five-year operating cash (114,404) went back to shareholders as buybacks plus RSU tax withholding, which is economically a repurchase. Dividends were trivial until the FY2027 increase to USD0.25 a quarter. The pattern changed sharply in FY2026 and H1 FY2027, with the same team in charge (Huang as CEO, Kress as CFO): - Acquisitions and licenses rose to 14.5bn in FY2026, mostly Groq. - Ecosystem stakes rose from 1.3bn to 17.4bn in FY2026, then 42.4bn in H1 FY2027 alone, about 57% of that half's 74.4bn operating cash. - NVIDIA issued 25bn of debt in June 2026. ## Investment rates per sales change Over the five years, sales rose 199,263, from 16,675 in FY2021 to 215,938 in FY2026 (`salesChangeFy21Fy26`). | Rate | Amount | Per unit of Δsales | Step | | --- | ---: | ---: | --- | | Incremental M&A (acquisitions + Groq; no material divestitures) | 15,937 | 8.0% | `acquisitionRate` | | Internal intangibles, all R&D | 52,693 | 26.4% | `researchRate` | | Internal intangibles, R&D step-up only | 14,573 | 7.3% | `researchGrowthRate` | | Fixed capital (capex 13,156 − D&A 8,933) | 4,223 | 2.1% | `fixedCapitalRate` | | Working capital | 30,317 | 15.2% | `workingCapitalRate` | | Ecosystem equity stakes (financial, shown separately) | 19,935 | 10.0% | `equityStakeRate` | NVIDIA invests through expensed R&D and working capital, not fixed assets. Each dollar of new sales needed about 2 cents of net fixed capital and 15 cents of working capital. The inflections: - **Working capital as the main use.** It turned from a small use into the main one after FY2024. DSO rose from 51 to 60 days and inventory reached 31.6bn at July 2026 (`dsoQ2`). - **Ecosystem stakes.** They now exceed every other operating use. - **Capex.** It rose to about 2.5% of sales (TTM 7,474 against 3,687 of D&A) as NVIDIA builds its own AI and R&D infrastructure. Contracted capex is 8bn. ## Reinvestment and returns | Measure | Value | Step or basis | | --- | ---: | --- | | Owner earning power, FY2021 GAAP operating NOPAT | 4,453 | `operatingNopatFY2021` | | Owner earning power, FY2026 | 110,676 | `operatingNopatFY2026` | | Owner earning power, TTM at 17% | 163,991 | | | Five-year reinvestment rate | 22.5% | Growth investment 50,477 (net capex + working capital + acquisitions) / matching NOPAT 224,603 (`reinvestmentRateFy22Fy26`). Including equity stakes, 31%. | | Lagged five-year ROIIC on tangible capital | 472% | ΔNOPAT 106,223 / Δcapital FY2021→FY2025 of 22,521 (`roiicLaggedFy26`) | | Rolling three-year ROIIC from the FY2023 trough | 530% | ΔNOPAT 106,263 / Δcapital FY2022→FY2025 of 20,052 (`roiicLaggedFy23Fy26`) | | Operating leverage | 63.2% | ΔOperating income / Δsales, FY2021→FY2026 (`operatingLeverageFy21Fy26`). Operating margin went from 27% to 60%. | The identity, growth ≈ reinvestment rate × ROIIC, does not hold here, and the reason is informative. NOPAT grew at a 90% compound rate. 22.5% × 472% gives 106%, roughly the right order. But the ROIIC is not a capital-productivity measure: tangible capital did not create the earnings. Expensed R&D did, along with price and mix on a product in shortage and operating leverage on largely fixed R&D. Treat all five years of R&D as unamortized capital, an upper bound on the capital base. After tax, that adds 43,735 of investment, which lifts the reinvestment rate to about 40% and cuts the incremental return to about 140% (106,223 / (22,521 + 52,693)). Those are still extraordinary, but they are a rent on scarcity and integration, not a reinvestment engine that compounds at a stable rate. So the forecast does not use a Harvia-style reinvestment × ROIIC grower. It funds growth capital explicitly and judges margins separately. The mix by use: - **Organic tangible investment** (net capex and working capital): 34,540, earning the returns above. - **Acquisitions:** 15,937. Their earnings cannot be measured separately (see below). - **Equity stakes:** 19,935 over five years plus 42,404 in H1 FY2027. Their return so far is unrealized marks: 23.7bn of H1 FY2027 gains, and July carrying value of about 99bn against about 62bn of purchases since FY2023. Marks are not owner cash. - **Payouts:** 117,404 including RSU withholding and dividends. The forward diagnostics are generated from the shared engine in `research/10-capital-efficiency.md`. ## Acquisition record | Deal | Date | Price | What was bought | Result | | --- | --- | --- | --- | --- | | Mellanox | Completed FY2021 | Within 8,524 of FY2021 acquisition cash | InfiniBand and Ethernet networking | Networking revenue 31,376 in FY2026, more than 10× FY2021 (`dataCenterNetworkingFY2026`). Profit not disclosed separately. Strategically the most successful deal: it raised content per gigawatt. | | Arm | Signed Sep 2020, terminated Feb 2022 | 1.25bn prepayment forfeited; 1.35bn charge | Would have been the CPU architecture owner | Regulators blocked it. NVIDIA kept a 20-year license (`nvda-arm-termination`). | | Small deals | FY2022–FY2025 | 1,402 in total | Software and teams | Not separately measurable | | Groq (non-exclusive license and related consideration) | Dec 2025 (FY2026) | 13,000 paid in FY2026 + 3,921 accrued, about 16.9bn | Inference IP and staff | Goodwill rose from 5,188 to 20,832. Management credits Groq with raising revenue per gigawatt. No separable profit is disclosed. | | Other FY2026 acquisitions | FY2026 | 1,535 | Not itemized in the excerpts read | Not measurable | | Hugging Face (pending) | Agreed 2 Sep 2026; close expected H1 2027 | About 11,900 + up to 1,000 equity retention | Open AI-model platform | Commits to supporting other silicon vendors. No revenue disclosed (`nvda-hugging-face-8k`). | In aggregate, about 30bn of 2025–2026 deal and license spend buys ecosystem position and talent, not disclosed profit streams. A deal yield cannot be measured with matched definitions. Nothing indicates NVIDIA is a serial acquirer whose value comes from deal arithmetic. The underwriting treats Groq's remaining 986 and Hugging Face's 11,900 as year-1 cash claims with no separately credited earnings. Any benefit has to show up through the scenario margins. ## Current cash definitions and funding bridge These definitions feed the forecast. **Trailing free cash flow.** Reported trailing operating cash is 134,360. Subtracting 7,354 of capital purchases and 120 of related principal reproduces the issuer's 126,886 of free cash flow. That measure adds back 7,241 of stock compensation. A separately labeled owner-cash proxy deducts it, giving 119,645. This proxy and the forward operating-profit model are different measures, and the forecast does not deduct stock compensation twice. **Starting point.** Re-underwritten 3 Nov 2026. The baseline is the trailing four quarters to the 27 October anchor: reported Q4 FY2026, Q1 and Q2 FY2027 plus the Q3 guidance midpoint (108bn of sales, 74% gross margin, 9.2bn of opex). That gives 353,964 of revenue and 232,289 of GAAP operating income, a 65.6% margin. The earlier vintage annualized the Q3 guide to 432bn and then grew it 50%, counting the near-term ramp twice. Operating profit still excludes the large unrealized equity gains that boosted net income (`nvda-q2-fy2027-release`, `nvda-q4-fy2026-release`). **Opening liquidity.** July cash plus debt securities was 56,586. The pro forma bridge reserves: - 6,048 for the October dividend. - 1,500 for the SB Energy equity stake. - 1,100 for the KYEC advance. That leaves 47,938 of liquid funds against 33,500 of debt principal. No working-capital catch-up is reserved, because working capital starts at the reported July balance (75,002) and grows with forecast revenue. Year 1 also pays the Groq balance (986) and the Hugging Face consideration (11,900). Unreported later cash generation, buybacks and new financing are not invented (`nvda-supply-oct2026`, `nvda-hugging-face-8k`). **Equity commitments.** The full 25bn of reported conditional equity commitments is reserved in year 1, an accelerated funding assumption. The cost paid becomes a separately retained investment, subject to the scenario recovery haircut. Locked public equity, private equity and equity-method interests are not cash available to pay operating bills. **Supply and capacity commitments.** These are 279bn and mostly buy inputs for future sales, which margins, capex and working capital must fund. Subtracting every gross contract again as debt would count ordinary costs twice. Unused partner capacity and guarantees are different exposures and are tested in the downside as a cash loss. **Terminal value.** The terminal denominator is unlevered GAAP NOPAT. The valuation deducts net debt, adds separately retained investments and deducts other claims once. Issued debt keeps paying its estimated financing cost in annual cash flow. Dividends stay at USD1 per share a year; dividends already paid are excluded. ## Post-July financing bridge —13 November2026 The full2bn September Nscale subscription now has a separate first-year reserve, sourced from the counterparty's amended registration rather than an issuer headline. July reported cash and the25bn pool are not silently rolled to November. Funded cost joins retained financial assets once; no operating earnings or appreciation is credited. Exact overlap0–2bn remains a sensitivity to reconcile at17November accounts, when actual cash, holdings and remaining commitments must be rolled together. Nscale's preliminary losses are not measured mature project returns and its debt is not NVIDIA debt. # Investment thesis **Current assessment,15 November2026:** atUSD162.85, five-year base IRR is4.87%, weighted−0.71%; ten-year weighted3.72%. Operating paths, fixed exits and45%/45%/10% weights remain unchanged. The post-July Nscale reserve is paid once and retains only the existing case recovery. Customer funding and security/legal outcomes remain unresolved. [Current evidence](../updates/2026-11-15-daily-review.md); [13 November funding revision and fixed-price effect](../updates/2026-11-13-funding-revision.md); [prior probability revision](../updates/2026-11-12-daily-review.md). Historical tables below retain their original quote and weights. NVIDIA is an exceptional business at a cyclical peak. Its unit economics are extraordinary: about 190–225% after-tax return on tangible capital (`research/01-business.md`). The risk is not capital intensity. It is three things: - The price per unit of compute once the largest buyers' own chips mature. - A digestion once the buildout must be funded from customers' own cash rather than debt, equity and NVIDIA's balance sheet. - Margins that sit far above NVIDIA's own history. The 3 November re-underwrite starts from a guidance-inclusive four-quarter estimate to the 27 October anchor (three reported quarters plus unreported Q3 guidance): - **Revenue:** 354bn, using reported Q4 FY2026, Q1 and Q2 FY2027 plus the Q3 guidance midpoint. - **GAAP operating income:** 232bn, a 65.6% margin. This replaces the annualized Q3 run rate (432bn), to which the earlier vintage then added another 50% of growth. A sourced demand ceiling, built from customers' cash and financing, bounds every scenario. Mechanisms, the ceiling and the outside view are in `research/08-underwriting-plan.md`. | Input | Downside | Base | Upside | | --- | --- | --- | --- | | Year-1 growth against trailing 354bn | −15% | +28% | +58% | | Annual growth from year 2 | 3% | 5% | 8% | | GAAP operating margin, reached in year 3 | 30% | 50% | 58% | | Capital spending / sales | 6% | 4% | 3% | | Operating capital / sales | 30% | 22% | 18% | | Normalized tax rate | 23% | 20% | 17% | | Exit enterprise value / unlevered NOPAT | 13x | 18x | 22x | | Retained investment recovery | 40% | 80% | 100% | | Loss stress on partner capacity and guarantees | 25% | 0% | 0% | By year 5 these produce: | | Downside | Base | Upside | | --- | ---: | ---: | ---: | | Revenue (demand ceiling 844bn) | 339bn | 551bn | 761bn | | NOPAT (trailing about 193bn) | 78bn | 220bn | 366bn | At the USD217.94 quote of 27 October and the fixed 27 October 2031 exit: | | Downside | Base | Upside | Weighted | | --- | ---: | ---: | ---: | ---: | | 5-year XIRR | −24.5% | −1.2% | 12.9% | −4.4% (40/45/15) | | 10-year XIRR | −10.5% | 3.2% | 11.5% | 1.8% (40/45/15) | The weighted figure is the IRR of probability-weighted dated cash flows. The before/after comparison is in `updates/2026-11-03-reunderwrite.md`. These paths are not confidence intervals. ## Pre-mortem - **One-foot or three-foot hurdle?** Three-foot. The product economics are understandable. The value depends on post-2027 AI capital spending, custom-silicon substitution and AI-lab credit, none of which a filing can settle. - **Cyclical or secular?** Both. AI compute is a secular need, but NVIDIA's revenue is a capex cycle. NVIDIA had downturns in FY2009 (revenue −16%), FY2020 (−7%, operating income −25%) and FY2023 (operating income −58%). Today the customers' capex exceeds their own cash (`research/02-history.md`). - **How could it be a value trap?** By valuing shortage-era earnings as permanent. Those earnings could fall at the same moment the ecosystem stakes and guarantees lose value. - **Cigar butt or productive asset?** A productive, appreciating asset: R&D, CUDA and supply relationships, with about 79% of merchant and custom AI-accelerator revenue. - **Agency cost at the original underwriting date?** Pay was small beside the founder's then ~USD190bn stake. That historical alignment no longer describes the interim CEO's authority or incentives. Material in control of resources: revenue-weighted incentives alongside about USD99bn of stakes, USD25bn of commitments and USD108.5bn of guarantees (`research/03-management.md`). - **Good or bad execution?** Assume good product execution. Assume average-to-weak capital allocation: buybacks regardless of price, and lab investments justified by affordability rather than return. - **Turnaround?** No. - **Big promises with little record?** Partly. The 70% FY2028 outlook exceeds what the demand ceiling allows. Financed demand of about a quarter of the business has no record through a downturn. - **IP a larger competitor could squash?** No competitor is larger in AI compute. But each of the five largest customers is larger as a company, and Google has a complete alternative stack. - **Single product or fad?** Concentrated: about 90% of profit is one product cycle. Past network buildouts ended in digestions (Cisco −15%, Applied Materials −31% after 2001). ## Exit multiple **Own history.** A verified historical price and multiple series has not been reconstructed, and none is invented. Today: - Market value is about USD5.29tn (24,285m diluted shares at USD217.94). - Net of about 23bn of net liquidity and about 99bn of ecosystem stakes, enterprise value is about 5.17tn. - That is about 27x the trailing four-quarter NOPAT of 192.8bn. The estimated trailing figure includes the unreported Q3 guidance midpoint; it is not reported earnings. **Peers.** Peer multiples are not reconstructed; that is left open rather than guessed. **Koller check.** Value = NOPAT × (1 − g/RONIC)/(r − g). The required return r is 10% because the earnings are cyclical and concentrated. No case uses a peak multiple on peak earnings. | Case | Inputs | Implied multiple | Applied to | | --- | --- | ---: | --- | | Downside | g 3%, RONIC 40% | 13x | Trough-normalized earnings (30% margin). At zero growth, 1/r would give 10x. | | Base | g 5%, RONIC 50% | 18x | Normalized 50%-margin earnings | | Upside | Long-run g 6%, RONIC 60% | 22.5x, capped at 22x | Earnings that stay near peak margin | Sensitivities, for information: - Base at 22x: 2.1%. - Base with a 55% margin: 0.7%. - Downside at 11x: −26.4%. ## Bear case The downside is "the price we could sell at after five years if we weren't forced sellers". Its mechanism: 1. Customers' AI capex already outruns their cash (`research/02-history.md`). After the 2025–2026 build, financing tightens and buyers absorb capacity. 2. Revenue falls 15% below the trailing four quarters in year 1, about 30% below the Q3 run rate. That is the base-rate range for capex-cycle suppliers: Cisco −15%, Applied Materials −31%, Micron −49%, NVIDIA FY2009 −16%. 3. Revenue then recovers at 3% a year. 4. Hyperscalers move internal inference to their own chips. Financed labs stretch payments. NVIDIA defends volume with price. 5. Operating margin reverts within three years to 30%, NVIDIA's pre-boom average. 6. Working capital rises to 30% of sales. 7. Equity stakes recover 40% of carrying value. 8. 25% of the 164.5bn of guarantees and partner capacity is paid in year 3. 9. The exit is 13x trough-normalized NOPAT. Five-year XIRR: −24.5%. Tails it omits: - **A Micron-scale −49% collapse.** A −30% first year gives −27.4% (information only). - Equity stakes going to zero. - Guarantee calls above 25%. - A Taiwan supply disruption. - Revenue failing to recover at all. ## Why the opportunity exists **Evidence on attention.** - **Size:** NVIDIA is among the largest listed companies (about USD5.3tn) and a member of the S&P 500. - **Ownership:** BlackRock (7.43%) and Vanguard (7.31%) are the largest holders; insiders own 3.94% (`nvda-proxy-2026`). - **Coverage:** essentially every large broker covers it. - **Not measured:** daily trading value, short interest and retail ownership were not retrieved. Neglect is impossible. **Recent events that drove sellers.** - The August gross-margin reset. - The debate about "circular financing". - Custom-silicon announcements. These are real risks, not noise. ## Variant view at the original USD217.94 quote **No mispricing in our favour; if anything, the price assumes more than customers can fund.** At USD217.94: - The base earns about −1% a year for five years. - The probability-weighted return is about −4%. - A 12% base return would need a price near USD118, and a 12% weighted return about USD100. The market appears to price NVIDIA's current margin and growth as durable, or to expect customers' AI spending to keep exceeding their own cash for years. This research concludes otherwise: - Customers' cash plus observed financing caps year-1 revenue near 576bn (stage 8). - Past capex-cycle suppliers, NVIDIA included, have had digestions (outside view). - NVIDIA's own pre-boom margins were 27% operating and 62% gross (stage 7). At that historical quote the business quality did not provide a sufficient margin of safety. The current ranked comparison and probabilities are in the 12 November review; this historical comparison is not a current quote or return. **Catalyst.** None hard. The Q3 FY2027 report on 17 November 2026 is an evidence date. ## What would prove us wrong | Observation | Refutes | Where and when | | --- | --- | --- | | Hyperscaler free cash flow turning positive while capex keeps rising more than 30% a year | The demand ceiling, which would be too low | Alphabet, Amazon, Meta and Microsoft quarterly releases from Q4 2026 | | FY2028 revenue above about 576bn with operating margin above 60% | The base's customer-funded bound; would support the upside | FY2028 results, about February 2028 | | DSO back below 50 days with revenue still growing over 50% | The financing-dependence concern | Q3 FY2027 (17 Nov 2026) and later CFO commentary | | Gross margin above 70% through FY2029 despite custom-silicon growth | The base's margin normalization | Quarterly releases | | Hyperscale share of Data Center falling below 45%, external TPU and Trainium sales in the tens of billions, or guarantee calls and stake impairments | Confirms the downside rather than refuting it | Alphabet and Amazon 2027 filings; NVIDIA 10-Q notes 5–10 | `valuation/prospective-checks.json` was frozen on 3 November against the earlier model hash (`f5ae5c50024e`). It stays unedited, so its predictions also score the superseded vintage. Permanent-loss paths and unverified inputs are in `research/06-risks.md`. ## Judged view — 5 November 2026 Low in precise forward returns; platform advantage supported, customer-paid demand and durable margins unverified. Retain existing coarse horizon-specific weights after reassessment: scenarios already represent these adverse/favorable operating mechanisms. Numerical odds are subjective, not verified frequencies; unresolved gaps prevent high conviction. No tuning to return. The current dated scenario returns and the evidence for each changed or retained input are in [the independent bull–bear judgment](12-debate.md). This review supersedes earlier undated/current return statements above; dated historical/source examples remain calibration. 17 November Q3 actuals replace guidance-inclusive starting estimate; reconcile customer-funded purchases, RPO/cash, receivables, guarantees and utilization. # Risks and disconfirming evidence **15 November update.** Original class-action complaint pp.1–3/17 and derivative complaint pp.1–3/33–34 were inspected: the former seeks damages against NVIDIA and Huang; the latter seeks recovery for NVIDIA and governance changes. Allegations and requested remedies are not findings or a quantified net company loss. The Insurer's14November report describes executive Side A cover; the actual policies were not available and corporate reimbursement is not presumed. Megaspeed's emailed statement is reported by newswires, not independently verified clearance. Current odds remain45/45/10. [Evidence and limits](../updates/2026-11-15-daily-review.md). Dated entries below preserve earlier assessments. **Demand and customer funding.** Large infrastructure orders can precede profitable use. Three direct customers represented 16%, 15% and 13% of H1 sales; direct and end customers are not identical. H1 receivables absorbed USD24.590bn and inventory USD10.204bn of cash. Extended payment terms and financially weaker clouds make conversion and credit quality central to valuation. Growth in token usage is insufficient without paid, collectible demand. **Platform and competitive risk.** CUDA and system integration are advantages, but AMD, custom accelerators and software portability can reduce pricing or share, particularly for repeatable inference workloads. AMD's August release describes Helios/MI400, ROCm and named cloud/customer partnerships. Those are competitive claims, not independent proof of superior economics. [AMD Q2 issuer results](https://ir.amd.com/financial-information/sec-filings/content/0000002488-26-000121/q22026991.htm). **Margins and supply.** Memory, packaging, networking and faster product transitions create cost, quality and inventory risk. Supplier advances and USD279bn supply/capacity commitments are real economic capital despite external manufacturing. Export restrictions already caused product charges; the Q3 outlook excludes China data-center compute. The scenarios do not assume restoration of a full China market. **Ecosystem correlation.** July equity marks were about USD99bn. Supply and deployment are increasingly supported by investments and guarantees. Maximum phased guarantees of USD108.5bn and USD56bn of partner cloud/third-party lease arrangements are not immediate debt, but can become cash losses. The downside pays a chosen USD41.125bn stress in year 3. This is not an estimated default probability or full worst-case exposure. Optional further support is not exercised. **Customer/supplier checks.** Microsoft's July discussion reported substantial CPU/GPU capex alongside growing cloud cash flow. TSMC's October report described strong leading-edge demand and further capacity spending. These sources corroborate current activity while also showing the industry's funding needs. They do not prove NVIDIA's future share, margins or return on customer investment. [Microsoft](https://www.microsoft.com/en-us/investor/events/fy-2026/earnings-fy-2026-q4), [TSMC](https://investor.tsmc.com/schinese/encrypt/files/encrypt_file/reports/2026-10/eba4fac72873bf546b669bd2b5d973df5c93215d/TSMC%203Q26%20Transcript.pdf). Coverage does not establish actual utilization or profitability for every model maker or cloud. Private contracts, later cash movements and unquantified supply advances remain unresolved. Reported investment marks may also be difficult to realize, particularly when operating demand weakens at the same time. ## Permanent-loss paths (reviewed 3 Nov 2026) 1. **Substitution at the largest buyers.** Hyperscalers are about 55% of Data Center revenue and each builds its own accelerators. Google now sells TPU systems externally to nearly 100 customers, Amazon will sell Trainium outside AWS from late 2027, and Broadcom's AI revenue grew 221% (`googl-q3-2026`, `amzn-q3-2026`, `avgo-q3-fy2026`). If internal inference moves off NVIDIA, price per gigawatt falls, not just volume. That is the downside's permanent-margin mechanism. 2. **Financed demand.** AI labs that NVIDIA supports with equity and credit are about a quarter of next year's business (`aiLabFinancedShare`). If their revenue or funding disappoints, NVIDIA loses sales, receivables (DSO already 60 days), equity value (about 99bn carrying) and guarantee exposure (108.5bn maximum) at once. These losses are correlated, not diversified. 3. **Customer cash, not supply, as the binding constraint.** Alphabet, Amazon, Meta and Oracle had negative free cash flow in their latest quarters while capex roughly doubled. Continued growth needs open debt and equity markets. 4. **Supply commitments in a downturn.** 279bn of commitments, mostly memory and concentrated in the next three fiscal years, become onerous if demand pauses. In FY2026, H20 provisions alone were 4.5bn. 5. **Regulation.** Export rules removed China data-center compute. A Taiwan supply disruption is an omitted tail. 6. **Capital allocation.** Buybacks of about USD20bn a quarter regardless of price, plus deal and license spend of about 30bn (Groq, Hugging Face) with no measurable profit. ## Unverified inputs - **Profit split by economic unit** (hyperscale against ACIE). Estimated from revenue; NVIDIA does not disclose it. - **Customer-side gigawatt economics.** USD60bn per GW is a management figure. Power cost and economic life are this repository's assumptions. No neocloud or lab P&L has been reconciled. - **Oracle's FY2027 capex guide and Broadcom's FY2027/FY2028 AI-revenue and gigawatt remarks.** These come from third-party call summaries and automated transcripts. - **The FY2026 sources-and-uses residual** (2,987) is not traced. - **NVIDIA's own historical valuation multiples, peer multiples, trading value and short interest.** Not retrieved. - **The terms of the lab stakes**, including whether NVIDIA's equity carries preferences, and the exact funding dates of the 25bn of commitments. - **No fieldwork.** No interviews or channel checks were done. ## Surviving debate attacks — 5 November 2026 Low in precise forward returns; platform advantage supported, customer-paid demand and durable margins unverified. Reconstructed the independent isolated stress at the same quote/date/horizons; do not treat stress size as a probability or a measured current fact. Discriminating evidence: 17 November Q3 actuals replace guidance-inclusive starting estimate; reconcile customer-funded purchases, RPO/cash, receivables, guarantees and utilization. The [complete debate](12-debate.md) sizes these risks using the shared forecast; a sensitivity is neither a measured fact nor a calibrated probability. ## External scrutiny,6November2026 The [House committee chairs’5November release](https://foreignaffairs.house.gov/news/press-releases/chairmen-mast-moolenaar-and-crawford-request-jensen-huang-brief-congress-on-link-to-chinese-intelligence-asset) confirms a request for Jensen Huang to discuss Megaspeed contacts and information-security controls. Reply requested11November; briefing requested18November. Attendance and the underlying allegations remain unconfirmed in the checked official sources. External official pages and dates are in `researchScope.externalDisclosureWatch`. **8 November: issuer inquiry confirmed.** The [board's full statement](https://nvidianews.nvidia.com/news/statement-from-nvidia-board-of-directors), paragraphs1-5, establishes a new independent-director/counsel investigation. It also states continued confidence in management and no present reason to believe national security or company secrets were compromised. The [CEO's response](https://nvidianews.nvidia.com/news/a-message-to-our-company) commits to cooperation. An inquiry is a material governance development, not a finding or an all-clear. Potential cash mechanisms are leadership disruption, compromised intellectual property, customer/partner loss, wider export restrictions and legal/remediation costs. Neither statement quantifies those effects or announces a new restriction or management replacement. Keep operating inputs and both horizons'40/45/15 weights after explicit reassessment; do not invent a loss amount or a precise additional event probability. China data-centre compute remains excluded, but indirect routing, non-China restrictions and information-security damage are separate risks. The three operating paths do not bound this tail. Revisit the whole distribution when findings, actual contract/cash changes or a leadership decision supply discriminating evidence. [Dated assessment](../updates/2026-11-08-daily-review.md). ## 9 November follow-up The [dated review](../updates/2026-11-09-daily-review.md) records the filed-suit notice and sources distinguishing allegations, requests and findings. The complaint/docket was not directly reviewed. No new restriction, leadership decision or quantified company loss is established. Keep the existing operating paths and explicit unquantified tail; reassess on discriminating cash/contract or governance evidence. ## 11 November evening follow-up [Evening evidence and assessment](../updates/2026-11-11-evening-review.md) distinguish reported conditional congressional escalation from an issued demand or investigative finding. Retain the unquantified governance tail and revise the distribution when evidence changes cash, contracts, leadership or restrictions. ## 11 November late follow-up Reuters reports that the board met late on 11 November to decide whether to remove the CEO. It attributes the preparations to the WSJ, a second relationship to the NYT, and a replacement search to Bloomberg. Nothing is confirmed by the issuer. Leadership continuity is now an active, near-term risk rather than a remote tail. Its cash channels are execution cadence, talent retention, customer/government relationships and remediation cost; none is yet observed. [Dated assessment](../updates/2026-11-11-late-review.md). ## 12 November confirmed transition The [8-K, Item 5.02](https://www.sec.gov/Archives/edgar/data/1045810/000104581026000082/nvda-20261111.htm) resolves the prior uncertainty about a leadership decision. It does not resolve the investigation, securities allegations or operating consequences. Interim continuity and permanent succession remain distinct questions. Both horizons now use 45%/45%/10%; annual drivers and 13x/18x/22x exits remain unchanged to avoid a second arbitrary charge for the same uncertainty. The [review](../updates/2026-11-12-daily-review.md) records potential coordination, talent, product and customer mechanisms plus unquantified legal/security tails. Reported acceptance of a congressional briefing is not confirmation of the originally requested 18 November date. ## 13 November customer-funding and governance evidence Nscale's actual financing needs must be separated from NVIDIA's July pool and from the operating demand ceiling. The extra2bn reserve is paid once and receives only the existing scenario asset recovery. Whether an earlier commitment overlaps is a0–2bn cash-reserve range, not an invented probability. Project funding for future service agreements remains uncommitted; loss-making expansion can create value, but its funding and mature project return need proof. The Singapore MHA statement and US Senate Finance preservation letter are primary process evidence, not findings of spying, a new export ban, damages or issuer clearance. House acceptance/subpoena status was unconfirmed before its8pm EST13November deadline. Reported Senate December3/10 requests and potential CEO candidates remain unconfirmed in original committee/issuer notices. Recheck those originals; keep requests distinct from issued process and permanent appointments. # Assumption register Reported facts retain periods, units and source IDs in `financials/facts.json`. `valuation/calculations.json` connects those facts and explicit judgments to the model. The arithmetic is shared; the choice of definitions and inputs follows NVIDIA's business. There is no supplied external NVDA investor forecast to reproduce. **9 November review.** The [dated assessment](../updates/2026-11-09-daily-review.md) adds a filed shareholder suit and unpriced investment announcements to the source record. No new financial report, quantified loss, binding restriction or operating interruption sizes a numerical revision. All financial inputs and both horizons’40/45/15 weights are retained with explicit tail limits; assessment evidence/date change. Unspecified investment amounts are neither free nor added twice to the existing reserve. | Input or definition | Evidence and calculation | Judgment and change trigger | | --- | --- | --- | | Trailing revenue | FY2026 + H1 FY2027 - H1 FY2026 = USD302,970m. | Preserve fiscal dates. Do not relabel a forecast as reported trailing sales. | | Reported FCF and owner cash | Operating cash 134,360 - capital cash 7,474 = 126,886; less recurring SBC 7,241 = 119,645. | Owner-cash proxy differs from issuer FCF. Forward GAAP profit already charges SBC, so no duplicate subtraction. | | Starting point (re-underwritten 3 Nov 2026) | Trailing four quarters to the 27 Oct anchor: TTM revenue 302,970 − Q3 FY2026 57,006 + Q3 FY2027 guide 108,000 = 353,964. Operating income 197,579 − 36,010 + (108,000 × 74% − 9,200) = 232,289 (`revenueTrailingToAnchor`, `operatingProfitTrailingToAnchor`). | Before: annualized Q3 guide 432,000 / 282,880. Year-1 growth was then added to an already-annualized peak quarter, counting near-term growth twice. The trailing window ends at the anchor, so year-1 growth applies once. Revise when Q3 actuals replace the guide (17 Nov 2026). | | Revenue path | Bottom-up year 1: hyperscalers 20% of a 1,065 cash-plus-financing budget (213); ACIE about 6 GW × 35 (210); Edge 30. Total 453 = +28%. Downside sized from capex-cycle declines (Cisco −15%, Applied Materials −31%, Micron −49%, NVIDIA FY2009 −16%). | Year 1 −15% / +28% / +58%, then 3% / 5% / 8% (`growthFadeYears` 2). Before: +10/+50/+70% fading to 0/5/10% over three years from 432,000. Every year must stay under the demand ceiling; the upside year 1 (559) sits just under 576. Management’s +70% FY2028 (about 690) exceeds the ceiling and is not used. | | Operating margin | Pre-boom FY2016–FY2022 operating margin averaged about 27% (range 15–37%). Gross margin averaged 62.1% over FY2018–FY2022. Opex is now about 9–12% of sales against 25–35% pre-boom. Management guides 72–73% gross margin for FY2028. | Normalize within three years (`marginTransitionYears` 3) to 30% / 50% / 58%. Before: 35/55/65% over five years. Base = pre-boom gross margin minus opex at today’s scale. Downside = pre-boom average. Revise on sustained gross margin above 70% or opex intensity shifts. | | Operating capital | Reported July proxy 75,002 (core receivables, inventory and prepaids less operating accruals, plus residual noncurrent operating assets less long deferred revenue). | Starts at the reported balance (before: scaled to the 432,000 run rate with a 9,181 catch-up reserved from cash). Then 30% / 22% / 18% of revenue. | | Capex and D&A | TTM D&A / sales about 1.22%; historical capital spending about 2.47% of sales. Contracted capex is USD8bn. | Spend 6/4/3% of sales on infrastructure and recurring PP&E/intangible capital, above historical intensity. Do not equate accounting D&A with maintenance or subtract the same contracted spending twice. | | Taxes and financing | FY tax guide 16–18%; June debt issuance and contractual rates are disclosed. | Start at 17%, transition to 23/20/17%; estimate 4.5% continuing debt cost, refinancing and no new net borrowing. Carry the historical noncurrent tax liability separately. | | Cash available | July cash and debt securities 56,586 less October dividend 6,048, SB 1,500 and KYEC 1,100 = 47,938 (`proformaOpeningCashTrailing`). | No catch-up reserve with a trailing start. Not an observed October balance; later cash generation and buybacks are not invented. | | Supply and ordinary operations | Supply/capacity 279,000; own R&D cloud 29,000 and uncommenced own leases 25,000. | Fund product inputs through margins and capital, not by subtracting all contracts as extra debt. Unused capacity/cancellations can still require revision. | | Equity commitments | Reported conditional equity commitments 25,000. | Reserve all in year 1 and add funded cost to retained assets. Conditional does not mean already paid. Future follow-on investments are not automatically free. | | Post-July Nscale subscription (13 November review) | Counterparty S-1:September15 NVIDIA subscription1,000 plus further1,000 expected around16November. | Reserve2,000 once in year1 outside theJuly26 pool; funded cost enters the existing40/80/100% retained-value rule. No operating earnings or note income. Pool overlap0–2,000 remains a sensitivity; reconcile current cash, held securities and remaining commitment together at17November accounts. | | Anthropic subscription overlap (15 November review) | November10 S-1 pp.23/347/F-67: conditional IPO subscription10,000. The November18,2025 original partnership already announced up to10,000 before the July pool. | Provisionally treat this as fulfillment within the25,000 pool; composition is unitemized, so inclusion is a judgment. Extra0–10,000 funding sensitivity changes5y weighted IRR by0 to−0.01922pp. Do not add a second liability merely from a later agreement. Reconcile cash/holdings/remaining commitments at17November; no IPO proceeds or valuation gain assumed. | | Retained securities | Public equity including locked noncurrent holdings plus nonmarketable/equity-method securities = 98,897, plus the separate SB reserve. | Recover 40/80/100% of marks and funded cost; no appreciation or income. This is a valuation stress, not a known private investment exit price. Disposal taxes/restrictions can reduce value. | | Contingent partner exposure | Guarantees 108,500 plus partner cloud/third-party leases 56,000. | Base/upside no payment; downside chosen 25% loss paid once in year 3. No claim of zero risk or common default probability. | | Shares and distributions | Latest quarter weighted-average diluted shares 24,285m; declared dividend USD0.25 quarterly. | Hold net shares flat, leave economic SBC cost charged and give no net buyback accretion. USD1 annual dividend on approximate future dates. Actual grants/settlements/repurchases require later reconciliation. | | Terminal valuation | Koller (1 − g/RONIC)/(r − g) at r 10%: downside g 3%, RONIC 40% → 13x; base g 5%, RONIC 50% → 18x; upside g 6%, RONIC 60% → 22.5x, capped at 22x. | 13 / 18 / 22x on unlevered GAAP NOPAT (before: 15 / 22 / 28x). No peak multiple on peak earnings. Net cash and retained investments enter once. | | Hugging Face consideration (added 3 Nov 2026) | 8-K of 3 Sep 2026: about 11,900 to stockholders plus up to 1,000 equity retention (`huggingFaceConsideration`). | Paid in year 1 as a cash claim, alongside Groq's remaining 986. No acquired earnings are credited because no revenue is disclosed. The retention equity is future stock compensation inside GAAP opex. Revise on close, a price adjustment or a failed approval. The 27 Oct vintage omitted this. | | Cycle and windfall normalization | Q2 FY2027 margins are 75% gross and 66% operating, against FY2022's 64.9% and 37.3% and FY2023's 56.9% and 15.7%. Management guides gross margin to 71–72% in Q4 and 72–73% in FY2028 on memory costs. | The opening run rate is a shortage-era margin. Operating margins of 30%, 50% and 58%, reached in year 3, are the through-cycle judgment (revised 3 Nov from 35/55/65% over five years). The downside's reversal is charged as permanent substitution. Do not raise the base margin on one strong quarter. | | Historical capital-allocation measurements | Five-year rates per Δsales: M&A 8.0%, R&D 26.4% (7.3% step-up only), fixed capital 2.1%, working capital 15.2%, equity stakes 10.0%. Reinvestment 22.5%; lagged ROIIC 472% (`calculations.json` steps). | Diagnostics only. They show that tangible capital does not drive the earnings, so no ROIIC-based grower is used. They are not bound to model inputs. | | Supply prepayments and extended terms | Other noncurrent operating assets about 9,965; DSO 60 days; extended terms with investment-grade customers. | Inside the operating-capital proxy (22% of sales in the base, 30% in the downside). A move back to about 50 days DSO would release about 12bn at the run rate. Not credited. | | Demand ceiling (added 3 Nov 2026) | Five hyperscalers’ operating cash 793,016 (annualized latest quarter; Amazon TTM) + observed financing 272,408 (Amazon debt + Oracle equity, × 4) = 1,065,424. × 25% maximum share (observed 21.0%) = 266,356; + ACIE 7 GW × 40,000 = 280,000; + Edge 30,000. Total 576,356, growing 10% a year (`demandCeilingYear1`, `ceilingGrowth`). | Bound to `model.demandCeiling`; the engine rejects any scenario above it. Revise when hyperscaler cash flow or financing pace changes materially, or when NVIDIA’s observed share of their capex moves above 25%. | | Outside view | NVIDIA downturns FY2009, FY2020 and FY2023; Cisco FY2002, Applied Materials FY2002, Micron FY2023 (primary filings). | Five-year revenue CAGRs −0.9% / +9.2% / +16.6% from the trailing 354bn. Generic large-company base rates are pending (no sourced copy of the Base Rate Book). | The five-year exit is 27 October 2031. The 3 November re-underwrite is recorded in `updates/2026-11-03-reunderwrite.md`. Every changed estimate requires a dated old/new value, evidence, economic explanation and return effect isolated from stock-price/calendar movement. Scenarios are not probabilities. The decade extension does not constitute ten years of product-cycle research. 11 November evening: [governance evidence reassessed](../updates/2026-11-11-evening-review.md). No supported numerical cash, margin, funding or ownership revision; all annual drivers and both horizons' weights retained. The pending process is not a quantified loss estimate. 11 November late: reported board deliberation on CEO removal. Not yet an issuer fact; annual drivers retained under review. Revision trigger: an issuer-announced departure, interim CEO or reduced role. [Assessment](../updates/2026-11-11-late-review.md). ## Current probability revision — 12 November 2026 | Input | Before | Current | Evidence and judgment | | --- | --- | --- | --- | | Five-year downside/base/upside weights | 40%/45%/15% | 45%/45%/10% | Filed succession decision, Item 5.02; near-ceiling upside needs uninterrupted coordination as well as funded demand. Interim leadership supports retaining the base, while a permanent successor remains unknown. | | Ten-year downside/base/upside weights | 40%/45%/15% | 45%/45%/10% | The same early disruption can have lasting product/share and capital-allocation effects; no automatic decade recovery is credited. Separate long-horizon assessment, not a measured transition frequency. | | Annual financial paths, commitments and terminal valuation | Existing source-linked paths and 13x/18x/22x endpoints | Retained | No quantified order, margin, compensation or funding change disclosed. No separate arbitrary key-person haircut. | The [dated review](../updates/2026-11-12-daily-review.md) and [change record](../updates/2026-11-12-probability-change.json) retain old inputs and same-price/date effects. Reassess on the permanent successor, technical-management structure, actual orders, product cadence and 17 November accounts. Unknown loss amounts remain unknown; three paths do not exhaust possible outcomes. # Underwriting plan Written 3 Nov 2026 and revised the same day after the owner challenged the first version as extrapolation (`updates/2026-11-03-reunderwrite.md`). The model vintage stays anchored at 27 Oct 2026, with exits fixed at 27 Oct 2031 and 2036. Money is in USD bn unless stated. Rolling forecast years run October to October; NVIDIA's fiscal years end in late January. ## Opportunity ledger | # | Opportunity | Starting point (evidence) | Why NVIDIA wins / customer need | Reachable scale five years out (estimate) | Investment and cash cost | Timing and limits | Evidence against | | --- | --- | --- | --- | --- | --- | --- | --- | | 1 | **Volume:** hyperscaler AI systems (Vera Rubin, then its successor) | Hyperscale 48.7 in Q2, about 195 annualized (`hyperscaleRevenueQ2`). Top-five capex about 810–830 in 2026 (company guidance), "significantly" or "meaningfully" higher in 2027 (`googl-q3-2026`, `amzn-q3-2026`). | Rented cloud capacity has to run what cloud customers' code targets (CUDA). Rubin has purchase orders from every major hyperscaler (`nvda-q2-fy2027-call` p.4). | The five's own cash plus observed financing is about 1,065 in year 1. At the observed 21% share that is about 225; in the base, 20% (213); ceiling 25% (266). Management's 1.3tn for 2027 exceeds their cash plus financing. | Working capital about 20–22% of incremental sales; supply commitments (279, mostly memory) paid through cost of goods. Ceiling: 25% of the hyperscalers' cash-plus-financing budget, about 266 in year 1. | Supply-constrained through FY2028 by management's account. Limited by memory and CoWoS capacity and by customers' power and shells. | Internal inference moving to TPU, Trainium, Maia and MTIA. Broadcom's AI revenue +221%. Microsoft cut its 2026 capex estimate from about 190 to about 170. Customers' free cash flow is negative. | | 2 | **New customers:** ACIE (neoclouds, sovereigns, enterprises) | About 40 in Q2, +138% year on year. Neocloud installed base 3 → 8 GW during 2026 (p.3). Sovereign +35% sequentially (p.5). | These customers lack their own chips. NVIDIA supplies reference designs (DSX) and financing partners. | Ceiling: at most 7 GW added in year 1 × 40 per GW = 280. Base: about 6 GW at a blended 35 per GW = 210. | Extended terms (DSO 60); guarantees of 3.5 for neocloud land, power and shells; equity-method stakes in financiers (3.3). | Bound by the availability of debt and equity for neoclouds and by power. | Highly sensitive to credit markets. Neocloud returns are unmeasured here. | | 3 | **Financed AI labs** (OpenAI, a second frontier lab, others) | About 25% of next year's business (`aiLabFinancedShare`). OpenAI commitments about 12 GW (p.6). | Labs have demand but lack investment-grade balance sheets, and NVIDIA supplies the credit. | 12 GW × 40 per GW is up to about 480 of content through 2030, if delivered on NVIDIA systems at Rubin pricing. That is an upper bound, not a forecast. | Equity: nearly 50 invested plus 25 committed. Guarantees: 105 (SB Energy) plus credit enhancement for about 2 GW. Shareholders bear these costs. | Funding phases through 2030; SB Energy phases from FY2029. | OpenAI's own XPU (1.3 GW in 2027 per Broadcom's call, unverified) and Anthropic's TPU, Trainium and AMD purchases. Revenue that depends on NVIDIA's own credit. | | 4 | **Price and content per gigawatt** | Management sizes content at about 18 (Hopper), 25 (Blackwell) and 40 (Rubin) per GW, including the Vera CPU, NVLink and networking (pp.4 and 10). | Each generation delivers more tokens per watt, which justifies price per GW. | Already in the Rubin figure. Successors are not assumed to raise content further in the base. | Memory costs: gross margin down to 71–72% in Q4 and 72–73% in FY2028 (p.7). | The margin reset is already visible. | Memory suppliers taking value; competitors' tokens-per-watt claims (Maia 300 +40%). | | 5 | **Replacement of the installed base** | Hopper, Blackwell and Rubin installed across hyperscalers and neoclouds. Prior generations still rent at a premium (`amzn-q3-2026`). | Annual cadence plus power limits: replacing old GPUs is how a power-capped site grows output. | Supports the mature growth rate from year 2: 3%, 5% and 8% by scenario. | Ordinary working capital. | Starts to matter from about 2029 as 2025–2026 fleets age (five-to-six-year accounting lives). | Customers stretching useful lives; Amazon is "finding ways to extend" them. | | 6 | Networking and CPUs | Networking 31.4 in FY2026 (+142%). Management sees about 20 of server-CPU demand for Vera. | Attached to systems | Included in #1–#4; not added separately, to avoid double counting. | Same as above | | Ethernet competition (Broadcom) | | 7 | China data-center compute | Excluded from guidance. H20 and H200 shipments are under 1% of Data Center. | | **Zero in every scenario** | | Depends on regulation | Export rules | | 8 | Edge: gaming, workstations, automotive | 22.5 of Graphics segment revenue in FY2026 | Mature franchise | Low single-digit growth, inside total revenue | | | Memory prices weigh on PCs | | 9 | Acquisitions (Hugging Face, Groq balance) | 11.9 + 1.0 retention; Groq 0.986 outstanding | Ecosystem reach | **No separate earnings credited** | Cash paid in year 1 | Close expected H1 2027 | Regulatory approval | **Management targets as a bound.** The roughly 70% FY2028 outlook (February 2027 to January 2028) is a supply-constrained planning number that implies FY2028 revenue of about 690. The demand ceiling built below caps year-1 revenue at 576. Management's plan therefore requires hyperscaler capex beyond their own cash plus the financing pace observed so far, or a larger NVIDIA share than today's. Management cites 1.3tn of top-five capex in 2027, but the companies themselves give no 2027 number. Executives' ranges are overprecise, and here the plan exceeds what customers' budgets fund. So the base is set from the bottom-up customer budget, not from the plan. ## 12 November opportunity reassessment The [filed succession decision](https://www.sec.gov/Archives/edgar/data/1045810/000104581026000082/nvda-20261111.htm), Item 5.02, changes who coordinates opportunities 1–6, not their measured customer budgets or installed base. In hyperscale design wins and Rubin/system content, slower architecture and supply decisions could concede lasting share. In financed labs, a changed approval process could reduce risky deployment as well as delay sales; an assumed improvement in lending discipline is not credited without actual decisions. Existing CUDA users, product teams and funded orders support continuity. Reachable units, margins, capital needs and ceilings are retained pending operating evidence. The joint uninterrupted upside therefore loses five percentage points to downside, while the customer-funded base keeps 45%. See the [dated impact audit](../updates/2026-11-12-daily-review.md). This judgment was recorded before inspecting the revised return. ## Demand ceiling Built in `valuation/calculations.json` (steps `hyperscalerOcf` to `demandCeilingYear1`) and bound to `model.demandCeiling`. The engine rejects any year above it. | Component | Year 1 | Evidence | | --- | ---: | --- | | Hyperscalers' own operating cash, annualized | 793 | Microsoft 58.0 (Q1 FY27) × 4; Alphabet 42.3 × 4; Meta 32.8 × 4; Oracle 23.1 × 4 (Jun–Aug); Amazon TTM 168.3. Alphabet, Meta and Oracle are derived as free cash flow plus capex. | | External financing at the observed pace | 272 | Amazon 48.1 of Q3 debt plus Oracle 20.0 of Q1 equity, × 4. Microsoft repaid debt. | | Hyperscaler capex budget | 1,065 | Their latest-quarter capex is already 231 (926 annualized), above their own cash | | × largest share reaching NVIDIA | 25% | Observed 21%: Q2 hyperscale revenue 48.7 / next-quarter capex 231. Plus up to 4 points for CPU and networking attach. TPU, Trainium, Maia, MTIA, buildings, power and storage take the rest. | | Hyperscaler ceiling | 266 | | | Non-hyperscaler capacity × full Rubin content | 280 | At most 7 GW in year 1 (5 GW added in 2026, p.3) × 40 per GW. These buyers do not design their own chips. | | Edge | 30 | Q2 run rate 7.2 × 4 | | **Ceiling, year 1** | **576** | Grows 10% a year: customers' operating cash about +15%, with financing flat in absolute terms. Year 5: 844; year 10: 1,359. | The ceiling is an upper bound, not a forecast. It uses the full share in every component at once. ## Outside view Generic large-company base rates (sales growth and margin distributions by size bucket) are **pending**. A sourced copy of the Base Rate Book could not be obtained, so no remembered percentiles are used. The comparison below uses only primary filings. **NVIDIA's own downturns** (10-K values via `nvda-xbrl-companyfacts`): | Downturn | Revenue | Operating income | | --- | --- | --- | | FY2009 | 4,098 → 3,425m (−16%), then −3% in FY2010 | 836m → −71m | | FY2020 | 11,716 → 10,918m (−7%) | 3,804 → 2,846m (−25%) | | FY2023 | Flat | 10,041 → 4,224m (−58%); margin 37% → 16% | That is three downturns in about fifteen years. Pre-boom operating margins (FY2016–FY2022) ranged from 15% to 37% and averaged about 27%. Gross margin averaged 62.1% over FY2018–FY2022. **Other suppliers into customer capex peaks:** | Supplier | Peak year | Revenue decline | Source | | --- | --- | --- | --- | | Cisco | FY2001 → FY2002 | 22.3bn → 18.9bn (−15%) | `csco-fy2002-results` | | Applied Materials | FY2001 → FY2002 | 7.3bn → 5.1bn (−31%) | `amat-fy2002-10k` | | Micron | FY2022 → FY2023 | 30.8bn → 15.5bn (−49%) | `mu-fy2023-results` | **This model's paths against that record.** Five-year revenue CAGRs run from the trailing 354; margins are GAAP operating. | Path | Revenue path | 5-year revenue CAGR | Margin path | 5-year NOPAT CAGR (from 192.8) | | --- | --- | ---: | --- | ---: | | Downside | Year 1 −15% against trailing (about −30% against the Q3 run rate), then +3% | −0.9% | 65.6% → 30% in three years | −16.5% | | Base | +28%, then 5% | +9.2% | → 50% | +2.7% | | Upside | +58%, then 8% | +16.6% | → 58% | +13.7% | The downside's decline sits between Cisco's and Applied Materials'. Its margin is NVIDIA's own pre-boom average. The base still grows revenue 9% a year from an already enormous base. That is plausible only because year 1 completes orders already placed (Rubin purchase orders from every hyperscaler) and later years grow at half the ceiling's rate. Its margin is far above NVIDIA's history. The named mechanism: scale has cut opex from about 25–35% of sales before the boom to about 9–12%, while gross margin returns to the pre-boom 62%. The upside's 16.6% revenue CAGR at this scale has no precedent among the suppliers above. It is allowed only because it stays inside the customer-funded ceiling, and it gets 15% weight. **Humility check.** How would a disciplined value investor underwrite a supplier at a cyclical peak? This repository's method (`docs/method/growth-opportunities.md`, stage 7 and stage 9) answers it: - **Normalize the margin.** Do not capitalize the peak. - **Bound revenue by customers' funds.** Not by management plans. - **Assume that at least one digestion occurs within the horizon.** That is what the three NVIDIA downturns and the other suppliers' records show. - **Price the exit on normalized earnings at a multiple the Koller identity earns.** - **Then ask whether today's price leaves a margin of safety even in the base.** Done that way, the base earns about −1% a year for five years at USD217.94. A 12% base return would need a price near USD118, and a 12% weighted return about USD100. When the critical forecast (post-2027 AI capex) cannot be known, the method says to widen the range or decline to underwrite, not to fit a growth rate to the price. The honest finding is that the price assumes far more than customers can currently fund. ## Scenario mechanisms The three cases are coherent business outcomes. Inputs and their `inputNotes` are in `valuation/model.json`; the arithmetic is in `valuation/calculations.json`. Year 1 carries the cycle position, and from year 2 revenue grows at the mature rate. Margins normalize within three years. | Driver | Downside | Base | Upside | | --- | --- | --- | --- | | Mechanism | **Digestion.** Customers' capex already exceeds their cash, so after the 2025–2026 build financing tightens and buyers absorb installed capacity. Hyperscalers move internal inference to their own chips. Financed labs slow purchases. NVIDIA cuts price to defend volume while memory stays expensive. | **Customer-funded growth.** Year 1 completes the Rubin ramp at the pace customers' cash and observed financing allow. Thereafter replacement plus inference demand grow 5% while custom silicon takes share at the hyperscalers. Competition and memory costs normalize margins. | **Share held near the ceiling.** Customers keep spending at the edge of their cash plus financing. NVIDIA keeps its share and most of its pricing. | | Year-1 revenue (vs trailing 354; ceiling 576) | 301 (−15%) | 453 (+28%) | 559 (+58%) | | Growth from year 2 | 3% | 5% | 8% | | Year-5 revenue (ceiling 844) | 339 | 551 | 761 | | Operating margin, reached in year 3 | 30%. Pre-boom average; gross margin about 50% with opex about 18% of a smaller base. Permanent, because substitution, not under-use, causes it. | 50%. Gross margin back to the pre-boom 62%, opex about 12%. | 58%. Gross margin about 68%, opex about 10%. | | Capex / sales | 6% | 4% | 3% | | Operating capital / sales | 30% (stretched terms, excess inventory) | 22% | 18% | | Tax rate | 23% | 20% | 17% | | Retained-investment recovery | 40% | 80% | 100% | | Contingent loss | 25% of 164.5 (about 41) in year 3 | 0 | 0 | | Financing and dilution | No new net borrowing. Debt refinanced at 4.5%. Net shares flat. GAAP stock compensation charged. No buyback accretion. | Same | Same | | Exit (Koller: (1 − g/RONIC)/(r − g), r = 10%) | 13x (g 3%, RONIC 40%) on trough-normalized NOPAT | 18x (g 5%, RONIC 50%) on normalized NOPAT | 22x (long-run g 6%, RONIC 60%), capped because the margin stays near peak | **Temporary under-use versus permanent impairment.** A pause in hyperscaler orders alone would be temporary under-use, like FY2023: inventory provisions, a few weak quarters, then recovery with the next generation. The downside instead pairs the cyclical decline with permanent substitution. It therefore reverts margins to the pre-boom level and grows only 3% after the trough. Annual granularity places the decline in year 1. In reality it would begin after the visible orders ship, which moves cash timing but not the year-5 exit. **Funding before cash.** Growth working capital and capex, the25bn July equity commitment pool plus the2bn post-July Nscale funding reserve and both acquisition claims (Groq 0.986, Hugging Face 11.9) are paid in year 1 before any of their earnings. Neither acquisition is credited with earnings. Supply commitments flow through cost of goods. **Not modeled:** - China revenue. - Future investment gains. - Interest on accumulated cash. - Buybacks above the stock-compensation offset. - New borrowing. ## 13 November funding update Nscale's amended S-1 (pp108–110,161–162; F-64) establishes aSeptember subscription after NVIDIA'sJuly balance/commitment anchor. Reserve2bn once; do not give this financing an extra increment of operating revenue or treat Nscale's own debt as NVIDIA's debt. Its Anthropic infrastructure funding is still being arranged, so a signed service contract is not a fully financed GPU delivery. Keep the existing demand ceiling and downside financing stress; the public filing does not isolate a defensible change in aggregate NVIDIA shipments or margins. A negative-FCF project is not automatically bad: the missing question is funded delivery and matched mature cash return. Verify the expected16November closing and roll actual NVIDIA cash/holdings/remaining commitments together on17November. ## 15 November customer and commitment check Anthropic's original S-1 describes a multi-chip orchestration foundation that lowers its switching effort. The existing ledger's CUDA advantage is therefore customer-specific, not an immovable industry barrier. This supports the already-modeled substitution/share and margin fade; one advanced lab does not measure every customer's switching time or NVIDIA's future absolute shipments. Lower compute cost can also expand paid demand. The article's exact three-day benchmark was not verified in the primary pages. The same filing's10bn conditional NVIDIA subscription may fulfill the program announced in2025. The unitemized July pool cannot prove inclusion or a new incremental liability. Keep the explicit overlap judgment and0–10bn funding sensitivity in the [15November review](../updates/2026-11-15-daily-review.md). Customer reported operating cash is positive despite large noncash valuation losses, but future compute commitments and conditional IPO funding remain distinct from collected cash or fully financed delivery. No incremental revenue or terminal investment premium follows from these disclosures. # Capital productivity and reinvestment Forecast assessment:2026-10-27, operating re-underwrite3November; funding reserve updated13November2026. The model now starts from trailing four-quarter figures, adds a sourced demand ceiling and uses cycle-based scenarios. Money is in USD millions. These calculations use the researched scenarios and the shared forecast ledger. They are forward diagnostics, not independently measured historical ROIIC or realized acquisition IRRs. Local sensitivities may differ. | Period | Scenario | Capital return / proxy | Period-weighted reinvestment | New growth capital | Matching earnings | | --- | --- | ---: | ---: | ---: | ---: | | 5 years | downside | -111.3% | 22.0% | 102989.674 | 468262.855 | | 5 years | base | 23.7% | 10.8% | 115829.262 | 1074853.710 | | 5 years | upside | 144.0% | 7.5% | 120456.162 | 1615799.053 | | 10 years | downside | -49.2% | 23.2% | 207741.458 | 896022.645 | | 10 years | base | 37.1% | 10.1% | 238226.564 | 2352931.976 | | 10 years | upside | 127.6% | 6.9% | 270691.493 | 3936544.962 | Return definition: Operating ROIIC · partial capital proxy. Numerator 27485.803 / denominator 115829.262. Earnings basis: Unlevered operating NOPAT. The forecast runs from 2026-10-27 to 2031-10-27; a ten-year selection uses its own summed ledger through 2036-10-27. The measured historical record (FY2021–FY2026: lagged ROIIC 472% on tangible capital, reinvestment 22.5%, operating leverage 63%) is in `research/04-capital-allocation.md`; it is mostly price and margin on expensed R&D, so it is not used to extrapolate these forward figures. Average period reinvestment is total new growth investment / total matching earnings, not the arithmetic mean of annual rates. Old acquisition claims, dividends, debt repayment and financial stakes do not acquire a second increment of operating profit. Five-year base funding: cash generated before new investment 1068733.260, less distributions 121425.000, less old claims 12886.000, plus opening cash drawn / (cash built) 0.000, plus net borrowing / (deleveraging) -791592.998. This funds growth capital 115829.262 and separately funded financial stakes27000.000 (July pool25000 plus post-July Nscale reserve2000). Reconciliation error is below 0.000001. Rates above 100% are funded, not clipped. Forecast NOPAT change / recorded net physical and working-capital investment: 23.7%. The base NOPAT rises only modestly because the margin normalizes from 65.6% to 50% while revenue grows. Changes in margins and tax affect this numerator, while expensed growth investment is omitted from this denominator. It is a partial forecast diagnostic, not a measured acquisition-cohort return or an all-in economic ROIIC. Limits: - This operating-capital proxy uses net capex and working capital. Expensed R&D and other intangible growth investment are not capitalized; it is not all-in economic ROIIC. Changes in margins and tax also affect the numerator. - Financial stakes have no assigned operating earnings and are excluded from operating ROIIC; their funding is shown separately. - These are period-specific forecast diagnostics, not independently measured historical ROIIC or a project XIRR. The matched, lagged historical measurement is in `research/04-capital-allocation.md`; it is a separate diagnostic, not a calibration of these forecast ratios. ## Expensed growth investment The physical-capital proxy omits expensed R&D. A separate sensitivity treats forecast R&D above a fixed run-rate sustaining amount as after-tax growth investment, amortizing each new cohort over an explicitly assumed life. Both earnings and capital change by the same net amount; shareholder cash flows and XIRRs stay unchanged. The split, constant revenue intensity and lives are assumptions, not issuer useful-life disclosures. Prior R&D capital and other growth expenses are not fully reconstructed, so this remains a partial economic-capital diagnostic. | Assumed life | ROIIC proxy | Weighted reinvestment | Growth capital | | --- | ---: | ---: | ---: | | 3 years | 22.1% | 12.6% | 138288.642 | | 5 years | 22.5% | 13.3% | 146544.061 | Definitions and original methodology links are in `docs/method/cash-flow-and-reinvestment.md`; source locators are in this company’s registry. Probabilities are separate judgments recorded in `research/11-scenario-probabilities.md`. # Scenario probabilities **Current review15 November2026:** both horizons retain45%/45%/10%. The Nscale reserve is a separately funded cash use with existing recovery assumptions. The newly inspected complaints seek unquantified relief; reported insurance terms and Megaspeed's statement do not establish a customer cancellation, permanent successor or enacted export restriction. Fundraising risk is already part of the joint downside. Coarse odds remain low-confidence and do not quantify every legal or financing tail. [Current review](../updates/2026-11-15-daily-review.md); [funding revision](../updates/2026-11-13-funding-revision.md). Assessment dates: - **2026-10-27:** initial weights. - **2026-11-03 (first review):** weights unchanged. - **2026-11-03 (re-underwrite):** the scenarios were redefined after the owner's challenge, so the weights were reassessed. They now apply to the new paths: digestion, customer-funded growth, and share held near the ceiling. - **2026-11-09:** filing notice and current public developments reviewed; numerical weights retained with the unquantified tail explicit. [Decision](../updates/2026-11-09-daily-review.md). - **2026-11-12:** confirmed succession; five percentage points move from upside to downside at both horizons. Annual cash paths and terminal endpoints retained. [Decision and impact](../updates/2026-11-12-daily-review.md). - **2026-11-08:** reassessed after the issuer confirmed an independent board investigation. Both horizons retain40/45/15; the reasons and additional unquantified tail limitation are recorded in the model. These are subjective judgments held with low confidence. Financial sources support the mechanisms, not measured numerical odds, and no external author provided or endorsed them. | Horizon | Downside | Base | Upside | | --- | ---: | ---: | ---: | | 5 years | 45% | 45% | 10% | | 10 years | 45% | 45% | 10% | Previous weights through 11 November: 40/45/15 at both horizons. Initial weights before the 3 November re-underwrite: 5 years 35/40/25; 10 years 40/35/25. **8 November governance assessment.** The [board statement](https://nvidianews.nvidia.com/news/statement-from-nvidia-board-of-directors), paragraphs1-5, confirms a process, while retaining confidence in leadership; the [CEO response](https://nvidianews.nvidia.com/news/a-message-to-our-company) confirms cooperation. These sources do not establish a business restriction, leadership replacement or measurable change in customer-funded orders, cash margins or claims. The material risk record changes; numerical path weights remain unchanged. This is a low-confidence judgment pending findings, not a probability of exoneration. Leadership disruption, information loss and wider enforcement can create losses beyond the modeled downside; the grid is not a complete quantification of them. Revise paths and odds together if evidence establishes any of those mechanisms. Old weights and source dates remain in dated records and version history. **Historical five-year reasoning (3 November; superseded weights).** A digestion within five years is close to as likely as not: - NVIDIA itself had revenue or profit downturns in FY2009, FY2020 and FY2023, three in about fifteen years. - Every capex-cycle supplier read here had a 15–49% revenue decline after a customer spending peak: Cisco FY2002, Applied Materials FY2002, Micron FY2023. - Customers' capex now exceeds their own cash. - About a quarter of next year's demand comes from AI labs NVIDIA finances. - Custom silicon is growing about twice as fast as NVIDIA's Data Center business. The base keeps a plurality because year 1 completes orders already placed and the base never exceeds what customers can fund. The upside needs customers to keep spending at the ceiling for five years and NVIDIA to hold share, so it originally received 15%; the succession revision reduces that to 10%. **Historical ten-year reasoning (3 November).** At least one digestion over a decade is likely, and the competitive position against in-house silicon is less visible. Every path already contains its recovery growth after year 1, so no weight moves to the upside for the longer horizon. **The three paths:** - **Downside:** a digestion after the 2025–2026 build. Revenue falls about 30% below the Q3 run rate and margin returns to the 30% pre-boom level, sized from NVIDIA's own and other suppliers' reported downturns. - **Base:** growth funded by customers' own cash and observed financing, with share lost to custom silicon and margins normalized to 50%. - **Upside:** NVIDIA holds share while customers spend near the demand ceiling; margin stays near 58% and the exit is capped at 22x. Price movement alone did not change the weights. The change follows the redefinition of the scenarios and the base-rate evidence recorded in `research/08-underwriting-plan.md`. Review triggers: - Customer AI monetization/capex and concentration. - Gross margins, alternative silicon, supply commitments and inventory. - Receivables, economic R&D, partner stakes and contingent support. Each path represents a broad outcome region, not the probability of hitting an exact share price or IRR. Paths are a coarse approximation of a continuous joint distribution; the displayed downside and upside are not bounds. Joint driver dependencies must be considered, rather than multiplying marginal driver guesses. Scenario weights sum to 100% within this approximation. Severe outcomes remain possible outside the modeled paths. Price movement alone does not change these weights. A changed causal premise, acquisition capacity, financial report or scenario definition does. Local assumption sensitivities suspend the researched weights until the full distribution is reassessed. No weighted-average IRR is presented as an expected shareholder return. Sources: nvda-q2-fy2027-10q, nvda-q2-fy2027-release, nvda-q2-fy2027-call, nvda-q2-fy2027-cfo, nvda-xbrl-companyfacts, msft-fy2027-q1, googl-q3-2026, amzn-q3-2026, meta-q3-2026, orcl-q1-fy2027, avgo-q3-fy2026, csco-fy2002-results, amat-fy2002-10k, mu-fy2023-results, nvda-sb-energy. Method: [Damodaran, chapter 33](https://pages.stern.nyu.edu/~adamodar/pdfiles/val3ed/c33.pdf), printed pp.894-897. See `valuation/model.json` for machine-readable weights and `docs/method/cash-flow-and-reinvestment.md` for the capital definitions. **11 November evening reassessment.** [Dated evidence and decision](../updates/2026-11-11-evening-review.md): retain 40%/45%/15% at both horizons. The new process evidence does not identify a cash-loss amount or a defensible standalone probability. This leaves the existing omitted tail explicit; it is not an exoneration probability. **11 November late reassessment.** [Dated evidence and decision](../updates/2026-11-11-late-review.md). Both horizons stay at 40%/45%/15%, now explicitly under review. Reports that the board is weighing removal of the CEO are attributed and unconfirmed, and no transmission into orders, R&D cadence or margins is yet observed. Reweighting on headlines would fit odds to news rather than evidence. On an issuer-confirmed leadership change, reassess the downside weight, paths and terminal multiple together. **12 November current reassessment.** The issuer-confirmed departure activates the prior review trigger. Architecture, talent, supply and customer-financing decisions now face an interim coordination period. Transfer a coarse five percentage points from the joint near-ceiling upside to downside, which already contains share, margin and correlated funding stress. Installed CUDA, operating teams and funded customer demand support keeping the base at 45%. Over ten years, lost early design wins and capital decisions can persist, so no automatic reversion restores upside weight. Exact numerical odds are unmeasured and low confidence. No separate cash loss or exit-multiple haircut is added. The [same-price/date audit](../updates/2026-11-12-probability-change.json) holds every annual row and capital metric fixed. Confirmed economic costs or gains require revising the paths themselves; these weights do not price every legal/security tail. # Bull–bear review — NVIDIA, 5 November 2026 Independent advocates prepared opening cases without seeing the other draft, then exchanged point-by-point cross-examinations. A separate judge who did not author the prior model assessed the competing claims. Source reconstructions retain their stated provenance; inaccessible private reports were not newly read. Quoted sensitivities test arguments and are not automatically adopted forecasts. Fixed quote USD 200.13, market time 2026-11-04T21:00:01.000Z; valuation date 2026-11-05. Original forecast exits retained. | Horizon | Before bear/base/bull | After bear/base/bull | Weighted before → after | | --- | --- | --- | --- | | 5 years | -23.24% / 0.53% / 14.88% | -23.24% / 0.53% / 14.88% | -2.77% → -2.77% | | 10 years | -9.75% / 4.15% / 12.54% | -9.75% / 4.15% / 12.54% | 2.74% → 2.74% | ## Bull case As of 5 November 2026. No opposing advocate draft was read. Existing company research, fact/calculation ledgers and exact registered source locators were examined; original issuer sources additionally checked live: Lumine Q3, CSU Q2, Diploma Q3 transcript, NVIDIA CFO commentary, TerraVest Highland/KBK, WOSG FY26. Harvia Q3 PDF retrieval failed; its registered extraction remains the evidence limit. Private source interpretations below are the repository's documented reconstruction, not a new claim to have read an unavailable private original. The strongest bear case is that major customers' AI capital spending is externally financed and concentrated, while custom silicon compresses NVIDIA's shortage margins. GPU demand can grow secularly and still suffer a capex digestion, correlated guarantee losses and a poor shareholder return. **NVDA-B1 — The installed software/system advantage buys time.** CUDA, networking and full-rack delivery reduce time-to-deployment and total customer engineering cost; sovereign/enterprise inference can broaden demand beyond hyperscalers. Test58% first-year revenue growth versus 28%, holding later 5% growth and capital/credit assumptions. Sources `nvda-q2-fy2027-call`, pp.3–7,10; `nvda-q2-fy2027-cfo`, hyperscale/ACIE and inventory discussion. It stays below the repository's funded customer-demand ceiling. Management's70% preliminary FY2028 view is not independent demand evidence. Broadcom's AI growth (`avgo-q3-fy2026`, release segment/outlook) is simultaneous evidence of expanding demand and substitution. **NVDA-B2 —50% may over-normalise the present advantage.** Test58% operating margin versus 50%, keeping revenue and funded investment fixed. R&D software reuse, hardware/software optimisation and platform ecosystem can preserve rents even as memory costs rise; 71–73% gross-margin management expectations support a bounded possibility, not proof (`nvda-q2-fy2027-call`, p.7). The pre-boom27% operating-margin outside view describes a different mix, but today's65.6% likewise cannot be a permanent baseline. No claim of customer spending-independent growth is made. **NVDA-B3 — A finite competitive lead can support 18×.** Test22× versus 18×.18× requires about 14 post-exit years at 8% earnings growth, 50% marginal return and 10% required return before mature no-growth value. The mechanism is CUDA, interconnect and developer adoption; the challenge is customers' own chips and model-efficiency gains. Source `nvda-q2-fy2027-10q`, MD&A pp.25–32 and commitments note 10. Both14years and 50% are explicit forward judgments, not measured duration. The starting 353.964bn revenue includes the unreported Q3 guidance midpoint: it is a guidance-based trailing estimate, not four reported quarters. Correct that label; replacing it mechanically with an older reported quarter would not improve the forward judgment. ### Isolated shared-engine effects Saved quote 200.13; market timestamp 2026-11-04T21:00:01.000Z. Fixed valuation date 2026-11-05; model 6728a011341f; exits remain the original model vintage. These are sensitivities, not joint forecasts or probability revisions. Baseline base IRR 0.53% / 4.15% (5/10 years). | Claim | Exact input/path | 5-year base | 10-year base | Change, pp (5/10) | | --- | --- | ---: | ---: | --- | | NVDA-B1 | `cases.base.inputs.revenueGrowth` | 4.81% | 6.35% | +4.28 / +2.21 | | NVDA-B2 | `cases.base.inputs.operatingMargin` | 3.50% | 5.71% | +2.96 / +1.57 | | NVDA-B3 | `cases.base.inputs.exitMultiple` | 3.91% | 5.68% | +3.38 / +1.54 | ## Bear case ### Strongest opposing case CUDA, networking and a complete rack system lower customer deployment time, while inference expands compute demand beyond the original training cycle. Hyperscalers have placed orders, NVIDIA has abundant cash and high tangible returns, and custom accelerators can coexist with a much larger total market. A near-term credit collapse cannot be asserted merely because customers are investing heavily. ### Bear case The binding risk is who pays for the next unit of compute and who captures its economics. Customers' capex includes buildings, power, custom silicon, CPUs and other equipment; it cannot all be mapped to NVIDIA sales. The demand ceiling in the current opportunity ledger makes these distinctions, but some of its GW/content figures and future customer funding still come from management propositions and third-party transcript summaries. Customer investment announcements are not deployed, utilized, cash-profitable capacity. The ongoing laboratory/neocloud P&L reconciliation is a material gap, not a reason to extrapolate demand. [nvda-q2-fy2027-call p.3–10; research/08-underwriting-plan.md Opportunities and customer sources/uses; googl-q3-2026, amzn-q3-2026, meta-q3-2026, orcl-q1-fy2027 cash-flow tables.] NVIDIA's largest buyers also have incentives and resources to substitute their own chips for repeatable workloads. Current cloud demand is evidence for compute, not for perpetual merchant-GPU share or price per watt. The base's50% GAAP operating margin already assumes lasting operating leverage far above the pre-boom average near 27%. The unit moat can survive while that margin does not. [nvda-fy2026-10k history and concentration; nvda-q2-fy2027-10q MD&A pp.25–32; googl-q3-2026 TPU discussion; amzn-q3-2026 custom-chip discussion; avgo-q3-fy2026 operating results.] The outside view includes NVIDIA's own earnings declines and Cisco, Applied Materials and Micron inventory/capex cycles. They establish that digestions are possible; their percentages are not interchangeable forecasts. A first-year15% revenue stress and 40% mature operating margin test customer delay and bargaining without claiming AI disappears. [csco-fy2002-results; amat-fy2002-10k; mu-fy2023-results; research/08-underwriting-plan.md Outside view.] Lab equity, purchase/capacity commitments and guarantee exposures are correlated with those customers' demand. An80% recovery on investments and zero calls in the base should be challenged together; the model's sensitivities show these claims matter less to the five-year value than operating margin and exit price, so they should not distract from the main earnings risk. An18x exit requires 5% profitable perpetual growth beyond 2031 in the Koller calculation. No finite advantage duration is evidenced; constant secular demand does not prove constant supplier rents. [nvda-q2-fy2027-10q notes 5–10; research/05-thesis.md Exit multiple.] ### Contested claims - NVDA-B1 —Test year 1 revenue−15% against+28%, leaving subsequent 5% growth fixed. Orders and finance availability can slip without permanent demand destruction. - NVDA-B2 —Test40% operating margin versus 50%; a disclosed sensitivity between historical 27% and current peak, not an observed normalized level. - NVDA-B3 —Test13x exit against 18x to shorten the premium for durable excess returns. - NVDA-B4 —Test40% investment recovery plus 25% claim/capacity stress jointly, preserving demand in this isolated sensitivity to avoid disguising correlated risks as independent probabilities. The current issuer release was reopened. The pending 17 November quarter must replace the guidance-inclusive baseline when published; Q3 guidance is not reported revenue. This review does not validate private contracts, lab profitability or every funding source. ### Shared-engine effects Fixed saved quote USD200.13; market timestamp 2026-11-04T21:00:01.000Z; valuation date 5 November 2026; fixed original exit calendar. Base before: 0.53% / 4.15% (5/10years). These are one-change sensitivities to the base, except named bundles; existing scenario probabilities are not revised or claimed as validated. | Claim | Change | 5-year base IRR | 10-year base IRR | 5-year effect | 10-year effect | | --- | --- | ---: | ---: | ---: | ---: | | NVDA-B1 | Year1 demand decline 15% | -7.26% | -0.02% | -7.79pp | -4.17pp | | NVDA-B2 | Operating margin 40% from year 3 | -3.74% | 1.83% | -4.27pp | -2.32pp | | NVDA-B3 | Base exit 13x | -4.45% | 1.88% | -4.98pp | -2.27pp | | NVDA-B4 | Investments40% recovery and 25% claim/capacity stress | 0.16% | 4.01% | -0.38pp | -0.13pp | ## Rebuttals ### Bull response 5 November 2026. Written after independently frozen openings were exchanged. Every numbered bear claim is addressed below; original bull openings and sensitivities remain unchanged. 1. **Bear first-year−15%: retain as adverse, reject as default.** Actual hyperscale/ACIE revenue and orders show broad installed demand (`nvda-q2-fy2027-cfo`, revenue discussion), and customers possess recurring operating cash plus disclosed funding. They do not prove utilisation or returns on every marginal rack. My58% upside uses the existing funding ceiling, but the ceiling itself includes forward management GW/content assumptions. End-user economics and credit conditions remain surviving tests. No statement that orders equal cash-profitable demand is defended. 2. **Bear40% normalized margin: concede uncertainty, rebut precise historical transplantation.** Pre-boom27% operated with a different product mix, scale and platform position; current 75% gross margin is equally unsuitable as permanent proof. Memory pass-through and custom silicon can lower rents while CUDA survives. The base50% and upside 58% are explicit judgments bounded by unit economics, not measured through-cycle returns. Stress40% is valid; a single capex-cycle analogue does not establish it centrally. 3. **Bear13× versus 18×: concede finite duration not established.** My14-year diagnostic is deliberately demanding. Product lead, CUDA and networking support a premium, but hyperscalers also fund competing systems. A lower multiple requires a named cash-growth/reinvestment/advantage-duration path, not merely aversion to the quote. Historical supplier boom/bust examples establish possibility, not the exact future drawdown or capitalisation. 4. **Bear correlated 40% stakes recovery/25% calls: concede material joint risk.** Preserve the actual contractual perimeter and avoid charging a purchase commitment already funded in cost of sales twice (`nvda-q2-fy2027-10q`, notes 5–10). The isolated balance-sheet stress's modest IRR effect does not neutralise correlation with demand/margins. The existing adverse case already combines these mechanisms; weights are joint scenario judgments, not independent event odds. 5. **Baseline evidence attack: concede prose correction, rebut mechanical staleness.**353.964bn includes 108bn Q3 guidance, not reported sales. SEC CFO Outlook explicitly calls 108bn expected, 74% gross margin and 9.2bn GAAP opex. A guide-based trailing estimate is permissible if separated from reported facts and replaced 17 November. Reverting silently to an older trailing base without rebuilding customer-funded future sales is not a method improvement. Conviction high in present product economics, low in long-horizon price/value confidence. ### Bear response 5 November 2026, after independent openings were exchanged. 1. **Bull58% first-year growth: hardware/system lead conceded, cash-demand claim unresolved.** Orders, actual hyperscale/ACIE sales and available customer financing support a strong near-term book; they do not establish final-customer utilization or profitable cash demand. The bull concedes the ceiling still uses management GW/content assumptions and private lab accounts remain unavailable (`nvda-q2-fy2027-cfo`, revenue/outlook; `08-underwriting-plan`, customer sources/uses). Retain58% as conditional upside and−15% as digestion stress; neither is selected centrally by analogy. 2. **Bull58% margin: modern mix matters, precise durability unproven.** I concede a pre-boom27% average is not a central forecast for today's software/rack/network mix. Equally, present peak gross margin and scale do not establish 58% GAAP operating margins indefinitely.40% was a sensitivity between historical and current conditions, not an observed normal level. Base50% remains a reasoned conditional judgment until memory prices, custom silicon and customer value capture distinguish them (`nvda-q2-fy2027-10q`, MD&A; callp.7; source-linked own-history record). 3. **Bull18/22x exit: duration uncertainty survives.** CUDA and interconnect can persist while the share of customer profit accruing to NVIDIA shrinks. Fourteen post-exit years 8% growth at 50% marginal return is a demanding joint claim, and the bull concedes no observed moat life. My13x stress is not uniquely central either. Disclose both, keep continuation assumptions explicit, and do not let a high multiple on a lower margin hide peak-rent extrapolation. 4. **Correlated investment/guarantee stress: agreed.** Investment recovery and guarantee calls can deteriorate with demand, so do not treat their odds as independent. Preserve the contractual perimeter and do not double-charge supply commitments already in cost of sales (`nvda-q2-fy2027-10q`, notes 5–10). Isolated balance-sheet sensitivity is smaller than margin/price sensitivity, which does not refute severe joint losses. 5. **Guidance-inclusive baseline: factual label correction required, not an automatic cash revision.**353.964bn includes 108bn unreported Q3 guidance. It is a guided/estimated trailing bridge, not four reported quarters. I agree with the bull that silently substituting an older quarter without rebuilding future customer demand would be mechanical staleness. Reported facts and estimates must be separated now and the report replaced 17 November. No other central numerical change is established by this exchange. High conviction in current product economics does not establish a margin of safety atUSD200.13; moderate-to-low conviction in post 2027 cash/terminal forecasts remains appropriate. ## Judgment Existing estimates receive no presumption merely because they are already in the model. Positive company evidence, the relevant outside view and the capital/funding bridge determine each verdict; neither arbitrary conservatism nor an unsupported favorable sensitivity replaces it. The evidence and reconstruction below explain numerical changes or explicit no-change conclusions claim by claim. | Contested claim | Verdict | Evidence/locator | Reason and model action | | --- | --- | --- | --- | | bull advocate: NVDA-B1: cases.base.inputs.revenueGrowth | reject | nvda-q2-fy2027-call pp.3–10; nvda-q2-fy2027-10q MD&A pp.25–32 and notes 5–10; googl-q3-2026 and amzn-q3-2026 customer cash/customsilicon discussion; nvda-q2-fy2027-cfo. Orders, CUDA/networking and funding ceiling support ongoing growth but not adopting management 70% as independent demand.58% remains upside. | Reconstructed the isolated proposed input/path from independent opening and shared engine, at frozen saved quote/date/horizons; source mechanism separated from unsupported numerical extrapolation. | | bull advocate: NVDA-B2: cases.base.inputs.operatingMargin | reject | nvda-q2-fy2027-call pp.3–10; nvda-q2-fy2027-10q MD&A pp.25–32 and notes 5–10; googl-q3-2026 and amzn-q3-2026 customer cash/customsilicon discussion; nvda-q2-fy2027-cfo. Current product/system advantage supports margins above preboom mix; 50% normalizes current 65.6%.58% belongs upside, with recurring R&D still charged. | Reconstructed the isolated proposed input/path from independent opening and shared engine, at frozen saved quote/date/horizons; source mechanism separated from unsupported numerical extrapolation. | | bull advocate: NVDA-B3: cases.base.inputs.exitMultiple | reject | nvda-q2-fy2027-call pp.3–10; nvda-q2-fy2027-10q MD&A pp.25–32 and notes 5–10; googl-q3-2026 and amzn-q3-2026 customer cash/customsilicon discussion; nvda-q2-fy2027-cfo. 18x requires roughly 14 post-exit years at 8%growth/50%marginalreturn/10%requiredreturn. CUDA/networking provides a mechanism; its duration remains unmeasured and 22x is not justified merely by growth. | Reconstructed the isolated proposed input/path from independent opening and shared engine, at frozen saved quote/date/horizons; source mechanism separated from unsupported numerical extrapolation. | | bear advocate: NVDA-B1: Year1 demand decline 15% | reject | nvda-q2-fy2027-call pp.3–10; nvda-q2-fy2027-10q MD&A pp.25–32 and notes 5–10; googl-q3-2026 and amzn-q3-2026 customer cash/customsilicon discussion; nvda-q2-fy2027-cfo. A15% digestion is credible and already adverse; recent funded orders and current customer capacity commitments do not establish an imminent central contraction. | Reconstructed the independent isolated stress at the same quote/date/horizons; do not treat stress size as a probability or a measured current fact. | | bear advocate: NVDA-B2: Operating margin 40% from year 3 | reject | nvda-q2-fy2027-call pp.3–10; nvda-q2-fy2027-10q MD&A pp.25–32 and notes 5–10; googl-q3-2026 and amzn-q3-2026 customer cash/customsilicon discussion; nvda-q2-fy2027-cfo. Customer bargaining and customchips are genuine; normalizing 65.6% to 50% already shares that risk.40% stays an adverse path until workload/mix evidence discriminates. | Reconstructed the independent isolated stress at the same quote/date/horizons; do not treat stress size as a probability or a measured current fact. | | bear advocate: NVDA-B3: Base exit 13x | reject | nvda-q2-fy2027-call pp.3–10; nvda-q2-fy2027-10q MD&A pp.25–32 and notes 5–10; googl-q3-2026 and amzn-q3-2026 customer cash/customsilicon discussion; nvda-q2-fy2027-cfo. 13x quantifies shorter advantage, not a uniquely established terminal value. Retain18x as explicitly conditional finite-duration judgment, low conviction. | Reconstructed the independent isolated stress at the same quote/date/horizons; do not treat stress size as a probability or a measured current fact. | | bear advocate: NVDA-B4: Investments40% recovery and 25% claim/capacity stress | reject | nvda-q2-fy2027-call pp.3–10; nvda-q2-fy2027-10q MD&A pp.25–32 and notes 5–10; googl-q3-2026 and amzn-q3-2026 customer cash/customsilicon discussion; nvda-q2-fy2027-cfo. Correlated funding/investment risk is admitted in downside; no verified claim call or recovery observation supports assigning 40%/25% centrally. | Reconstructed the independent isolated stress at the same quote/date/horizons; do not treat stress size as a probability or a measured current fact. | Retain existing coarse horizon-specific weights after reassessment: scenarios already represent these adverse/favorable operating mechanisms. Numerical odds are subjective, not verified frequencies; unresolved gaps prevent high conviction. No tuning to return. ## Conviction Low in precise forward returns; platform advantage supported, customer-paid demand and durable margins unverified. These are forward estimates with surviving attacks, not confirmed outcomes. Neither numerical calibration, a completed checklist nor passing software tests establishes investment skill. Reported facts stay separate from forward judgments, and adverse paths are not worst-case bounds. **What would change the view:** 17 November Q3 actuals replace guidance-inclusive starting estimate; reconcile customer-funded purchases, RPO/cash, receivables, guarantees and utilization. [Exact inputs, isolated sensitivities, model hashes and same-price/date before/after returns](../updates/2026-11-05-debate-review.json). ## Public Primary & Secondary Sources - `nvda-q2-fy2027-10q` (primary): NVIDIA Q2 FY2027 Form 10-Q — https://www.sec.gov/Archives/edgar/data/1045810/000104581026000075/nvda-20260726.htm (published 2026-08-26) - `nvda-q2-fy2027-release` (primary): NVIDIA Q2 FY2027 results and Q3 outlook — https://nvidianews.nvidia.com/news/nvidia-announces-financial-results-for-second-quarter-fiscal-2027 (published 2026-08-26) - `nvda-fy2026-10k` (primary): NVIDIA FY2026 annual report — https://www.sec.gov/Archives/edgar/data/1045810/000104581026000021/nvda-20260125.htm (published 2026-02-25) - `nvda-q2-fy2027-call` (primary): NVIDIA issuer-hosted Q2 FY2027 earnings transcript — https://investor.nvidia.com/files/content_files/TRANSCRIPT_-NVIDIA-Corp-NVDA-US-Q2-2027-Earnings-Call-26-August-2026-5_00-PM-ET.pdf (published 2026-08-26) - `nvda-supply-oct2026` (primary): NVIDIA supply-chain capacity expansion — https://nvidianews.nvidia.com/news/nvidia-and-supply-chain-partners-expand-capacity-to-accelerate-ai-infrastructure-buildout (published 2026-10-21) - `nvda-sb-energy` (primary): NVIDIA SB Energy investment and guarantee announcement — https://nvidianews.nvidia.com/news/nvidia-guarantees-sb-energy-s-ports-pike-technology-campus-in-ohio-to-exclusively-host-nvidia-ai-compute (published 2026-08-17) - `msft-fy2026-q4` (primary): Microsoft FY2026 Q4 earnings discussion — https://www.microsoft.com/en-us/investor/events/fy-2026/earnings-fy-2026-q4 (published 2026-07-29) - `tsmc-q3-2026` (primary): TSMC Q3 2026 earnings discussion — https://investor.tsmc.com/schinese/encrypt/files/encrypt_file/reports/2026-10/eba4fac72873bf546b669bd2b5d973df5c93215d/TSMC%203Q26%20Transcript.pdf (published 2026-10-15) - `amd-q2-2026` (primary): AMD Q2 2026 issuer results — https://ir.amd.com/financial-information/sec-filings/content/0000002488-26-000121/q22026991.htm (published 2026-08-04) - `nvda-price-2026-10-27` (market data): NVDA delayed market quote — https://finance.yahoo.com/quote/NVDA/ (published 2026-10-27) - `capital-method-roiic` (methodology): Mauboussin and Callahan: Return on Invested Capital — https://www.morganstanley.com/im/publication/insights/articles/article_returnoninvestedcapital.pdf (published 2022-10-06) - `capital-method-cash` (methodology): Dan Shuart, Eagle Point Capital: A Framework For Spotting Value Traps — https://eaglepointcapital.substack.com/p/spotting-value-traps-the-anti-compounders (published 2023-08-26) - `scenario-method` (methodology): Aswath Damodaran: Probabilistic Approaches in Valuation, chapter 33 — https://pages.stern.nyu.edu/~adamodar/pdfiles/val3ed/c33.pdf - `nvda-fy2025-10k` (primary): NVIDIA FY2025 annual report (Form 10-K) — https://www.sec.gov/Archives/edgar/data/1045810/000104581025000023/nvda-20250126.htm (published 2025-02-26) - `nvda-fy2024-10k` (primary): NVIDIA FY2024 annual report (Form 10-K) — https://www.sec.gov/Archives/edgar/data/1045810/000104581024000029/nvda-20240128.htm (published 2024-02-21) - `nvda-fy2023-10k` (primary): NVIDIA FY2023 annual report (Form 10-K) — https://www.sec.gov/Archives/edgar/data/1045810/000104581023000017/nvda-20230129.htm (published 2023-02-24) - `nvda-fy2022-10k` (primary): NVIDIA FY2022 annual report (Form 10-K) — https://www.sec.gov/Archives/edgar/data/1045810/000104581022000036/nvda-20220130.htm (published 2022-03-18) - `nvda-xbrl-companyfacts` (primary): SEC XBRL company facts for NVIDIA (CIK 1045810) — https://data.sec.gov/api/xbrl/companyfacts/CIK0001045810.json (published 2026-08-26) - `nvda-q4-fy2026-cfo` (primary): NVIDIA Q4 FY2026 CFO commentary — https://www.sec.gov/Archives/edgar/data/1045810/000104581026000019/q4fy26cfocommentary.htm (published 2026-02-25) - `nvda-q2-fy2027-cfo` (primary): NVIDIA Q2 FY2027 CFO commentary — https://www.sec.gov/Archives/edgar/data/1045810/000104581026000073/q2fy27cfocommentary.htm (published 2026-08-26) - `nvda-proxy-2026` (primary): NVIDIA 2026 proxy statement (DEF 14A) — https://www.sec.gov/Archives/edgar/data/1045810/000104581026000036/nvda-20260512.htm (published 2026-05-12) - `nvda-agm-2026-votes` (primary): NVIDIA 2026 annual meeting results (Form 8-K item 5.07) — https://www.sec.gov/Archives/edgar/data/1045810/000104581026000056/nvda-20260624.htm (published 2026-06-30) - `nvda-hugging-face-8k` (primary): NVIDIA agreement to acquire Hugging Face (Form 8-K item 8.01) — https://www.sec.gov/Archives/edgar/data/1045810/000104581026000078/nvda-20260902.htm (published 2026-09-03) - `nvda-arm-termination` (primary): NVIDIA Arm transaction termination (Form 8-K, Feb 2022) — https://www.sec.gov/Archives/edgar/data/1045810/000104581022000005/nvda-20220208.htm (published 2022-02-08) - `sec-nvda-crypto-2022` (primary): SEC order: NVIDIA cryptomining disclosure settlement — https://www.sec.gov/news/press-release/2022-79 (published 2022-05-06) - `nvda-huang-form4` (primary): Jen-Hsun Huang Form 4 filings — https://www.sec.gov/cgi-bin/browse-edgar?action=getcompany&CIK=1197649&type=4&owner=include&count=100 (published 2026-09-18) - `msft-fy2027-q1` (counterparty primary): Microsoft FY2027 Q1 results and call — https://www.microsoft.com/en-us/investor/events/fy-2027/earnings-fy-2027-q1 (published 2026-10-28) - `googl-q3-2026` (counterparty primary): Alphabet Q3 2026 results and call — https://www.sec.gov/Archives/edgar/data/0001652044/000165204426000080/googexhibit991q32026.htm (published 2026-10-28) - `amzn-q3-2026` (counterparty primary): Amazon Q3 2026 results — https://s2.q4cdn.com/299287126/files/doc_earnings/2026/q3/earnings-result/AMZN-Q3-2026-Earnings-Release.pdf (published 2026-10-29) - `meta-q3-2026` (counterparty primary): Meta Q3 2026 results — https://www.sec.gov/Archives/edgar/data/0001326801/000162828026066700/meta-09302026xexhibit991.htm (published 2026-10-21) - `orcl-q1-fy2027` (counterparty primary): Oracle FY2027 Q1 results — https://investor.oracle.com/files/content_files/1q27-pressrelease-September_FINAL.pdf (published 2026-09-10) - `avgo-q3-fy2026` (counterparty primary): Broadcom FY2026 Q3 results — https://www.sec.gov/Archives/edgar/data/0001730168/000173016826000076/avgo-08022026x8kxex99.htm (published 2026-09-02) - `nvda-q4-fy2026-release` (primary): NVIDIA Q4 and FY2026 results release — https://nvidianews.nvidia.com/news/nvidia-announces-financial-results-for-fourth-quarter-and-fiscal-2026 (published 2026-02-25) - `csco-fy2002-results` (base-rate primary): Cisco fourth quarter and fiscal 2002 results — https://newsroom.cisco.com/c/r/newsroom/en/us/a/y2002/m08/cisco-systems-reports-fourth-quarter-and-fiscal-year-2002-earnings.html (published 2002-08-06) - `amat-fy2002-10k` (base-rate primary): Applied Materials FY2002 Form 10-K — https://www.sec.gov/Archives/edgar/data/0000006951/000000695103000003/form10k.htm (published 2003-01-08) - `mu-fy2023-results` (base-rate primary): Micron fiscal Q4 and FY2023 results (Form 8-K exhibit 99.1) — https://www.sec.gov/Archives/edgar/data/723125/000072312523000051/a2023q4ex991-pressrelease.htm (published 2023-09-27) - `nvda-q3-fy2027-call-date` (primary): NVIDIA sets conference call for third-quarter fiscal 2027 results — https://nvidianews.nvidia.com/news/nvidia-sets-conference-call-for-third-quarter-financial-results-6928873 (published 2026-10-28) - `avgo-backstop-8k-2026-10` (counterparty primary): Broadcom Form 8-K Item 2.03: residual value support for financed AI racks — https://www.sec.gov/Archives/edgar/data/1730168/000119312526434271/d148323d8k.htm (published 2026-10-27) - `orcl-notes-2026-11` (counterparty primary): Oracle free writing prospectus: 10.0bn senior notes — https://www.sec.gov/Archives/edgar/data/1341439/000119312526443918/d87441dfwp.htm (published 2026-11-02) - `orcl-equity-2026-11` (counterparty primary): Oracle prospectus supplement: 10.0bn common stock offering — https://www.sec.gov/Archives/edgar/data/1341439/000119312526442729/d181425d424b5.htm (published 2026-11-02) - `meta-forward-2026-10` (counterparty primary): Meta prospectus supplement: forward sale of Class A shares — https://www.sec.gov/Archives/edgar/data/1326801/000119312526436410/d94886d424b5.htm (published 2026-10-27) - `meta-preferred-2026-10` (counterparty primary): Meta Form 8-K: 6.50% mandatory convertible preferred — https://www.sec.gov/Archives/edgar/data/1326801/000119312526438017/d183592d8k.htm (published 2026-10-29) - `meta-notes-2026-11` (counterparty primary): Meta free writing prospectus: 15.0bn senior notes — https://www.sec.gov/Archives/edgar/data/1326801/000119312526445646/d186765dfwp.htm (published 2026-11-03) - `msft-notes-2026-10` (counterparty primary): Microsoft free writing prospectus: 25.0bn notes — https://www.sec.gov/Archives/edgar/data/789019/000119312526439794/d799291dfwp.htm (published 2026-10-29) - `googl-euro-notes-2026-11` (counterparty primary): Alphabet free writing prospectus: EUR 6.0bn notes — https://www.sec.gov/Archives/edgar/data/1652044/000119312526443857/d134038dfwp.htm (published 2026-11-02) - `nvda-house-briefing-request-2026-11-05` (primary): House committee chairs request Jensen Huang briefing — https://foreignaffairs.house.gov/news/press-releases/chairmen-mast-moolenaar-and-crawford-request-jensen-huang-brief-congress-on-link-to-chinese-intelligence-asset (published 2026-11-05) - `nvda-coxe-form4-2026-11-06` (primary): Tench Coxe Form4 filed6November2026 — https://www.sec.gov/Archives/edgar/data/1045810/000119764726000013/wk-form4_1793999520.xml (published 2026-11-06) - `nvda-board-investigation-2026-11-08` (primary): NVIDIA Board announces independent investigation — https://nvidianews.nvidia.com/news/statement-from-nvidia-board-of-directors (published 2026-11-08) - `nvda-ceo-message-2026-11-08` (primary): NVIDIA CEO message supporting board inquiry — https://nvidianews.nvidia.com/news/a-message-to-our-company (published 2026-11-08) - `nvda-rosen-filed-notice-2026-11-09` (primary): Rosen Law Firm: NVIDIA securities class action filed (Business Wire distribution) — https://sanantonio.newsnetmedia.com/story/557412/equity-alert-rosen-law-firm-files-securities-class-action-lawsuit-on-behalf-of-nvidia-corporation-investors-nvda/ (published 2026-11-09) - `nvda-rebellions-skt-2026-11-08` (primary): NVIDIA/Rebellions/SK Telecom NVLink Fusion collaboration — https://nvidianews.nvidia.com/news/nvidia-rebellions-and-sk-telecom-announce-collaboration-to-advance-sovereign-ai-infrastructure-with-nvidia-nvlink-fusion (published 2026-11-08) - `nvda-korean-startups-2026-11-09` (primary): NVIDIA: Korean startup investment forum and announcements — https://blogs.nvidia.com/blog/korean-startup-investments-msit/ (published 2026-11-09) - `tsmc-october-2026-revenue` (primary): TSMC October 2026 revenue report — https://pr.tsmc.com/english/news/3349 (published 2026-11-10) - `nvda-reuters-congress-2026-11-11` (reporting): Reuters: House Republicans discuss conditional subpoena — https://cruisin929.com/2026/11/11/house-republicans-push-for-nvidias-huang-to-testify-float-subpoena-over-spy-allegations/ (published 2026-11-11) - `nvda-transition-2026-11-12` (primary): NVIDIA announces CEO transition — https://nvidianews.nvidia.com/news/nvidia-announces-ceo-transition (published 2026-11-12) - `nvda-transition-8k-2026-11-12` (primary): NVIDIA transition Form8-K, Item5.02 — https://www.sec.gov/Archives/edgar/data/1045810/000104581026000082/nvda-20261111.htm (published 2026-11-12) - `nvda-senate-preservation-2026-11-11` (primary): Warren and Coons records-preservation letter to administration — https://www.banking.senate.gov/imo/media/doc/20261111_warren_coons_letter_to_wh_commerce_state_re_nvidia_record_retentionpdf.pdf (published 2026-11-11) - `nvda-moolenaar-transition-2026-11-12` (primary): Moolenaar statement on NVIDIA leadership transition — https://chinaselectcommittee.house.gov/media/press-releases/moolenaar-statement-on-jensen-huang-s-resignation-from-nvidia (published 2026-11-12) - `nvda-nbc-briefing-2026-11-12` (reporting): NBC News: Huang agrees to brief lawmakers; date still being arranged — https://www.aol.com/articles/jensen-huang-agrees-brief-congress-153230000.html (published 2026-11-12) - `nvda-geforce-2026-11-12` (primary): NVIDIA GeForce NOW: Dawn of War IV and weekly games update — https://blogs.nvidia.com/blog/geforce-now-thursday-warhammer-dawn-of-war-iv-coming/ (published 2026-11-12) - `nvda-house-second-request-2026-11-12` (primary): House Foreign Affairs, China Select and Intelligence chairmen: second request to Huang, NVIDIA general counsel and BIS under secretary — https://foreignaffairs.house.gov/news/press-releases/chairmen-mast-moolenaar-and-crawford-send-second-request-to-jensen-huang-and-nvidia-to-brief-congress-on-the-chinese-intelligence-asset (published 2026-11-12) - `nscale-s1-2026-11-12` (primary): Nscale amended IPO registration: financing, NVIDIA subscription and Anthropic funding — https://www.sec.gov/Archives/edgar/data/2110365/000119312526458722/ck0002110365-20261112.htm (published 2026-11-12) - `nvda-mha-2026-11-12` (primary): Singapore MHA: investigations into Alice Huang and Megaspeed — https://www.mha.gov.sg/media-room/newsroom/singapore-investigations-into-alice-huang-huang-le-and-megaspeed-international-pte-ltd/ (published 2026-11-12) - `nvda-finance-letter-2026-11-12` (primary): Wyden and Neal records-preservation letter to Commerce — https://www.finance.senate.gov/wp-content/uploads/2026/11/11122026wydenneallettertolutnickonnvidia.pdf (published 2026-11-12) - `nvda-stevens-2026-09-02` (primary): Mark Stevens Form4 filed2September2026 — https://www.sec.gov/Archives/edgar/data/1045810/000119903926000012/wk-form4_1788387031.xml (published 2026-09-02) - `nvda-stevens-2026-09-08` (primary): Mark Stevens Form4 filed8September2026 — https://www.sec.gov/Archives/edgar/data/1045810/000119903926000014/wk-form4_1788901755.xml (published 2026-09-08) - `nvda-stevens-2026-09-22` (primary): Mark Stevens Form4 filed22September2026 — https://www.sec.gov/Archives/edgar/data/1045810/000119903926000016/wk-form4_1790110579.xml (published 2026-09-22) - `nvda-euronews-scott-2026-11-13` (secondary): Euronews: ex-NVIDIA CEO Huang and Commerce Secretary Lutnick summoned to testify (13 November 2026) — https://www.euronews.com/2026/11/13/ex-nvidia-ceo-huang-and-us-commerce-secretary-howard-lutnick-summoned-to-testify-before-co (published 2026-11-13) - `nvda-class-complaint-2026-11-09` (primary): Kistenmacher v NVIDIA, filed securities complaint — https://www.courthousenews.com/wp-content/uploads/2026/11/kistemacher-vs-nvidia-complaint.pdf (published 2026-11-09) - `nvda-derivative-complaint-2026-11-11` (primary): Vanegas derivative complaint on behalf of NVIDIA — https://kehoelawfirm.com/wp-content/uploads/2026/11/nvidia-derivative-action-nvda.pdf (published 2026-11-11) - `nvda-insurer-2026-11-14` (secondary): The Insurer: reported NVIDIA Side A D&O tower — https://www.theinsurer.com/ti/news/exclusive-inigo-and-beazley-share-lead-nvidias-up-to-950-million-side-a-do-tower-2026-11-14/ (published 2026-11-14) - `anthropic-s1-2026-11-10` (primary): Anthropic amended S-1,10November2026 — https://www.sec.gov/Archives/edgar/data/2130357/000162828026072484/anthropicpbc-sx1ax2.htm (published 2026-11-10) - `anthropic-partnership-2025-11-18` (primary): Anthropic: Microsoft/NVIDIA strategic partnerships,18November2025 — https://www.anthropic.com/news/microsoft-nvidia-anthropic-announce-strategic-partnerships (published 2025-11-18) --- # Sygnity (SGN.WA) - Legal Name: Sygnity S.A. - Exchange: Warsaw (Poland) - Sector: Vertical software and IT services - Currency: PLN - Reviewed: 2026-11-15 - Model Framework: Software owner cash and disciplined acquisitions - Financial Period: September 2026 reported earning-power bridge, cash and claims; October g.on pro forma (anchored 2026-10-27) - Research Status: 8 of 11 research checks complete. Still to complete: Security and evidence base; Opportunities and funded scenarios; Variant view and catalysts - Static Report URL: https://longview-investing.vercel.app/research/sygnity.html - Interactive Model URL: https://longview-investing.vercel.app/#/company/sygnity - JSON API: https://longview-investing.vercel.app/api/companies/sygnity ## Executive Summary Maintain and develop critical software for regulated customers, then deploy surplus owner cash into independent software businesses. Project execution, employee economics and actual deal returns determine compounding. ## Key Operational Watch Items - Maintenance/implementation demand and organic versus acquired cash growth - Working capital, current cash tax, employee awards and reserve needs - Related-party pricing, target cash yields, actual deployment and minority rights ## Valuation Scenario Assumptions ### Downside Case Project/organic cash growth and the acquisition pipeline stall. Retain cash at face value; an 11.5x exit (zero-growth value at the market cost of equity, averaged with the 2022 trough). No separate permanent margin haircut is silently added. - Key Drivers: Common cash payout: 0.0% | Exit / after-financing owner cash: 11.41× | Net dilution beyond charged compensation: 0.0% | Available cash deployment fraction: 100.0% - Organic revenue growth: -2.0%, 0.0%, 0.0%, 0.0%, 0.0% - Owner-cash margin on organic revenue: 24.8%, 23.3%, 23.3%, 23.3%, 23.3% - Price paid per unit of acquired revenue: 1.06×, 1.06×, 1.06×, 1.06×, 1.06× - Owner-cash margin of acquired businesses: 16.2%, 16.2%, 16.2%, 16.2%, 16.2% - Acquired revenue growth after purchase: 0.0%, 0.0%, 0.0%, 0.0%, 0.0% - Deal budget (capital available for acquisitions): 0.00, 0.00, 0.00, 14.02, 14.02 ### Base Case 6.51% through-cycle organic owner-cash growth, deal spending starting at the observed FY2025 pace and growing 3.4% a year (flat deal count), the family 6.61x purchase price, and a 16x Koller exit at a 10.45% market cost of equity. No new borrowing or free buyback accretion. - Key Drivers: Common cash payout: 0.0% | Exit / after-financing owner cash: 15.71× | Net dilution beyond charged compensation: 0.0% | Available cash deployment fraction: 13.6% - Organic revenue growth: 9.5%, 7.0%, 7.0%, 5.5%, 5.5% - Owner-cash margin on organic revenue: 24.8%, 24.8%, 24.8%, 24.8%, 24.8% - Price paid per unit of acquired revenue: 1.06×, 1.06×, 1.06×, 1.06×, 1.06× - Owner-cash margin of acquired businesses: 16.2%, 16.2%, 16.2%, 16.2%, 16.2% - Acquired revenue growth after purchase: 0.0%, 0.0%, 3.4%, 3.4%, 3.4% - Deal budget (capital available for acquisitions): 26.29, 27.19, 28.11, 29.07, 30.06 ### Upside Case 11.0% organic growth and twice the base deployment fraction, growing 15% a year. Same purchase price and the same 16x exit: no extra re-rating or cheap-deal uplift. - Key Drivers: Common cash payout: 0.0% | Exit / after-financing owner cash: 15.71× | Net dilution beyond charged compensation: 0.0% | Available cash deployment fraction: 27.3% - Organic revenue growth: 13.0%, 11.0%, 11.0%, 9.0%, 9.0% - Owner-cash margin on organic revenue: 24.8%, 24.8%, 25.3%, 25.3%, 25.3% - Price paid per unit of acquired revenue: 1.06×, 1.06×, 1.06×, 1.06×, 1.06× - Owner-cash margin of acquired businesses: 16.2%, 16.2%, 16.2%, 16.2%, 16.2% - Acquired revenue growth after purchase: 0.0%, 0.0%, 3.4%, 3.4%, 3.4% - Deal budget (capital available for acquisitions): 35.74, 44.20, 44.20, 58.40, 58.40 ## Research Notebook # Business model Researched 27 October 2026; economic engine, unit economics and history rebuilt from five years of filings on 3 November 2026. Money is PLN millions unless stated. Source IDs refer to `sources/registry.json`, fact IDs to `financials/facts.json`, and named calculations to `valuation/calculations.json`. Sygnity builds, implements and services software for the Polish state, the central bank, banks, utilities and, since 2024, a few acquired product niches. Accounting, payment, welfare, labour-market and billing workflows run on systems Sygnity wrote, which gives a reason to keep paying the incumbent. Regulation and changes in customer processes create development work. Switching means migration, retraining, testing and operational risk; it remains possible, and public tenders can be lost. The fiscal year moved from October–September to the calendar year: FY2020/21 and FY2021/22 end in September, FY2022/23 ran 15 months to December 2023, and 2024 onward are calendar years. Calendar 2021 and 2022 below are derived from the audited fiscal years and the Q1 interim reports; calendar 2023 is the audited comparative column (`revenueCY2021`–`revenueCY2023` and the matching cash facts). ## Value-creation mechanism Sygnity's value comes from owning the code and domain knowledge of systems that a ministry, the central bank or a distribution utility cannot easily replace. The incumbent then gets most of the multi-year maintenance and change-request work on those systems: 48-month renewals across six Ministry of Family systems in 2024 (`pracaGovContractMax` and the PWI, AC and ZC contracts, sgn-fy24-report pp.32-33), and two annexes taking the NBP accounting-system contract to PLN135m gross for 2024–2027, the second "due to an increase in the Employer's needs" (sgn-nbp-contract). The work is paid largely in advance or on delivery and needs almost no capital, so owner cash is close to after-tax operating profit. Since TSS Europe (Topicus, in the Constellation family) took control in 2022, two more steps were added: reprice and prune the franchise, then send the surplus cash into small vertical-software companies. **Most of the value so far came from the repricing step, not from acquisitions.** Between calendar 2021 and 2025, NOPAT rose by 44.8 (`nopatChangeWindow`). Deals explain about 8.5–13.6 of that: the 2024 deals' annualized profit or cash proxy of 6.7–11.8 (`deals2024AnnualProfit`, `deals2024CashProxy`), plus DocLogix and one month of Mednow. The remaining 31–36 came from the existing business, while working capital released 24.8 of cash (`workingCapitalReleasedWindow`). The repricing was a one-time step that is now largely complete. The forward mechanism has to be margin retention against wage inflation, plus deployment of cash that otherwise sits idle. ## Business-model profile Comparable firms are chosen by these traits, not by the industry label. **Growth vehicle: Mixed: organic and acquisitions.** Revenue rose from 205.9 in 2021 to 350.3 in 2025. Companies bought in 2024–2025 contributed roughly 51 of 2025 revenue (Edrana, Sagra, DocLogix, Mednow at their run rates), so the organic base grew by about 80 over the period, with negative net capital. Deals began only in 2024 under the current owner and were 8.0% of revenue over 2023–2025. The model grows the organic base and adds deals at measured entry yields, on a record of only two active deal years. **Capital intensity: 0.6% of sales, asset-light.** Capital spending is about 0.6% of revenue: computers and offices for software engineers. Tangible capital is negative (−24.3 in 2025) because customers prepay and staff costs are billed promptly. Growth needs people and, since 2024, acquisition cash (8.0% of revenue over 2023–2025). **Comparables.** A comparable firm writes and maintains mission-critical software for government, central-bank, banking and utility customers, earns from long implementation and maintenance relationships, needs almost no fixed capital, and has recently started bolt-on acquisitions. Its revenue depends on the timing of public procurement. Not the label: "IT services" suggests consulting priced by the hour; Sygnity owns the systems its customers run and earns product-like maintenance margins. Declared in `valuation/model.json` (`businessProfile`). `npm run check` measures deal and capital intensity from the reported facts, and the outside view grades the forecast only against firms that share them. ## Economic snapshot Calendar 2025 (audited) unless stated; calculations named in brackets. | Measure | Value | Definition | | --- | ---: | --- | | Revenue (`revenue2025`) | 350.3 | +19.5% on 2024; about 14% excluding businesses bought in 2025 (sgn-presentation) | | Gross margin | 46.5% (FY to Sep 2022: 28.5%) | gross profit / revenue (`grossMargin2025`, `grossMarginFY2022`) | | EBIT margin | 25.5% (FY2022: 12.0%) | statutory operating profit (`ebitMargin2025`) | | EBIT before acquisition amortization | 96.7 (27.6%) | plus 7.2 amortization of acquired intangibles (`ebitBeforePpa2025`) | | Incremental EBIT margin | 39.7% (2024→2025) | ΔEBIT 22.7 / Δrevenue 57.1 (`incrementalMargin2025`) | | Trailing June 2026 EBIT | 106.0 (26.2% of 405.3) | FY2025 + H1 2026 − H1 2025 (sgn-h126 p.9) | | Operating working capital | −38.5 | receivables 44.7 + contract assets 17.8 + contract costs/inventory 1.3 − trade and other liabilities 61.0 − contract liabilities 41.3 (`operatingWorkingCapital2025`) | | Receivable / contract-liability days | 65 / 43 | (receivables + contract assets) / revenue; customer prepayments / revenue | | PP&E / right-of-use assets | 5.5 / 8.6 | | | Tangible capital including leases | −24.3 | working capital + PP&E + ROU (`tangibleCapital2025`) | | Return on tangible capital (leases capitalized) and ROCE (leases expensed) | not meaningful | capital is negative under both definitions: customers and employees fund the business | | Return on capital including goodwill and acquired intangibles | 34.1% after tax (36.8% before acquisition amortization) | NOPAT 72.5 at 19% / 212.6 (`returnOnCapitalIncGoodwill2025`) | | Same, excluding goodwill from mergers before 2024 | 131% | 72.5 / 55.5 (`recentCapital2025`); 157.2 of goodwill dates from the EMAX and Polsoft era | | Debt / leases / cash | 0 bank debt (last 12.0 loan repaid in 2025), bond 0.1 / leases 9.2 / cash 187.7 (225.1 at June 2026) | sgn-fy25 p.6, sgn-h126 | | Revenue / EBIT per average FTE | 0.407 / 0.104 (FY2022: 0.309 / 0.037) | 860 FTE in 2025, 752 of them in implementation (`revenuePerFte2025`) | Normalization before returns: capex (1.8) was below own-asset D&A (2.6 = 13.96 total − 4.10 right-of-use − 7.24 acquisition amortization), so maintenance spending is not understated in a way that matters at this scale. There are no minorities. Mednow is in the 2025 accounts for one month only, and the model adds its missing months separately. Cash tax is the main normalization: 2025 tax paid was 8.6 against 18.2 of current tax, because loss carry-forwards ran out (no tax losses remained at June 2026, sgn-h126 note 12.3; H1 2026 paid 24.3). The model charges current tax. The effective rate of 14.9% reflects the IP-box regime, so returns at the 19% statutory rate are slightly conservative. ## Economic segmentation Sygnity reports one IT segment (sgn-fy25 note 7). Its management report gives revenue by customer sector, and the revenue note splits servicing from implementation. Neither split is a profit split. Re-cut by how value is created: | Economic unit | Revenue evidence | Profit evidence | Demand driver | Character | | --- | --- | --- | --- | --- | | Public-administration franchise (Ministry of Family welfare and labour systems, Poczta Polska ZST, police CSD) | 96.9 in 2024 (33%); FY2022 87.7 on 98 projects (`sectorPublic2024`, `sectorPublicFY2022`) | not disclosed | statutory programme changes, EU-funded digitization, 48-month maintenance renewals | recurring core, with tender risk on new builds | | Central-bank and financial-market franchise (NBP ZSK accounting, brokerage and depository systems, bank reporting) | 84.8 in 2024 (29%); FY2022 69.8 | not disclosed | regulation (NBP, KNF, KDPW), bank IT budgets | recurring; NBP alone is up to about 27/yr net under its 2024–2027 contract | | Utilities billing and metering (PGE multiZBYT, SUD, AMS) | 64.5 in 2024 (22%); FY2022 48.0 | not disclosed | energy-market regulation: Solidarity Shield price caps (2023 act), CSIRE data hub, balancing rules | recurring base plus regulation-driven waves; part of the growth is cyclical | | Acquired product companies (Sagra retail SaaS, Edrana ERP Baltics, DocLogix, Mednow hospital IT, g.on Germany) | retail 20.0 in 2024 (Sagra about 90%); about 51 of 2025 revenue (estimate from the 2024 run rates plus the 2025 contributions) | roughly 15 of about 111 portfolio owner cash, about 13%: Edrana 2.3 and Sagra 4.3 annualized net profit, DocLogix about 0.8, plus model proxies of 0.5 for Mednow's missing months and 4.7 for g.on | customer vertical demand; SaaS and maintenance | small, growing by purchase | Across revenue streams, 2025 servicing was 144.2 (41%, +27% on 2024), implementation 189.0 (54%, +13%) and licences 15.8 (4.5%) (sgn-fy25 note 9). "Implementation" includes change requests on Sygnity's own installed systems, which behave like recurring development rather than one-off projects. The NBP annex is development work added to a maintenance contract. Conclusions. Revenue share by sector says nothing about profit share, and the issuer does not disclose sector margins, so we do not allocate profit among the three franchises. About 87% of owner cash still comes from the incumbent franchises and about 13% from acquired businesses. Utilities grew fastest from FY2022 to 2024 (48.0 to 64.5, +34%) during the Solidarity Shield and CSIRE adaptation wave. Up to about 16 of revenue, perhaps 6–7 of EBIT at the group's incremental margin, may be regulation-timed rather than permanent. ## Unit economics The unit through which Sygnity adds value is the installed-system relationship, and the intervention that mattered is TSS's ownership since 2022. The before-and-after comparison is therefore the whole incumbent business per delivery employee, FY to September 2022 (the last year before TSS's influence) against calendar 2025. Contracts are not costed separately in public filings. | Per average FTE (PLN thousand) | FY2022 (685 FTE) | 2025 (860 FTE) | Change | | --- | ---: | ---: | --- | | Revenue | 309 | 407 | +32% | | Subcontractors and other third-party services | 94 (30.4% of revenue) | 88 (21.6%) | lower in money as well as share | | Payroll and benefits | 159 (51.5%) | 191 (46.9%) | +20%; average pay at the parent rose from 123 (FY2021) to 168 | | D&A | 16 | 16 | 2025 includes acquired-intangible amortization | | Gross profit | 88 | 189 | ×2.1 | | EBIT | 37 | 104 | ×2.8 | | Tangible capital | negative | negative | prepayments and payables exceed receivables | The levers were pruning and price, not buying power. Projects fell from 538 (FY2021) to 397 over the 15 months to December 2023 (`projectsFY2021`, `projectsFY2023`), while revenue per public-sector project rose from 0.89 (FY2022) to 1.17 (2024). The share of subcontracted work fell by about 9 points of revenue. Revenue per head rose faster than pay. mBank's October 2024 note records management's aim of winding down projects with insufficient profit (sgn-mbank-2024). A representative maintenance unit: the Praca.gov.pl contract (November 2024) has a 48-month maximum of 14.2 gross, about 11.5 net of VAT, or 2.9 a year. At the 2025 group revenue per implementation FTE of 0.466, that is roughly six people. At the group's 2025 cost structure it would earn about 0.7–0.8 of EBIT a year on no tangible capital. This is an illustration from group averages, not a disclosed contract margin. The large unit is NBP: up to 135 gross (about 110 net) over 2024–2027, roughly 27 a year, or 7–8% of 2025 revenue. Acquired companies are the second unit (`edranaNetPrice` … `sagraAnnualProfit`): | | Edrana Baltic (Feb 2024) | Sagra Technology (Mar 2024) | | --- | ---: | ---: | | Price net of acquired cash | 9.4 | 37.6 | | Revenue, annualized from 11 / 10 months | 14.9 | 26.3 | | Net profit, annualized | 2.3 (16%) | 4.3 (16%) | | Price / net profit | 4.0x | 8.7x | | Purchase-price intangibles | 11.3 | 40.7 | | Tangible capital acquired | receivables 1.7, no material PP&E | receivables 2.2, PP&E 0.6 | The two together cost 4.0x–7.1x a cash proxy, depending on whether the reported contributions bear purchase-price amortization (`deals2024LowerMultiple`, `deals2024UpperMultiple`). Their returns are entry yields, not post-integration results. Failure case: e-Taxes. Sygnity won the Ministry of Finance e-Taxes system in 2013 on a fixed-price tender and subcontracted much of it to Fast Enterprises. The resulting court disputes ran for years, through the period of the 2018 creditor restructuring, and in February 2021 Sygnity settled by paying USD3.75m (about PLN14m) into deposit (`fastEnterprisesSettlement`, sgn-fy21-report p.24). The levers failed because, on a new-build tender where price decides, Sygnity carried subcontractor and scope risk without incumbency. The same exposure remains wherever Sygnity bids for greenfield public systems (sgn-fy24-report p.55). Parent against subsidiaries, H1 2026 (added 3 November). The parent company is the organic franchise. In H1 2026 it earned a core operating margin (gross profit less selling and G&A) of 32.6% on 135.9 of revenue, up from 26.1%. The subsidiaries (SBS, Edrana, Sagra, DocLogix, Mednow), together with consolidation entries, earned about 9.9 of pretax on about 77.6 of revenue, roughly 13%, after acquisition amortization. The acquired units therefore run at well under half the parent's margin. Mednow's KPO hospital projects added revenue faster than profit: subsidiary over-time revenue rose +35.8 against a pretax gain of +3.9 (02-history.md). The repricing mechanism described above is a parent-company phenomenon. It has not been shown in the acquired companies, whose results are not reported separately. ## Independent checks 1. **Cash against accounts.** Derived owner cash, after capex, leases, interest and award cost (`ownerCashCY2025`), was 104.4 in 2025 against NOPAT of 72.5 at 19%. The gap is explained by 14.1 of working-capital release, cash tax 9.6 below current tax, and non-cash acquisition amortization of 7.2. The normalized June-trailing bridge removes the first two and gives 94.6 (`normalizedTrailingOwnerCash`). The unit view, high margin on negative capital, matches the group cash record: 2021–2025 growth released net capital (`totalCapitalDeployedWindow` is only 64.9 including 70.1 of deals). 2. **Outside forecasts as a test of the consensus model.** In October 2024 mBank forecast 2025 EBIT of 56.2 (17.6% margin) and net profit of 48.2, with a sell rating and a PLN39 target at PLN58 (`mbankEbit2025Forecast`). Actual 2025 EBIT was 89.5 and net profit 77.3. The consensus treated Sygnity as a tender-driven integrator with a stable margin. The repricing seen in the per-FTE table is what it missed. 3. **Adverse allocation.** Give the deals the most favourable reading: the 11.8 cash proxy for the 2024 deals, plus 0.8 for DocLogix and 1.0 for Mednow's December, is 13.6 of NOPAT. The existing business still produced at least 31.2 of the 44.8 NOPAT increase from 2021 to 2025 (70%) while releasing capital. Then treat all utilities growth since FY2022 (+16.4 of revenue) as temporary regulatory work at the 39.7% incremental margin. That removes about 6.5 of EBIT (7% of 2025), but the conclusion that the incumbent franchise was repriced survives. 4. **Peer check.** Asseco Poland, the largest Polish software house and a competitor in public and banking IT, trades at about 24x trailing earnings on an aggregator's 28 October 2026 figure (asseco-pe-aggregator; not reconciled to filings). No peer filings were read for margins, so the per-FTE comparison is internal only. Where the consensus may still disagree: the market may treat 2025–2026 margins as the new normal. We treat them as the result of a one-time repricing whose durability depends on renewals and wage inflation. The H2 2026 and H1 2027 EBIT margins will settle it (prospective checks). No fieldwork was done: no customer, employee or competitor interviews. Statements by management are evidence of what management says. Sources: sgn-fy21, sgn-fy21-report, sgn-fy22, sgn-fy22-report, sgn-fy23, sgn-fy23-report, sgn-fy24, sgn-fy24-report, sgn-fy25, sgn-h126, sgn-q121, sgn-q122, sgn-presentation, sgn-nbp-contract, sgn-mbank-2024, asseco-pe-aggregator, sgn-asseco-h126. # History, market & cycle Sygnity's history includes the ComputerLand/EMAX combination, the e-Taxes dispute and a 2018 creditor restructuring that was only completed in June 2021 (sgn-fy22 p.8 note). Its current economics should not be inferred only from the recovery off that distressed base. TSS Europe bought 72.68% in a 2022 tender at PLN12 a share (`tssCallPrice`, sgn-tss-call), which made Sygnity an acquisition platform inside the Topicus/Constellation family. Group affiliation brings experience and deal flow. It does not guarantee returns to minority shareholders. Acquisitions since then: Edrana Baltic (February 2024), Sagra (March 2024), DocLogix (July 2025), Mednow, formerly Comarch HIS (December 2025), and g.on (October 2026, from TSS). Closing dates matter. DocLogix is fully in the trailing June 2026 window, Mednow for seven months, g.on not at all. A newly bought company's full-year earnings cannot simply be added on top of its already consolidated contribution. The model uses calendar windows: FY2025 + H1 2026 − H1 2025, provisionally rolled to September with the 2 November preliminary release. The 2025 presentation's revenue bridge suggests about 14% growth excluding businesses bought in 2025. Some chart sub-bars do not add to their headline, so the numerical ledger uses statutory tables. The latest full statements are for H1 2026; the Q3 report is due 10 November 2026, and no Q3 actuals are invented. ## Demand drivers and cycle Who decides and who pays: | Customer | End user | Payer | Decision-maker | | --- | --- | --- | --- | | Ministries and agencies (welfare, labour market, foreigners' permits) | civil servants, labour offices, citizens | state budget, often EU or KPO co-financing | the ministry's IT department and procurement under public-procurement law | | NBP (central bank) | accounting and back-office staff | NBP | NBP IT; renewals and annexes rather than open retenders | | Distribution and sales utilities (PGE group) | billing and metering staff | the utility, which passes costs into regulated tariffs | utility IT, under deadlines set by the regulator (URE, PSE) | | Banks, brokers and depositories | operations staff | bank IT budget | bank IT and compliance, driven by KNF, KDPW and exchange rules | | Acquired niches (retailers, Baltic ERP users, hospitals, airports) | the business user | the customer | the owner or CIO | Demand splits into four parts: - **Replacement and maintenance:** servicing revenue, 144.2 in 2025 (41%). - **Regulation-driven development:** change requests forced by law. Examples are the 2023 electricity price-cap act (Solidarity Shield), the CSIRE energy data hub, PSE balancing rules and labour-market reforms (sgn-fy24-report pp.32-34). - **Discretionary new systems:** tenders. - **Product sales in acquired niches.** Measured cycle (calendar revenue, `revenueCY2020`–`revenue2025`): 218.7, 205.9, 219.1, 229.3, 293.2, 350.3. Revenue fell 6% in 2021, while projects were being pruned and the e-Taxes settlement closed. Servicing went from 75.0 (FY2021) and 75.6 (FY2022) to 70.6 a year over the 15-month FY2023, then to 113.8 in 2024 and 144.2 in 2025. H1 2026 servicing was 96.6 against 66.6 a year earlier, partly from acquisitions. The franchise has not shown a cyclical collapse. A test of public IT budgets through a recession is unavailable, because the 2020 shock coincided with Sygnity's own restructuring. We did not obtain a series of Polish public IT spending or EU-fund disbursement, which is the series that would measure the public-sector cycle. Market misperception and the series that tests it. In 2023–2024 the sell-side saw a tender-driven integrator with a stable mid-teens margin: mBank forecast 17.6% EBIT for 2025 (sgn-mbank-2024). The actual margin was 25.5%. The belief was refuted by the reported EBIT margin and the rising servicing share, which are published every half-year. The present risk is the opposite misreading: treating regulation-timed utility work and the post-repricing margin as permanent. ### Growth decomposition, 2024 to H1 2026 (added 3 November 2026) Sygnity reports one segment. The parent company's separate statements (Sygnity S.A., which holds none of the acquired businesses) isolate the organic core, and IFRS 15 timing separates long projects from the rest. PLN m: | | H1 2024 | H1 2025 | H1 2026 | H1 2026 vs H1 2025 | | --- | ---: | ---: | ---: | --- | | Group revenue | 131.4 | 158.4 | 213.5 | +55.0 (+34.7%) | | Parent products and services (organic) | 99.6 | 116.1 | 135.5 | +19.5 (+16.8%) | | of which parent maintenance | 35.7 | 41.4 | 51.1 | +9.7 (+23%) | | of which parent implementation | 60.7 | 70.5 | 81.2 | +10.7 (+15%) | | Subsidiaries and eliminations (group less parent) | about 32 | 41.8 | 77.6 | +35.8 | | Group revenue recognized over time | 15.5 | 20.5 | 67.1 | +46.5 | | of which subsidiaries | — | 6.3 | 42.1 | +35.8 | | Group revenue at a point in time | 115.9 | 137.9 | 146.4 | +8.5 | | of which subsidiaries | — | 35.5 | 35.5 | 0.0 | | Licences / hardware | 4.7 / — | 6.7 / 0.5 | 10.8 / 1.3 | subsidiaries' licences +4.9 | Sources: sgn-h126 pp.17, 44-45; sgn-h125 p.16; facts `parentRevenueH12024` … `pointInTimeH12026`. Quarterly over-time revenue was 11.3, 9.2, 13.9, 21.7 (2025), then 17.5 and 49.5 (Q1 and Q2 2026). Servicing was 30.8, 35.8, 36.0, 41.6, then 47.8 and 48.8. What drives it: - **Organic core.** The parent grew 13.4% in 2025 (221.5 to 251.3) and 16.5–16.8% in each first half since 2024. Maintenance grew 17% in 2025 and 23% in H1 2026. The parent drove most of the profit: pretax excluding intra-group dividends rose from 32.3 to 46.2 (+13.9), and its core operating margin from 26.1% to 32.6%. Subsidiaries added only +3.9 of pretax (`subsidiaryPbtIncreaseH1`). - **Acquired, and a deadline wave inside it.** All of the subsidiaries' +35.8 of revenue growth is over-time project revenue; their point-in-time revenue was flat at 35.5. It comes from Mednow (ex-Comarch HIS, consolidated from December 2025) and DocLogix (from July 2025). In Q1 2026 Mednow signed nine hospital contracts for its OPTIMED NXT system under the KPO recovery plan, worth 22.8 gross (sgn-q126 pp.33-34). The Ministry of Health requires those projects to be physically completed by 15 July 2026, with spending eligible to 31 August 2026 (kpo-d112-schedule). This is the structure SII flagged: a 2.3x jump in over-time revenue and a −19% fall in point-in-time revenue in Q2 (sii-2q26). We estimate the KPO wave added about 3.2 of pretax in H1 and about 3.8 of owner cash over the trailing year (`kpoWaveOwnerCash`), with a range of about 3–8. The wave ends after August 2026. - **Servicing.** Of the +30.0 servicing increase in H1, +9.7 is parent maintenance, which is organic. The remaining +20.3 is in subsidiaries: Mednow's hospital maintenance and DocLogix are new to the scope, but their split is not disclosed. SII's point that scope changes don't explain the jump is therefore partly answered (a third is organic) and partly open. - **Price versus volume.** This can't be separated from the disclosures. Group revenue per FTE was flat in 2025 (0.416 to 0.407 including acquisitions) and parent average pay rose 5.3%, so 2025 growth was mainly volume and scope. Renewals are being signed at higher ceilings (NBP +25 gross; the Ministry of Family SR/SW renewal at 4.9 gross for 24 months). - **Sectors (H1 2025 vs H1 2024).** Public +28% (57.7 vs 45.2), banking +16%, utilities −8% (29.2 vs 31.6) as the Solidarity Shield and CSIRE wave faded, and retail (Sagra) 12.3 (sgn-h125 p.21). H1 2026 sector figures are not published: the H1 2026 report has no management-report sales table. ### Funding and the timing of demand - **EU recovery funds (KPO/RRF).** The direct exposure is Mednow's hospital work, ending 31 August 2026. Asseco is the larger hospital-software vendor (healthcare revenue 217 against Comarch e-health's about 50 in 2024, itwiz-comarch-his). We found no other Sygnity contract described as KPO-funded. - **Cohesion funds (2021–2027).** We found no Sygnity contract linked to them. - **Polish budget IT spending.** The core is statutory ministry systems financed from the state budget, not EU funds. The tender aggregator lists 37 public awards to Sygnity S.A. worth 195 gross from January 2025 to July 2026, mostly Ministry of Family maintenance and development: WUP-Viator 7.9, CAS 8.4, SR/SW 4.9 and Mazowieckie eDOKweb (mimira-awards, secondary, not reconciled). We did not obtain budget-law line items for 2026–2027, so the budget cycle remains unmeasured. - **Backlog.** Sygnity has published no order backlog (portfel zamówień) in the reports reviewed for 2021–2026. - **Q3 report date.** The issuer's 2026 calendar and current report 29/2026 give 10 November 2026; the 12 November shown by TipRanks and Investing is wrong. Verdict: the 2025–2026 level is mostly new and durable at the parent (statutory maintenance and development, +17–23% maintenance growth), plus a temporary KPO hospital wave inside the acquired Mednow that ends in Q3 2026. The model removes the wave from starting earnings and raises base growth to the through-cycle organic rate (07-assumption-register.md). ## Competition Competition happens in two places. **At tender, for a new system,** Sygnity faces the leading Polish integrators (Asseco Poland above all), international IT firms and consultancies, and price is often decisive (sgn-fy24-report p.55; sgn-mbank-2024). **On an installed system,** competition is weak. The incumbent owns the code, the domain staff and the operational history, so renewals and change requests usually come to Sygnity by annex or extension. Examples are the NBP annexes, six 48-month Ministry of Family contracts in 2024 and Poczta Polska's two-year maintenance renewal (23.6 gross). Advantage in one sentence: Sygnity owns the knowledge and code of systems that the state, the central bank and utilities cannot switch without risking service continuity, so it captures the maintenance and regulatory change work on them. Evidence that this is durable is limited to contract continuity in the filings. We found no lost renewal of a major system in 2021–2026, but the filings would not necessarily disclose one, and we did no customer interviews. Competitor count: the issuer expects consolidation among the largest players and more foreign entrants (sgn-fy24-report p.55). Topicus, which controls Sygnity, also owns about 24.8% of Asseco Poland: 9.99% bought from Cyfrowy Polsat, plus 14.84% of treasury shares at PLN85 closed on 1 October 2025 under a shareholders' agreement with Adam Góral (topicus-asseco). Sygnity's controller is therefore a large holder in its main competitor. That may soften head-to-head price competition, but it also gives TSS a choice of vehicle for Polish deals and talent. Comarch, which began a strategic-options review in January 2025, sold its hospital-software unit (Comarch HIS) to Sygnity in December 2025 (itwiz-comarch-his), so one competitor exited one niche. Share of the Polish IT market is not measured here. The competitor that could hurt Sygnity is Asseco Poland. It has the scale and public-sector references to underbid when a ministry or NBP retenders a core system or replaces it in an EU-funded modernization. It hasn't done so visibly, because the customer bears the migration risk, and because the state's in-house IT centres, not rival vendors, are the more likely route to replacing outsourced maintenance. That second path is a hypothesis we have not tested. ## Counterparty incentives | Counterparty | What they want | What Sygnity gives them | Their outside options | What would change the arrangement | | --- | --- | --- | --- | --- | | TSS Europe / Topicus / Constellation (72.68% owner) | high-return deployment of cash into vertical software, long holding periods | a listed CEE platform with surplus cash, local M&A capacity and a profitable franchise | buy out minorities (the 2022 call reserved a future delisting "if an opportunity arises"); route CEE deals to other TSS units; sell group assets to Sygnity | a cheap price relative to value (buyout), or TSS needing a home for assets it owns | | Ministries and agencies | uninterrupted statutory services at a defensible price | continuity, domain staff, fast regulatory changes | retender (public-procurement law), insource to state IT centres, replace in EU-funded programmes | price rises that attract audit scrutiny, a failed delivery, a political push for state-owned IT | | NBP | reliable accounting and back-office systems | long-standing system knowledge | retender at the end of 2027 | delivery failure or a platform change | | Utilities | compliance with regulatory deadlines | billing and metering systems already adapted | other billing vendors, group IT | consolidation of utility IT platforms | | Employees (the scarce input) | pay, stability, career | pay up 37% per FTE 2021→2025 at the parent; incentive plan for up to 149 managers | other IT employers | wage inflation above price increases | Dependence runs both ways. A ministry depends on Sygnity for systems it cannot easily rebuild. Sygnity depends on renewals that can be contested by tender. TSS gives discipline and deal flow but can also extract value: g.on came from TSS for EUR7.2m, about 1.5% of market value, with no stand-alone financials disclosed. TSS had flagged it four times since 2022 (current reports 40/2022, 25/2023, 9/2024, 15/2024; sgn-gon). Headlines about a counterparty's action show what it did, not why. We have no evidence of TSS's motive in the g.on sale beyond the stated strategy. Sources: sgn-tss-call, sgn-fy21-report, sgn-fy22, sgn-fy22-report, sgn-fy23-report, sgn-fy24-report, sgn-fy25, sgn-h126, sgn-nbp-contract, sgn-gon, sgn-mbank-2024, sgn-presentation, sgn-calendar. # Management & incentives Two layers decide where Sygnity's cash goes. The six-person management board runs operations and proposes deals. TSS Europe (Topicus, in the Constellation family) owns 72.68%, appoints the supervisory-board majority and sets the capital-allocation frame. The model uses all 22,759,947 issued shares (unchanged since 2019) and charges award cost once. Money in PLN. ## Ownership and incentives | Holder | Shares (3 Sep 2026) | Value at PLN94 | 2025 pay | How acquired | Background | | --- | ---: | ---: | ---: | --- | --- | | TSS Europe B.V. | 16,542,399 (72.68%) | 1,555m | — | 2022 tender at PLN12 (about PLN199m) | Topicus/Constellation vertical-software acquirer | | Barca Global Master Fund | 2,050,679 (9.01%) | 193m | — | market | fund | | Maciej Różycki, CEO since May 2022 | 40,538 | 3.8m | 2.33m (base 0.90, bonus 1.39) | incentive transfers (25,663 at May 2026) | TSS investment and acquisition background | | Mariusz Jurak, vice-chair, board since before 2021 | 31,486 | 3.0m | 1.28m | incentive transfers | long-standing Sygnity executive | | Adam Larysz, board from 2025 | 20,552 | 1.9m | 1.79m | incentive transfers | board member responsible for a business sector | | Daniel Fryga, vice-chair from 2025 | 12,599 | 1.2m | 1.95m | incentive transfers | not reviewed | | Dorota Beksińska, board from 2025 | 10,485 | 1.0m | 1.20m | incentive transfers | board member responsible for a business sector | | Monika Zientarska, CFO since Sep 2023 | 6,959 | 0.7m | 0.94m | incentive transfers | finance | | Supervisory board: three TSS-linked members (unpaid), two independents (paid 0.14m each) | — | — | — | — | Eggink (audit chair) has decades of WSE board service | Sources: sgn-h126 p.27; sgn-rem25 pp.6-11; sgn-fy24-report (management and supervisory-board biographies). No open-market purchases or sales by board members were found in 2025–2026. Holdings rose only through incentive transfers, including 56,266 shares in May 2026. Whose interests dominate. Executives' wealth is mostly pay. If the stock compounded at 10% a year for five years, the CEO's 40,538 shares would gain about PLN2.3m, about one year of his 2025 pay. TSS's interest is almost entirely share value: a 10% move in the price changes its stake by about PLN156m. The management board as a whole cost PLN9.49m in 2025 (`boardPay2025`). That is 12% of 2025 net profit and 2.7% of revenue, a heavy charge for a company with 860 employees. Compensation metrics (sgn-rem25 p.14): consolidated net profit, EBIT, ROIC and revenue growth, plus four non-financial criteria (staff retention, succession, diversity, energy use). Variable pay is capped at 500% of fixed pay. 75% of each bonus is paid in shares, with lock-ups of up to five years. Targets are not published, so the implied hurdle can't be backed out. Including ROIC is better than earnings alone, though revenue growth and EBIT still reward size. The 2024–2028 programme can award up to 442,257 shares (1.9%), bought back or issued (`incentivePlanCap`). So far it has been funded by buybacks: 90,806 shares were bought for PLN6.4m in February–April 2026 at about PLN70. ## Track record and references - **Shareholder return under TSS:** from the PLN12 tender price (March 2022) to PLN94 (3 November 2026) is 7.8x, about 56% a year over 4.6 years. - **Earnings:** EBIT rose from 25.4 (FY to September 2022, 12.0% margin) to 106.0 trailing June 2026 (26.2%). Derived owner cash rose from 23.9 (2021) to 104.4 (2025). - **Valuation:** the year-end price less cash, divided by net profit, went from 13x (2022) to about 24x (2023–2025), then fell to about 19x now (`peExCash2022`–`peExCash2025`). - **Return on capital:** including goodwill and acquired intangibles, about 34% after tax in 2025 (01-business.md). Working capital released cash throughout. Decisions that failed or are unproven: - **Idle cash.** Cash went from 52 (December 2021) to 225 (June 2026) with no dividend, and only about 35 a year is being deployed. - **Related-party deal.** The g.on purchase from TSS has no disclosed stand-alone economics. - **Deal outcomes.** No deal's results are reported after its first year. Bench depth beyond the leader: the board grew from three to six in January 2025 (Beksińska, Fryga, Larysz; sgn-fy24-report). The TSS-linked supervisory members bring CSU-family M&A practice; Ioana Corutiu, TSS CFO, has led more than 25 due diligences and 11 integrations (sgn-fy24-report, supervisory-board biographies). References: none collected. No former employees, customers or counterparties were interviewed, so zero of zero references. The only outside view on record is mBank's 2023–2024 sell rating (target PLN35.80, then 39), and the stock has since more than doubled. ## Capital allocation principles Governing objective: Sygnity has stated none. The controlling shareholder's family manages for return on invested capital, and Sygnity's bonus includes ROIC. Since 2022 the behaviour is consistent with long-term value per share for TSS. For minorities, the open question is whether cash builds up or is returned. 1. **Zero-based allocation.** Partly met. Unprofitable projects were wound down (538 projects in FY2021, 397 over the 15-month FY2023). Capitalized development stopped: 5.2 in 2022, zero in 2025. Staff moved to the sectors that renew. There is no evidence of capital moving away from an incumbent franchise, because none is in decline. 2. **Fund strategies, not projects.** Met in form. The stated strategy is to "invest surplus cash in building a group of independent IT solution providers" (sgn-fy24-report p.39), and five small deals have followed it at a steady pace. 3. **Capital has a cost.** Met. There have been no share issues since 2019, and debt and restructuring liabilities of 69.6 were repaid in 2021–2025. Share awards are charged and bought back rather than diluting. The cost of idle cash is not acknowledged: about 194 of cash net of the g.on payment earns deposit rates. 4. **Zero tolerance for bad growth.** Met. Revenue fell 6% in 2021 while low-margin work was pruned, and the project count fell 26% to FY2023. 5. **Price versus value.** Unproven. Buybacks have only offset awards, and none were made for value even at 13x earnings in 2022. There has been no dividend, so cash has not been returned when deal flow was thin. Deals have been bought at 4–9x profit, which is cheap against Sygnity's own 19–24x multiple. That is the right direction for value per share, if the profit holds. ## Integrity and governance Evidence: - **Related-party deals.** One with TSS (g.on, EUR7.2m initial price, true-up undisclosed), announced under MAR with prior flags since 2022. - **Board.** Two of five supervisory members are independent, and an independent chairs the audit committee. TSS-linked members take no fee. - **Votes.** The AGM approved the 2024 remuneration report. - **Pay rules.** No deviations from the remuneration policy in 2025, and no clawbacks (sgn-rem25 pp.3 and 24). - **Retained profit.** The June 2025 AGM put the 2024 profit (60.1) into "capital for the payment of dividends and investment purposes" (sgn-fy25 p.7). That keeps a dividend possible without committing to one. - **Charges from the controller.** The group buys services from TSS Europe: 8.4 in 2024, 11.1 in 2025 and 8.4 in H1 2026 alone, against 5.4 in H1 2025 (`tssServicesPurchased2024` … `tssServicesPurchasedH12026`). The nature of the charge is not disclosed. In H1 2026 it equalled 18% of net profit, and it is growing faster than profit. It is already expensed, so owner cash is measured after it, but it is a channel for value to reach TSS ahead of minorities. A prospective check tracks it. - **Controller's stake in the main competitor.** Topicus holds about 24.8% of Asseco Poland (closed 1 October 2025, topicus-asseco). The three agency costs: - **Control of resources:** moderate. Bonuses on revenue growth and EBIT reward size, and the board sources deals. ROIC in the bonus and TSS's hurdle culture are the offsets, and the deal pace has stayed small (about 35 a year). - **Risk tolerance:** low concern. TSS is diversified, and management holds little stock. - **Time horizon:** mixed. Bonuses are annual, but paid mostly in locked-up shares. Voting and economic ownership coincide: one share, one vote. TSS controls both. The main agency cost for an outside holder is the controller's discretion over idle cash, related-party transfers and any future buyout price. How the valuation treats it: - Unplaced cash earns nothing and is not distributed. - Probabilities carry a 30% five-year downside weight. - No discount is applied for a possible take-under, because there is no evidence of one. The 2022 call document's delisting reservation is recorded as a risk in 06-risks.md. Sources: sgn-rem25, sgn-h126, sgn-fy24-report, sgn-fy25, sgn-tss-call, sgn-gon, sgn-mbank-2024. # Capital allocation PLN millions, calendar years. 2024 and 2025 are statutory years. 2023 is the audited calendar comparative (sgn-fy24 note 44). 2021 and 2022 are derived from the audited fiscal years and the Q1 interims; each year reconciles to the cash balance within PLN0.001m (facts `cfoCY2021` … `cashEndCY2023`). Rates and returns are named audit steps in `valuation/calculations.json`. ## Sources and uses of cash | | 2021 | 2022 | 2023 | 2024 | 2025 | Total | | --- | ---: | ---: | ---: | ---: | ---: | ---: | | Operating cash flow | 36.4 | 55.1 | 54.8 | 84.1 | 114.7 | 345.2 | | of which working capital released / (invested) | (7.5) | 18.1 | 0.2 | (0.1) | 14.1 | 24.8 | | of which incentive add-back (non-cash) | 1.0 | 0.7 | 1.3 | 0.9 | 4.1 | 7.9 | | of which tax paid | +1.0 refund | (3.0) | (0.6) | (1.8) | (8.6) | (13.0) | | Capex and capitalized development | (2.8) | (8.2) | (3.8) | (3.1) | (1.8) | (19.6) | | Lease principal | (7.6) | (7.8) | (5.7) | (4.9) | (4.0) | (30.0) | | Interest paid | (1.1) | (0.6) | (3.2) | (1.4) | (0.5) | (6.7) | | **Owner cash, after award cost (memo)** | **23.9** | **37.7** | **40.9** | **73.9** | **104.4** | **280.9** | | Acquisitions | — | — | — | (34.3) | (35.7) | (70.1) | | Debt, bonds and restructuring agreement, net | (15.6) | (18.0) | (12.0) | (12.0) | (12.0) | (69.6) | | Buybacks (award delivery) | — | — | (0.3) | (0.5) | (4.1) | (4.9) | | Dividends | — | — | — | — | — | 0 | | Other investing and financing, FX | 0.0 | 0.0 | 0.7 | 0.2 | (0.1) | 0.8 | | **Net change in cash** | **9.3** | **20.4** | **30.6** | **28.1** | **56.6** | **145.1** | | Cash at year end (42.6 at December 2020) | 51.9 | 72.4 | 102.9 | 131.1 | 187.7 | | The memo line is operating cash flow less the award add-back, capex and development, lease principal and interest. The non-cash add-back is charged there and the cash buybacks appear separately. 2021 operating cash includes the February 2021 e-Taxes settlement (about 14) and 7.5 of working-capital build. 2022 is flattered by 18.1 of working-capital release. Cash tax was minimal until 2025 because of loss carry-forwards. What the table shows about priorities. Of 345 of operating cash over five years: - about 20% (69.6) retired the legacy bonds, bank loans and restructuring-agreement claims; - 20% (70.1) bought companies, from 2024 only; - 9% went to leases and 6% to capex and development; - 1.4% bought back shares to deliver awards, and nothing was paid in dividends; - 42% (145.1) accumulated as cash. The allocator's revealed order was debt first (2021–2025, loans fully repaid in 2025), then deals at about 35 a year. The surplus beyond that was held, and cash then rose further to 225.1 by June 2026. ## Investment rates per sales change The window runs from calendar 2020 (revenue 218.7, `revenueCY2020`) to 2025 (350.3), a change of 131.6 (`salesChangeWindow`). | Rate per PLN of revenue change | 2021–2025 | Calculation | | --- | ---: | --- | | Incremental M&A (acquisitions, no disposals) | 0.53 | 70.1 / 131.6 (`maRateWindow`) | | Internally generated intangibles: capitalized development | 0.07 | 9.3 / 131.6 (`developmentRateWindow`); expensed development is not disclosed | | Fixed capital, gross capex (before D&A) | 0.08 | 10.3 / 131.6 (`fixedRateGrossWindow`); an upper bound | | Working capital | −0.19 (released) | 24.8 of cash released (`workingCapitalReleasedRate`) | | **Total** | **≈0.49** | | Definitions and limits: - **Fixed capital.** This rate is gross. Own-asset D&A was not separated from right-of-use and acquisition amortization for 2021–2023. In 2024–2025, capex plus development (4.8) was below own-asset D&A (about 7.4: total D&A less right-of-use depreciation and acquisition amortization), so the true net rate is close to zero or negative. - **Working capital.** The cash-flow-statement change, excluding working capital that came with acquired companies. - **Organic versus acquired revenue.** Acquired companies contributed roughly 51 of 2025 revenue (an estimate: Edrana and Sagra at their 2024 run rates of 14.9 and 26.3, DocLogix 5.0, Mednow 4.3). The organic change is therefore about 80, and it required negative net capital: −24.8 of working capital against +19.6 of capex and development. A zloty of acquired revenue cost about 1.4 of deal capital. A zloty of organic revenue cost nothing. Inflection points and who was in charge: - **2021–2022, under the pre-TSS board (Zborowski to December 2021, Jurak, Jędrzejewska).** Debt repayment dominated, and capitalized development rose to 5.2 in 2022. - **2022 working-capital release of 18.1.** Most of it came in Q1 FY2022 (21.3 in October–December 2021), before TSS took control in May 2022. - **From 2023 (Różycki CEO from May 2022, Zientarska CFO from September 2023).** Capitalized development fell to 2.9 in 2023 and zero by 2025. Deals started in 2024. Working capital released 14.1 in 2025 as customer prepayments rose by 11.0. No peer rates were computed. ## Reinvestment and returns | | 2021 | 2025 | | --- | ---: | ---: | | EBIT | 34.2 (16.6%) | 89.5 (25.5%) | | NOPAT at 19% | 27.7 | 72.5 (78.3 before acquisition amortization) | | Derived owner cash | 23.9 | 104.4 | | Normalized owner cash (model) | — | 94.6 trailing June 2026; 111.3 for the current portfolio | - **Reinvestment rate, 2021–2025.** Acquisitions of 70.1 against derived owner cash of 280.9 is 24.9% (`acquisitionReinvestmentRate`). Including capex and development and netting the working-capital release, net growth capital was 64.9 (`totalCapitalDeployedWindow`), about 22% of owner cash before capex. The rest went to debt (69.6) or stayed as cash (145.1). - **Incremental return.** NOPAT rose by 44.8. Against net growth capital of 64.9 that is 69% unlagged (`incrementalReturnWindow`). Lagged one year, capital 2021–2024 was about 41.5 and the return about 108%. These numbers do not measure capital efficiency. Most of the gain is operating leverage and repricing, and it was not bought with capital. - **Identity check.** NOPAT grew 27% a year from 2021 to 2025, while reinvestment (22–25%) × incremental return (69%) gives about 16%. The identity fails because the margin recovery needed no investment. Split by use, it reconciles: - **Volume at the old margin.** Organic revenue of about 299 in 2025 (350 less about 51 acquired), at the 2021 EBIT margin of 16.6%, gives EBIT of about 49.7 and NOPAT of 40.2. That is +12.5 of NOPAT from volume. - **Margin.** The remaining organic gain is about 19–24 of NOPAT, from the move to a 24–26% margin on the organic base. It needed negative capital. - **Acquisitions.** About 8.5–13.6 of 2025 NOPAT on 70.1 of deal cash, a 12–19% yield in the first full year. It is not yet a rolling or lagged measurement. - **Payouts.** None. Buybacks of 4.9 only offset awards. - **Forward implication.** The margin step can't be repeated at the same scale. Forward compounding comes from organic volume, which has been capital-light, plus deals at 12–19% entry yields on a measured base of two deals. The model's family-aggregate 6.61x (15.1%) sits inside that range. The model also assumes cash that can't be placed earns nothing, which is consistent with the record. ## Acquisition record | Closing | Deal (seller) | All-in price | Cash acquired | Revenue disclosed | Net profit disclosed | Multiple | Purchase-price intangibles | Later evidence | | --- | --- | ---: | ---: | --- | --- | --- | ---: | --- | | Feb 2024 | Edrana Baltic, Lithuanian and Latvian ERP (private seller) | 10.99 (EUR2.54m; 7.46 paid at closing) | 1.58 | 13.6 for 11 months | 2.15 for 11 months | 4.0x annualized net profit on net price | 11.3 | none disclosed after 2024 | | Mar 2024 | Sagra Technology, Szczecin, retail sales and analytics SaaS (founders) | 40.82 (31.66 at closing; cap 42.3) | 3.20 | 21.9 for 10 months | 3.60 for 10 months | 8.7x | 40.7 | none disclosed | | Jul 2025 | DocLogix, Vilnius, document management (private) | 7.42 (EUR1.75m) | 0.51 | 5.0 for 6 months | 0.39 for 6 months | 4.35x–9.04x cash proxy (existing DocLogix steps) | 8.1 | — | | Dec 2025 | Mednow, formerly Comarch HIS, hospital information systems (Comarch) | 24.40 at year end (22.61 after the June 2026 update) | 3.03 | 4.3 for December | 1.04 for December | not measurable from one month; model proxy at 6.61x | 26.2 | — | | Oct 2026 | g.on experience, Münster, geolocation infrastructure software (TSS, related party) | EUR7.2m initial, about 31.5 | not disclosed | not disclosed | not disclosed | model proxy at 6.61x | — | true-up pending | Aggregate. About 101.6 of cash over 33 months (February 2024 to October 2026) is about 37 a year, roughly a third of current owner cash. For the two deals with a full year's disclosure, the price was 4.0x–7.1x a cash proxy (`deals2024LowerMultiple`, `deals2024UpperMultiple`) and 1.15x annualized revenue. The lower bound credits back purchase-price amortization. The issuer does not report acquired companies' later results, and its organic-growth figure (about 14% in 2025) excludes only 2025 purchases. Edrana's and Sagra's 2025 growth is therefore inside "organic", and the measured deal yield can't be rolled forward. Post-integration margins are not disclosed. The measured multi-year deal yield remains a bound on two first-year results, so the model prices future deals at the CSU-family aggregate (6.61x) and uses Sygnity's own range as a cross-check. ### Deployment record (measured 4 November 2026) | Calendar year | 2021 | 2022 | 2023 | 2024 | 2025 | 2026 to date | | --- | ---: | ---: | ---: | ---: | ---: | ---: | | Deal cash | 0 | 0 | 0 | 34.3 | 35.7 | 35.1 (g.on 31.5 + deferred 3.6) | - **Average:** 14.0 over 2021–2025 including the dry years (`recordMean2021to2025`); 26.3 over the TSS era 2023–2026 (`tssEraMean`); 35.1 over the active era 2024–2026 (`activeEraMean`). - **Spread:** the best year (35.7) is 2.55 times the five-year mean, and the worst is zero. Within the active era the spread is only ±2% (34.3–35.7). - **Quiet runs:** the longest is three years (2021–2023). - **Reference class:** 2021–2022 fall in the creditor restructuring before TSS control, a different regime. 2023 was TSS's setup year; the first deal was signed in December 2023. The base uses the expected budget: active-era pace × the TSS-era active-year frequency of 0.75 = 26.3, grown 3.4% a year (debate verdict). It places no quiet year in a specific year, because no deal timing is evidenced. The downside repeats the three-year dry spell, then deploys the five-year mean. ## Current owner-cash bridge and deployable cash The starting point for valuation is consolidated cash after funding the existing business, not an adjusted EBITDA multiple. Development staff and delivery spending remain operating expenses. Sustaining purchases and lease cash are charged. Amortization of acquired customer relationships is not charged a second time as a cash replacement expense. | Trailing September 2026 bridge (FY2025 + 9M 2026 − 9M 2025, reported 10 Nov) | PLN m | | --- | ---: | | Reported operating cash flow | 160.767 | | Remove working-capital cash contribution | -48.437 | | Replace paid tax with current tax expense | +2.657 | | Retain employee-share economic cost | -7.695 | | Sustaining tangible/intangible cash investment | -1.450 | | Cash lease principal | -4.677 | | Financing paid outside operating cash | -0.073 | | Remove after-tax cash/financial interest | -3.671 | | Normalized trailing owner cash (current-tax definition) | 97.421 | | Add back current tax on the excluded working-capital timing | +9.270 | | September earning power | 106.691 | The tax adjustment replaces PLN36.354m paid with PLN33.697m of current expense. Current tax itself is inflated by tax on customer advances: the deferred-tax asset on deferred income rose 5.1 and the contract-revenue deferred-tax liability fell 4.2 over the year. The bridge already excludes the 48.4 working-capital inflow those advances created, so it excludes the tax on them too (19% of 48.4 is 9.2). The missing Mednow months (0.534) and the g.on proxy (4.767) give current-portfolio owner cash of 112.0. These are inferred additions, not reported target results; see 07-assumption-register.md. The measured KPO wave (10.2) is removed and the remaining measured base (96.5) is rolled 0.57 years at the 2025 organic growth of about 14% (+7.7), giving the 109.5 anchor run rate. Retained cash earns the measured 1.56% after-tax deposit yield, and deals earn half a year in their purchase year. September cash of 281.080 is reduced by the g.on initial payment (31.496; closed 2 October, after the balance sheet) and an operating reserve of 73.915 (three months of nine-month cash costs), leaving 175.669 before year-one claims: 9.219 of acquisition and tax payables and 24.447 for the after-tax reversal of the 2026 customer-advance build (contract liabilities 25.6 in September 2025, 85.8 in September 2026). Reserve cash, the joint venture and financial receivables remain assets at face value. Deal cash is a stated yearly budget (08-underwriting-plan.md). Base: 26.3 in year 1, growing 3.4% a year to 35.5 in year 10 (expected budget including quiet-year frequency: active-era pace 35.1 × 0.75). Downside: 0 for three years, then 14.0. Upside: 35.7, 44.2, 44.2, 58.4, 58.4, then 95.0. Idle cash earns the measured after-tax deposit yield (1.56%). The PLN80m facility is not added to cash. Sources: sgn-fy21, sgn-fy22, sgn-fy23, sgn-fy24, sgn-fy25, sgn-h126, sgn-q121, sgn-q122, sgn-presentation, sgn-gon, sgn-facility, sgn-fx. `research/10-capital-efficiency.md` holds the engine's forward capital-productivity diagnostics. # Thesis & valuation Sygnity can compound if two things hold: the repriced incumbent franchise keeps its margin as wages rise, and management buys more software cash at attractive all-in prices instead of letting cash pile up. The causal links are renewals and regulatory change work, cash conversion after staff costs, the actual deal pipeline, and value retained for all common shareholders. TSS affiliation is context, not a substitute for those tests. No supplied Sygnity thesis or target return exists; the company is underwritten independently from filings. The driver plan in 08-underwriting-plan.md was written before returns were calculated. | At PLN97.00 on 10 November 2026 (exits fixed at 27 Oct 2031 / 2036) | 5-year XIRR | 10-year XIRR | | --- | ---: | ---: | | Downside: revenue −2% then flat, owner-cash margin 24.8% → 23.3% from year 2, three dry deal years then 14.0 a year, 11.45x | −2.7% | 1.7% | | Base: organic revenue 9.5% / 7% / 5.5%, margin held at 24.8%, expected deal budget 26.3 (+3.4% a year; active-era pace × 0.75 active-year frequency) priced at 1.06x revenue with a 16.2% margin, 15.97x | 10.3% | 10.2% | | Upside: organic revenue 13% / 11% / 9%, margin 25.3% from year 3, deal budget growing 15%, 15.97x | 14.7% | 14.4% | | Probability-weighted cash flows (30/50/20; 35/40/25) | 8.08% | 9.27% | The 10 November roll to the full Q3 report (07-assumption-register.md, 10 November section) replaced the provisional September bridge with reported cash, tax, balances and claims, and measured the KPO wave at 10.2 of owner cash (was a judged 6.06). Run-rate owner cash fell from 113.3 to 109.5; Q3's retained cash raised opening funds, less a 24.4 year-one claim for the reversal of the 2026 customer-advance build. Margins are now the measured starting margin (24.8%, higher because the lower-margin KPO revenue left with its cash) with the same stated downside compression and upside gain. At PLN97 on 10 November, the five-year weighted figure moved from 8.74% to 8.08% (−0.66 points; ten years 9.62% to 9.27%). The changes were the anchor run-rate starting point (119.8 instead of 111.3), deals earning half a year in the purchase year, a measured 1.56% after-tax return on retained cash, 3.4% growth of acquired businesses, and a 12.2x downside exit. Before the calibration the figures were 1.5/12.4/15.1% and 10.2% weighted over five years. The growth-durability review on the same day removed the KPO hospital wave from starting earnings (119.8 to 115.7) and raised organic growth to 6.51%/11.0%. The five-year figures moved from 1.3/14.1/16.9% (11.5% weighted) to 0.6/14.7/18.8% (12.2% weighted). The bull–bear debate on the same day (12-debate.md) then raised the KPO wave to 6.06 (baseline 113.3), set base deal-capacity growth to 3.4% (a flat deal count), and replaced the 18x and 12.2x exits with a Koller value at a 10.45% market cost of equity (16.0x and 11.5x). Those changes give the table above. Revenue build (3 Nov): the model now projects revenue and margin on yearly paths (08-underwriting-plan.md, Yearly paths). Before it, the debate-judged figures were −0.6/11.9/16.1% and 9.7% weighted (five years). Lumpy deal budget (4 Nov): yearly deal cash from the measured record replaced the smooth cap, moving the five-year weighted figure from 9.65% to 9.39%. The base uses the expected budget including quiet-year frequency. There is no common dividend. Returns come from terminal per-share equity, including retained cash and separately held assets, after claims. The multiple prices normalized after-financing owner cash. The weighted figure is the IRR of probability-weighted dated cash flows, not an average of IRRs. ## Pre-mortem - **One-foot or three-foot hurdle?** One-and-a-half feet. The cash mechanics are simple: high margin, negative working capital, no debt. The durability of renewals with ministries and NBP can't be verified from outside, and contract margins are not disclosed. - **Cyclical or secular?** Mostly secular: maintenance and regulatory work on installed systems. There is a regulation-timed layer in utilities (Solidarity Shield, CSIRE), and public budgets are tied to EU-fund cycles that we did not measure. - **How could it be a value trap?** Cash accumulates (225 in June 2026, rising about 70 a year beyond deal spending) while the controller neither deploys nor distributes it. Minorities would then own a growing pile of cash earning about 1.6% after tax. The base case leaves 697 of cash on the balance sheet in year five. - **Cigar butt or productive asset?** A productive asset. Return on capital including goodwill is about 34% after tax, and growth has released working capital. - **Agency cost?** Yes. It is the controller's discretion over idle cash, related-party deals (g.on) and any future buyout price, plus management-board pay of 9.5 (12% of 2025 net profit). - **Good or bad execution?** Assume CSU-family discipline on deal prices (the record so far is 4–9x profit). Do not assume a repeat of the 2022–2025 margin step, which is a one-time repricing. - **Turnaround?** It was one, and it has succeeded: EBIT margin from 12% in FY2022 to 26%. Underwriting from here is not a turnaround bet, but it inherits the risk that a turnaround margin regresses. - **Big promises, little record?** The deal strategy has only two and a half years of record and five small deals, with no post-acquisition results reported. - **Unique IP a larger competitor could squash?** The intellectual property is system-specific domain code. Asseco could underbid at a retender, but the customer's switching risk protects the incumbent. - **Single product or fad?** No. Revenue is spread across public (33%), banking (29%), utilities (22%) and acquired niches. NBP is the largest identifiable relationship at up to about 7–8% of revenue. ## Exit multiple | Evidence | Multiple | Denominator | | --- | ---: | --- | | Sygnity year-end market value less cash / net profit: 2022, 2023, 2024, 2025 | 13.3x, 21.6x, 24.0x, 23.8x | `peExCash2022`–`peExCash2025` | | Today: PLN2,139m less 193.6 of cash after the g.on payment / provisional trailing September net profit 104.6 | 18.6x | equity ex-cash | | Today: the same / normalized portfolio owner cash 111.3 | 17.5x | model denominator | | Peak (30 July 2025, PLN118) | about 40x trailing profit | | | Asseco Poland, trailing P/E (aggregator, 28 Oct 2026; 17.7x in July 2026) | 24.2x | equity, not reconciled | | Pre-TSS (2019–2021, PLN2–10 during the creditor restructuring) | not comparable | distressed | Koller check: value = earnings × (1 − g/RONIC)/(r − g). Base terminal growth of 5% is made of about 3% organic growth (near-zero capital) and about 2% from deals at a 15% yield, which needs reinvesting about 13% of earnings. That gives (1 − 0.13)/(r − 0.05). **Debate judgment (3 November 2026).** The cost of equity is now estimated, not assumed: - **Build-up.** The 6.31% PLN 10-year yield on 3 Nov 2026 (te-pl10y), less Poland's 0.72% default spread, plus Damodaran's 5.33% total ERP (damodaran-ctryprem, January 2026), gives 10.92%. Subtracting the spread means country risk is counted once (damodaran-riskfree). - **Market-implied.** The 17.5x paid on 3 Nov implies 9.98%. - **Judged rate.** The two are weighted equally: r = 10.45% (`judgedCostOfEquity`). The base and upside exit is 0.87/(0.1045 − 0.05) = 16.0x (`baseExitMultiple`). The 18x it replaced needed r of about 9.8%, below both estimates. The zero-growth value at that r is 9.6x, so the downside is now 11.5x, the midpoint with the 13.3x end-2022 trough (`downsideExitMultiple`). It was 12.2x at the 9% owner hurdle, and 14x before 3 November. Higher growth in the upside still earns no higher multiple. The Asseco figure in the table is unusable: aggregators show 9–24x depending on the denominator. The bear case below tests 12x. ## Bear case The price we could sell at after five years if we were not forced sellers, given a specific onerous mechanism: - **Margin.** Wage inflation and tender pricing pull EBIT margin back from about 26% to 21%, roughly the 2024 level. That cuts owner cash by about 20%, to about 96 from the 119.8 run rate. - **Growth.** None after that. - **Deals.** TSS sends CEE deals elsewhere, or none clear the hurdle, so cash builds at zero return. - **Exit.** 12x, between 2022's 13.3x and the zero-growth value. Result at PLN94: 5-year XIRR −3.0%, 10-year +1.1% (−4.5% and 0.0% before the calibration). Net cash, now earning interest, and retained earnings limit the loss: the business still earns about 96 a year, kept as cash. Tails this omits: - an e-Taxes-type fixed-price failure with penalties and guarantee calls (13.1 of guarantees outstanding); - loss of the NBP contract at its end-2027 expiry or of the Ministry of Family systems; - related-party value transfer through repeated purchases from TSS; - a controller buyout at a price below value, or a delisting that impairs liquidity; - Polish wage inflation well above price increases for several years. ## Variant view This is the judged view after the bull–bear debate (12-debate.md), not either advocate's. **History.** The market's mispricing ran in our favor from 2022 to 2024. It treated Sygnity as a tender-driven integrator: mBank had a sell rating, forecast 17.6% EBIT for 2025, and the stock ended 2022 at 13x earnings. The unit economics in stage 3 (EBIT per employee from 37k to 104k) and the measured returns in stage 6 showed the repricing the market missed. **That gap has closed.** **Where the debate left the business.** It survived: - the parent company's organic growth of 13–17% and maintenance growth of 17–23% are measured and durable in the near term; - there is net cash, and no plausible permanent loss appears in the representative paths. What did not survive was the valuation at PLN94: - **Cost of equity.** A cost of equity built from the 6.31% PLN 10-year yield is about 10.9%. Even averaged with the rate the market implies today, it is 10.45%. That supports about 16x for a 5% grower, not the 18x the base used. - **KPO wave.** Q3 profit points to a larger deadline wave (6.06 of owner cash, not 3.85). - **Deal pace.** Three flat years support a flat deal count, not 10% capacity growth. **Returns.** On the judged model the base returned 11.9% and the probability-weighted figure 9.7% over five years at PLN94 on 3 November (10.5% over ten). After the 10 November Q3 roll the base is 10.3% and the weighted figure 8.1% at PLN97. That is below a 12% hurdle. The price implies about 3.3% organic cash growth at an 18x exit for a 9% return (no-deal diagnostic, 08-underwriting-plan.md). At 16x the implied growth is higher. **Verdict: fairly priced to slightly expensive for a controlled, low-float company. There is no mispricing in our favor at PLN97 (10 November 2026).** The five-year weighted return reaches 12% near PLN81 (was near PLN85 before the Q3 roll). The deal option on idle cash adds about 2.2 points of base return over no deals (11.9% against 9.6%). It rests on CSU-family purchases at about 15% entry yields. There is no hard catalyst. Possible ones: - a faster deal pace; - confirmation in the 10 November report that the KPO wave was small. ## Why the opportunity exists - **Size:** PLN2,139m market value (22.76m shares at PLN94). - **Float:** holders below 5% own 4.17m shares (18.3%), about PLN392m (`freeFloatShares`). TSS (72.68%) and Barca (9.01%) do not trade. - **Liquidity:** median daily trading over the last three months was PLN0.78m, mean 1.57m (`medianDailyValue3m`). Turning over the free float takes about 500 trading days at the median, or 250 at the mean. - **Coverage:** mBank covered the stock with a sell rating in 2023–2024. A search found no 2026 broker target. Short interest, index membership and retail ownership were not checked. - **What drove sellers:** the stock fell from 118 (July 2025) to the low 70s in early 2026 while earnings rose, as the multiple compressed from about 40x towards 20x. It has since recovered to 94. Neglect is real: low float, a controller, and coverage that was wrong and has lapsed. Part of the discount is justified, though. Minorities can't direct cash, deal terms or a future buyout. That is why the base exit is not set at Asseco's 24x or the stock's own 2023–2025 levels. ## What would prove us wrong | Observation | Where and when | What it would refute | | --- | --- | --- | | EBIT margin below 22% in H2 2026 or H1 2027 | FY2026 annual report (about April 2027); H1 2027 report (September 2027) | that the repricing is durable, which supports base cash growth | | September cash conversion: operating cash after working capital well below the provisional profit bridge (model 106.0; label-implied 100) | Q3 report, 10 November 2026 | the measured base under the 119.8 run rate. Scored 10 Nov: 97.4 on the frozen definition, entirely from current tax on customer advances; 106.7 timing-consistent | | Organic revenue growth below 8% in Q3 2026 (parent nine-month growth below 10%) | Q3 report, 10 November 2026 | the 14% run-rate roll (+8.2) and the 6.51% base. Scored 10 Nov: parent nine months +14.0%, Q3 alone +9.1%; consolidated organic split not disclosed | | Nine-month over-time revenue above 110, or Q4 2026 over-time revenue above 40 | Q3 report (10 Nov) and FY2026 report (about March 2027) | the KPO-wave reading; the wave would then return to the base. Nine months 101.7 (above the 80–95 band, not refuted); Q4 open | | Purchases from TSS Europe above about 17 in 2026, or a new related-party deal | FY2026 related-party note | minority fairness | | Acquisition cash below 20 in calendar 2027, or deals above 10x cash | FY2027 report | the deal pace and price behind about 2.7 points of base return | | A second related-party purchase from TSS without disclosed target economics | current reports | minority fairness, pointing to a lower terminal multiple | | NBP not renewing the ZSK contract beyond December 2027, or a lost Ministry of Family system | current reports in 2027 | franchise durability | | A special dividend or a regular payout | AGM resolutions, June 2027 | the zero-return cash drag; a favorable surprise | | Q3 subsidiaries' pretax increase below about 3.5 (KPO wave near the 3.85 tail), or above about 7 (near the bear's 7.4) | Q3 report segment and separate statements, 10 November 2026 | the judged 6.06 KPO wave (debate verdict 3). Scored 10 Nov: 9.65 (range 9.2–10.1); the wave is now the measured residual, 10.2 owner cash | | Polish 10-year yield below 5.3% or above 7.3% for a quarter | market data | the 10.45% judged cost of equity behind the 16x exit (debate verdict 2) | | Calendar 2027 deal cash above 45 (capacity growth well above 3.4%) | FY2027 report | the flat-deal-count base (debate verdict 5) | | A voluntary TSS tender or a delisting resolution | current reports | the minority-exit tail (debate verdict 7) | The Q3, margin, deal-pace, calendar-2026 and 2027 owner-cash, NBP and debate checks are recorded in `valuation/prospective-checks.json`. Permanent-loss paths and unverified inputs are updated in 06-risks.md. Sources: financial facts, named calculations and the method and assumption registers; sgn-fy25, sgn-h126, sgn-9m26-preliminary, sgn-mbank-2024, sgn-price-history, asseco-pe-aggregator, sgn-rem25, sgn-tss-call. No Harvia margin, ROIIC or growth schedule was imported. # Risks & disconfirming evidence Return effects are single-variable diagnostics on the recorded base (5-year base XIRR 12.3% on the revenue build), not probabilities. Figures marked "judged model" predate the revenue build. ## Permanent-loss paths 1. **Margin regression of the repriced franchise.** The EBIT margin doubled from 12% (FY2022) to about 26%, driven by pruning, less subcontracting and price, while pay per FTE rose 37% (2021→2025). If renewals reprice down at retender, or wages outrun prices, margins can return towards 2024's 22–23% or lower. The bear case (margin to about 21%, no deals, 12x) gives −3.6% over 5 years. 2. **Contract failure on a fixed-price build.** e-Taxes shows the mechanism: a tender won on price, subcontractor disputes, years of litigation and a settlement of about PLN14m in 2021. Issued guarantees were 13.1 at June 2026. A full call in year one lowers the base by about 0.1 points, but it does not capture the reputational and operational damage of a failure. 3. **Loss of a core relationship.** The NBP ZSK contract (up to about 27 a year net) runs to 31 December 2027. The Ministry of Family systems run on 48-month contracts from 2024. A retender loss, or insourcing to state IT centres, would remove high-margin recurring work. 4a. **KPO deadline cliff and growth quality.** The revenue build removes the wave as 18.5 of revenue at a 32.8% owner-cash margin. If the true wave revenue was larger, the removal is larger and the implied wave margin lower. About two-thirds of H1 2026 revenue growth came from acquired subsidiaries, all of it over-time project revenue that includes KPO hospital contracts ending 31 August 2026. The model removes the estimated wave (3.8 of owner cash). If the wave was twice as large, the 5-year base falls to 14.0%. If base growth stays at 5% rather than 6.51%, it falls to 13.5%. 4b. **Control mechanics (Polish Public Offering Act, in force 3 Nov 2026).** - **Squeeze-out:** possible only at 95% (art.82), which needs about 5.1m more shares, including Barca's 2.05m. Minorities can likewise demand a buyout (sell-out) only once a holder passes 95% (art.83). - **Delisting:** needs a resolution passed by 9/10 of votes cast, with half the capital present, after a tender for all remaining shares (art.91). Barca alone can block it by attending and voting against. - **Tender price floor:** the higher of the 3- and 6-month VWAP and the highest price paid in 12 months, or fair value set by an audit firm if trading is thin (art.79). No premium is required. - **What this means:** minority upside from a take-out is a negotiated premium, not a legal right. The downside is a tender timed after a drawdown at the floor, about PLN80–84 today. That is the mean daily close over 3 months (83.6) and 6 months (79.9), an approximation of art.79's average of daily VWAPs (sgn-price-history), against PLN94 now. The 2022 call reserved a future delisting. 4c. **Related-party charges.** Payments to TSS Europe were 11.1 in 2025 and 8.4 in H1 2026, and they are growing faster than profit. 4. **Controller and minority risk.** TSS holds 72.68%. The 2022 call reserved a future delisting "if an opportunity arises". A buyout below intrinsic value, repeated related-party purchases (g.on, EUR7.2m, no stand-alone financials) or a cash pile left idle would transfer value or depress the exit price. The float is about PLN392m, with median daily trading of PLN0.78m. Debate verdict 7 (unresolved): TSS is past the mandatory-tender threshold, so a tender would be voluntary and a squeeze-out needs Barca. A floor-price tender with a delisting threat after a drawdown (about PLN80–84 today, −6.6% a year if taken in year two) remains a coercive tail inside the downside weight. 5. **Deal economics.** The deal yield rests on two first-year results (Edrana, Sagra: 4.0x–8.7x net profit) and a family reference class (6.61x). On the expected yearly deal budget (4 Nov), the 5-year base is 11.7%, against 12.3% with a smooth 35.7 budget. On the revenue build, removing all future deals lowers the 5-year base from 12.3% to 10.1%, and a margin 2 points lower in every year (21%) gives 10.5%. Earlier figure: from 11.9% to 9.6% (judged model, 3 Nov 2026 debate). 6. **Valuation.** Since the debate the base and upside exit is 16.0x, a Koller value at a 10.45% judged cost of equity (05-thesis.md, 12-debate.md). At 14x the 5-year base is 9.6%, at 18x 14.0% and at 22x 17.9%. The Polish 10-year yield (6.31% on 3 Nov 2026) is the main external input: each point moves the multiple by about 3x. 7. **Regulation-timed demand.** Utilities revenue grew 34% from FY2022 to 2024 during the Solidarity Shield and CSIRE adaptations. Up to about 16 of revenue and 6–7 of EBIT may fade as those programmes end. ## Unverified inputs - **September 2026 cash conversion.** Resolved 10 November: reported pretax cash matched the profit bridge; only current tax on customer advances differed (97.4 strict, 106.7 timing-consistent). The open risk is now the advance build itself: 85.8 of contract liabilities at September (25.6 a year earlier). If the KPO-era advances unwind, cash falls without new revenue cash; the model charges 24.4 after tax in year one. - **Run-rate roll.** The +8.5 uplift to the anchor run rate assumes the 2025 organic growth of about 14% continued; without it the 5-year base is 10.5% on the judged model. The debate upheld the 14% (a past window filled with observed growth). The Q3 report should show the 2026 organic rate. - **Mednow and g.on earning power.** Mednow is inferred from one month of consolidation and its net price. g.on is a proxy at 6.61x its price, with no revenue disclosed. Removing both estimated additions lowers the 5-year base by about 0.9 points. - **Idle cash is partly customer money.** Contract liabilities rose 14.4 in H1 2026 to 55.8, probably including KPO advances that unwind as projects finish. Public receivables settle late in the year (Q4 2025 released about 17 of working capital). After the g.on payment, about 194 of cash at June less the 71 reserve leaves about 120 that is genuinely surplus. - **Cash return.** Retained cash earns the measured H1 2026 deposit yield, 1.56% after tax. A change in Polish rates moves it. - **Operating reserve.** 71.1 (one quarter of cash costs) is a judgment. Negative working capital means part of the cash is customer prepayments (contract liabilities of 55.8 at June 2026). - **Tax.** Current tax is charged. The IP-box effective rate (about 15%) is not assumed to persist beyond what current tax shows. - **Employee awards.** The economic cost is charged once and replacement buybacks are assumed to prevent dilution. The plan cap is 442,257 shares (1.9%). - **Sector margins and contract profitability.** Not disclosed. The unit economics are built from group averages. - **Customer and competitor evidence.** No fieldwork. Some customer pages (PGE) timed out, and no customer-budget figure was inserted into the model. ## Next review The 10 November Q3 report was read and rolled into the model. Next: the g.on price settlement and any IRGA-related change in TSS's 72.68% holding (current reports); Asseco's Q3 report (26 November) as peer context; the FY2026 annual report (date unpublished; FY2025 came on 24 March) for the Q4 over-time cliff, the unwinding of the 85.8 advance balance, the H2 margin, TSS purchases (9M 11.1) and the 2027 deal pace (prospective checks). The parent's Q3 slowdown to +9.1% is the leading growth question. Sources: sgn-q326, sgn-h126, sgn-9m26-preliminary, sgn-gon, sgn-facility, sgn-calendar, sgn-tss-call, sgn-fy21-report, sgn-fy24-report, sgn-nbp-contract, sgn-rem25, sgn-asseco-h126. # Thesis-to-model register No supplied author forecast/return exists. Exact source calibration is the cash-statement reconciliation and matching accounting perimeter, not an invented historical IRR target. Current scenarios are new dated underwriting. The pre-return plan and self-evaluation remain separate. | Driver | Observed anchor | Economic mechanism and chosen estimate | What changes it | Model field | | --- | --- | --- | --- | --- | | Earning power | FY2025 and H1 2026 consolidated CFO; sgn-fy25 printed p.8, sgn-h126 p.13; Q3 profit bridge sgn-q325 p.7, sgn-9m26-preliminary p.1 | Remove timing-related working capital, replace paid tax with current expense, charge share awards, capex/leases/financing, remove non-operating interest. PLN94.580m June trailing cash. The 2 November provisional September bridge gives PLN106.026m, and estimated target additions take the portfolio to PLN111.258m (27 October: PLN100.603m). | Actual cash conversion, tax, target results or recurring spending. | baseline.equityCashEarnings; underwritingBridge | | Organic growth | FY2025 revenue bridge and servicing/implementation tables; sgn-presentation slides 6-8, sgn-h126 p.18 | Installation/servicing demand and regulatory work support cash growth after ongoing costs. Base 6.51% (own 2020–2025 through-cycle rate), downside 0%, upside 11.0%; judgments below recent parent growth, not issuer guidance. Kept by the 3 Nov 2026 debate against the bull's 7.5% and the bear's 5.0%. | Customer retention, tender wins/losses, pricing, wages, maintenance mix and AI competition. | growthRate | | Employee economics | H1 p.13 CFO addback and pp.27-33 awards/buybacks | Charge economic award cost, keep all issued shares, assume replacements prevent additional dilution. No duplicate buyback expense or free retirement benefit. | Net issuance, award market value, buyback prices and scheme changes. | normalizedTrailingOwnerCash; shares; dilutionRate | | Operating cash | June cash and cost/contract notes; H1 pp.10, 19 | Hold three months of observed cash operating costs, PLN71.075m, outside deployment. The reserve remains a separately valued asset. | Delivery obligations, advances, payroll needs or restricted cash. | openingCash; retainedAssets | | Post-period purchase | October g.on release and NBP transaction-date FX | Estimate initial outflow at PLN31.496m. No acquired cash invented. Estimate target cash separately using the disclosed purchase-cash proxy. | Final consideration, actual settlement FX, target cash and operating results. | gonCashPrice; gonOwnerCash | | Missing Mednow cash | H1 p.23 updated PPA; annual pp.41-42; consolidation from 1 December 2025 | Use updated net capital and the acquisition proxy. The 2 November provisional bridge covers October 2025–September 2026, so only two months are missing: PLN0.527m (27 October June bridge: five months, PLN1.318m). No single December annualization. | Reported stand-alone full-year cash and updated allocation. | mednowMissingCash | | Purchase valuation | CSU-family 2020–2025 achieved aggregate (see calculations.json aggregate steps); DocLogix annual pp.41-42 as cross-check | 6.61× all-in owner cash (15.1% yield), a family reference class because Sygnity's own record is two to three deals. DocLogix implies 4.35×–9.04× (midpoint 6.70×). | Sygnity's own post-close deal cash, CEE price competition, the implementation mix. | acquisitionMultiple | | Deployment | FY2025 acquisition cash PLN35.736m (annual p.8); g.on about PLN31.5m (October 2026) | First-year purchases match the observed pace; annual purchases are capped at it, with capacity growth 0%/3.4%/15% by scenario (base: a flat deal count whose cash per deal grows with target earnings, since the 3 Nov 2026 debate; upside: CSU-family reference class). Unplaced cash is retained at face value; no dividend is declared. | Actual pipeline and execution capacity. Cash availability alone is insufficient. | reinvestmentRate; deploymentCapacity; deploymentGrowth | | Debt and claims | June balance/claims, H1 pp.10 and 28-33; facility amendment | Year-one acquisition/tax claims PLN6.507m; explicit retirement/litigation/bond claims. No extra debt draw; contingent guarantees stress-tested separately. | Settlements, financing use, guarantees called or changes in claims. | cashClaims; retainedClaims | | Exit value | Business-critical installed base versus project/control risk | Base/upside 16.0x normalized after-financing cash, the Koller value at a 10.45% judged market cost of equity (build-up from the PLN 10-year yield and Damodaran's ERP, averaged with the market-implied rate); downside 11.5x (1/r averaged with the 2022 trough). Cash/assets and claims counted once. Was 18x/12.2x before the 3 Nov 2026 debate. | Durable quality/runway, governance or financing change. | exitMultiple | | Distributions | June AGM equity statement | No declared common dividend; retain cash. Employee-share purchases do not become shareholder receipts. | Declared payout or changed allocation policy. | payoutRatio; distributionPolicy | Facts have stable IDs, periods, units and precise locators. Every model input has evidence and an explicit estimate classification where needed. Named calculations preserve the chain from observation to current model. Price changes affect entry cost, not this operating forecast or the terminal multiple. Dated subjective probability weights and revision triggers are recorded separately in research/11-scenario-probabilities.md. The ~15.35% resource control differs from the forecast period-weighted external reinvestment rate; see research/10-capital-efficiency.md. ## 2 November 2026 preliminary nine-month results Sygnity current report 29/2026 (2 November, p.1; Polish original) gives approximate 9M 2026 consolidated revenue PLN336.3m, EBITDA PLN102.2m, pretax profit PLN92.8m and net profit PLN75.9m. It contains no cash flow, balance sheet, tax note or acquisition note. The full report is confirmed for 10 November. | Input | 27 October (June bridge) | 2 November (provisional) | Basis | | --- | ---: | ---: | --- | | Trailing owner-cash base | 94.580 | 106.026 | June bridge plus Q3 2026 − Q3 2025 pretax profit (36.686 − 22.496) at the H1 2026 current-tax rate (21.98%), plus the D&A change, less the capex/lease changes. Q3 2026 costs are estimated as half of H1 | | Missing Mednow months | 1.318 (5 months) | 0.527 (2 months) | Window now ends September | | Portfolio owner cash | 100.603 | 111.258 | Plus unchanged g.on proxy 4.704 | | Base deployment fraction | ~16.12% | ~15.35% | First-year spend held at PLN35.736m; the denominator grew | Unchanged: opening cash, operating reserve and claims (still June balances less the g.on payment; unreported Q3 cash is not added), exit multiples 18×/14×, growth paths and probability weights. Mechanism and limits. Replacing a trailing quarter is not annualizing one quarter. Pretax profit already charges employee awards and accrued financing, so neither is deducted again. Unresolved: implementation-revenue timing within Q3, a small change in deposit interest inside pretax profit (H1 2026 cash interest was PLN2.25m), cash tax and working capital. Replace every estimate with the 10 November statements; if September CFO conversion differs, revise the base and record the effect. Return effect at the same PLN95.90 price and 2 November date (exits fixed): 5-year downside/base/upside 1.06%/12.94%/17.92% versus −0.79%/11.03%/15.96%; weighted-cash-flow XIRR 11.08% versus 9.18%. 10-year weighted 11.54% versus 10.63%. ## 3 November 2026: family deal rules applied (closing loop) The serial-acquirer rules adopted for CSU and Lumine (see `docs/learning/investor-learning-log.md`, 3 November) also apply here. Sygnity is controlled by Topicus, so the CSU-family record is a disclosed reference class, not a borrowed assumption about Polish IT services. | Input | Before | After | Basis | | --- | --- | --- | --- | | Purchase multiple | 6.70× (DocLogix midpoint) | 6.61× (family aggregate) | One small deal becomes a cross-check | | Deal capacity | none: the fraction of resources rose with cash, PLN35.7m → about PLN93m in year five | capped at PLN35.736m, growing 0%/10%/15% | FY2025 annual p.8; g.on shows a flat pace | | Unplaced cash | retained | retained (stated) | No declared dividend; cash historically accumulated | | Portfolio owner cash | PLN111.258m | PLN111.333m | Mednow and g.on proxies at the new multiple | Return effects at PLN95.90, 2 November 2026, exits fixed: 5-year downside/base/upside 1.06%/12.94%/17.92% → 1.07%/11.91%/14.64%; weighted 11.08% → 9.72%. 10-year weighted 11.54% → 9.61%. Probabilities are unchanged. Retaining unplaced cash at zero return is conservative: a future special dividend, as Topicus paid in 2025, would raise the return. The 10 November statements remain the next evidence for both earning power and deployment. ## 3 November 2026: pipeline evidence tested against the inputs (no change) Five years of filings were added (see `updates/2026-11-03-pipeline-completion.md`). Each finding was checked against the input it bears on: | Driver | New evidence | Effect on the estimate | | --- | --- | --- | | Purchase valuation (`acquisitionMultiple` 6.61x) | Edrana and Sagra 2024: 4.0x–7.1x a cash proxy, depending on amortization push-down (`deals2024LowerMultiple`, `deals2024UpperMultiple`) | inside Sygnity's own range; kept | | Deployment cap (35.7, +0/10/15%) | 34.3, 35.7, about 31.5: flat for three years | kept | | Organic growth (0/5/8%) | 2021–2025 NOPAT growth was about half margin repricing, a quarter volume and a quarter deals (research/04-capital-allocation.md) | 5% does not extrapolate the repricing; kept, with margin durability as a prospective check | | Exit (18x / 14x) | own year-end ex-cash P/E 13–24x; today 17.5x on the model denominator; Koller 17–22x at r of 9–10% and g of 5% | 18x kept; 14x downside is generous against an 11x zero-growth value, so the bear case is tested at 12x separately (05-thesis.md) | | Cash tax | loss carry-forwards exhausted (no tax losses at June 2026; H1 2026 paid 24.3) | confirms charging current tax | Return effect: none. At PLN94 on 3 November 2026, the 5-year weighted return is 10.17% and the 10-year 9.83%. ## 3 November 2026: author calibration against the CSU-family cash P/E table The owner's rule: deviating from a referenced investor needs a written, evidence-based reason. The table's forward columns (`references/csu-family-cash-pe-2026-11-02.json`) imply 109.6 of calendar-2026 and 134.7 of calendar-2027 owner cash; the previous model gave about 119 for 2027. | Input | Before | After | Named step | Evidence and verdict | | --- | --- | --- | --- | --- | | Starting owner cash (`baseline.equityCashEarnings`) | 111.3 (trailing sum plus proxies) | 119.8 (anchor run rate) | `runRateUplift`, `runRateOwnerCash` | The trailing window is centred 0.57 years before the anchor; 2025 organic growth was about 14%. Our trailing restart was a convenience. Adopt-modified. | | Purchase-year contribution of deals | none | half a year | `dealPartYear` | Deals close through the year (Feb, Mar, Jul, Dec, Oct). Adopt. | | Return on retained cash | 0% | 1.56% after tax | `retainedCashAfterTaxYield` | H1 2026 deposit interest 1.982 on average cash of 206; interest is already excluded from operating cash. Adopt. | | Acquired-business growth (base, upside) | 0% | 3.4% | `acquiredBusinessGrowth` | CSU-family 2021-2025 mean organic growth. Adopt. | | Downside exit | 14x | 12.2x | `downsideExitMultiple` | 14x sat above the 11.1x zero-growth value; midpoint with the 13.3x 2022 trough. Our own evidence. | | Base organic growth | 5% | 5% (kept) | `baseCashGrowth` | The label needs 7.4% only for a 1.6% residual, inside definitional tolerance. The margin has plateaued and the 2022-2025 repricing was one-time. Unresolved; prospective checks. | | Deal pace | 35.7, +0/10/15% | kept | — | Three flat years of observed pace. Reject the faster-deployment route. | Return effect at PLN94, 3 November 2026, exits fixed: | Horizon | Downside | Base | Upside | Weighted | | --- | --- | --- | --- | --- | | 5 years, before → after | 1.48% → 1.27% | 12.37% → 14.14% | 15.11% → 16.92% | 10.17% → 11.52% | | 10 years, before → after | 2.94% → 3.29% | 11.28% → 12.48% | 13.94% → 15.11% | 9.83% → 10.85% | Step by step for the five-year weighted rate: run rate +1.41, half-year deals +0.08, cash return +0.16, acquired growth +0.10, downside exit −0.39 points. Probabilities are unchanged (11-scenario-probabilities.md). ## 3 November 2026: growth-durability review (KPO cliff and organic growth) The review checked whether the 2025–2026 growth lasts (02-history.md, growth decomposition). | Input | Before | After | Named step | Evidence and verdict | | --- | --- | --- | --- | --- | | KPO hospital wave in starting earnings | included (119.8) | removed: −3.8 of owner cash, so the baseline is 115.7 | `kpoWaveOwnerCash`, `portfolioExKpo`, `runRateOwnerCash` | Mednow's KPO OPTIMED NXT contracts (22.8 gross, Q1 2026) must be completed by 15 Jul 2026, with spending eligible to 31 Aug 2026 (kpo-d112-schedule). Subsidiaries' over-time revenue rose +35.8 in H1 while their pretax rose only +3.9. Estimated as the subsidiaries' H1 profit increase, less DocLogix and a Mednow base, plus a Q3 tail of 50%; range about 3–8. | | Run-rate roll | applied to 106.0 | applied to 102.2 (excluding the wave) | `organicTrailingBase`, `runRateUplift` | The wave does not recur, so it is not rolled forward. | | Base organic growth | 5% | 6.51% | `baseCashGrowth`, `organicRatio2020to2025` | Sygnity's own 2020–2025 through-cycle organic revenue CAGR (299.7 / 218.7, fifth root) at a constant margin. The parent's organic growth was 13.4% in 2025 and 16.5–16.8% in each first half since 2024, with maintenance +17–23%. Outside view: CSU-family organic 3.4%. | | Upside organic growth | 8% | 11.0% | `upsideCashGrowth`, `parentOrganicGrowthRecent` | Midpoint of the through-cycle rate and the parent's recent 15.6%. | Return effect at PLN94, 3 November 2026, exits fixed: | Horizon | Downside | Base | Upside | Weighted | | --- | --- | --- | --- | --- | | 5 years, before → after | 1.27% → 0.62% | 14.14% → 14.73% | 16.92% → 18.81% | 11.52% → 12.18% | | 10 years, before → after | 3.29% → 2.95% | 12.48% → 13.24% | 15.11% → 17.03% | 10.85% → 11.83% | Step by step for the five-year weighted rate: KPO cliff −0.67, base growth +0.70, upside growth +0.63 points. Probabilities are unchanged. ## 3 November 2026: bull–bear debate judgment Fact check: the Polish 10-year yield was 6.31% on 3 Nov 2026, not the 6.47% both advocates used. | Input | Before | After | Named step | Evidence and verdict | | --- | --- | --- | --- | --- | | KPO wave in starting earnings | 3.85 (H1 plus a 50% Q3 tail) | 6.06; baseline 115.66 → 113.28 | `kpoWavePretaxQ3Residual`, `kpoWavePretaxQ3Judged`, `kpoWavePretaxTrailing` | The Q3 subsidiary profit residual (7.0 pretax) contradicts a half-size tail. The residual is an upper estimate, and the known contracts bound the total. Equal weight on both. Split | | Base and upside exit | 18x | 16.0x | `buildUpCostOfEquity`, `marketImpliedCostOfEquity`, `judgedCostOfEquity`, `baseExitMultiple` | Build-up r 10.92% (6.31% − 0.72% + 5.33%) and market-implied r 9.98% are averaged to 10.45%, giving Koller (1 − 0.13)/(r − 5%). 18x needed r ≈ 9.8%. Split, leaning bear | | Downside exit | 12.2x (1/9% hurdle with the 2022 trough) | 11.5x | `zeroGrowthMultiple`, `downsideExitMultiple` | The same buyer rate applies to every exit. Consistency | | Base deal-capacity growth | 10% (family) | 3.4% | binding to `acquiredBusinessGrowth` | Three flat years; a flat deal count whose cash per deal grows with target earnings. Split | | Base organic growth | 6.51% | 6.51% (kept) | `baseCashGrowth` | The bull's 7.5% fade averages 5.7% over ten years, and the bear's EU cap exempts defence and covers a third of revenue. Outside view kept | | Run-rate roll | 14% | 14% (kept) | `runRateUplift` | It fills a past window with observed growth; the referenced investor's column needs it. Bull | | Weights, cash return | 30/50/20, 35/40/25; 1.56% | kept | — | Several attacks survive; the cash yield is measured | Return effect at PLN94, 3 November 2026, exits fixed: | Horizon | Downside | Base | Upside | Weighted | | --- | --- | --- | --- | --- | | 5 years, before → after | 0.62% → −0.59% | 14.73% → 11.86% | 18.81% → 16.06% | 12.18% → 9.69% | | 10 years, before → after | 2.95% → 2.43% | 13.24% → 11.53% | 17.03% → 15.75% | 11.83% → 10.50% | Step by step for the five-year weighted rate: KPO wave −0.40, deal capacity −0.16, base and upside exit −1.76, downside exit −0.17 points. The weights were reassessed and kept (11-scenario-probabilities.md). Prospective checks were added for the Q3 subsidiary profit, the PLN yield, calendar-2027 deal cash, a June 2027 payout and a TSS tender. ## 3 November 2026: revenue build (yearly paths) The owner's instruction: an IRR must come from financial projections, with each assumption stated year by year. The model now uses the shared revenue build (`revenueBuild`, `inputs.yearly`). | Input | Before (scalar) | After (yearly path) | Named step or source | | --- | --- | --- | --- | | Starting revenue | none | 492.2 = (443.5 trailing − 18.5 KPO) × 1.080 + g.on 29.8 + Mednow 3.3 | `startingRevenue`, `kpoWaveRevenue`, `gonRevenueEstimate` | | Starting margin | implicit | 23.0% = 113.3 / 492.2; KPO wave 18.5 of revenue at 32.8% | `startingMargin`, `kpoWaveMargin` | | Organic growth, base | 6.51% constant | 9.5% (yr 1) · 7% (yrs 2–3) · 5.5% (yrs 4–10) | `baseYear1OrganicGrowth`; pl-mof-mtfsp; sgn-h126 | | Organic growth, upside / downside | 11.0% / 0% | 13 / 11 / 9% · −2% then 0% | path reasons | | Organic margin | implicit, constant | base 23.0%; upside 23.5% from yr 3; downside 21.5% from yr 2 | `startingMargin`; sgn-fy24 | | Deal price | 6.61x cash | 1.06x revenue at a 16.2% acquired margin (6.6x cash) | `ownDealPriceToSales`, `acquiredMarginOwnDeals` | | Acquired growth | 3.4% | 0% (yrs 1–2), then 3.4% | sgn-h126; `acquiredBusinessGrowth` | Calibration: with constant paths equal to the scalar inputs, the returns matched the debate model exactly (5-year −0.59/11.86/16.06%, 9.69% weighted). Return effect at PLN94, 3 Nov 2026, exits fixed: | Horizon | Downside | Base | Upside | Weighted | | --- | --- | --- | --- | --- | | 5 years, before → after | −0.59% → −2.08% | 11.86% → 12.28% | 16.06% → 16.18% | 9.69% → 9.65% | | 10 years, before → after | 2.43% → 1.63% | 11.53% → 11.40% | 15.75% → 15.11% | 10.50% → 10.07% | Five-year weighted attribution: organic growth paths +0.06, margin paths −0.12, acquired margin and price +0.02, acquired growth 0.00 points. The ten-year rate falls 0.43 points, mainly from the fading growth (−0.42). ## 4 November 2026: yearly deal budget Deal cash is now stated per year (`deploymentCapacity.basis: 'yearly-path'`). `deploymentGrowth` is dropped. | Input | Before | After | Named step or evidence | | --- | --- | --- | --- | | Base deal cash | 35.7 growing 3.4% (35.7 → 48.3) | expected budget 26.3 growing 3.4% (26.3 → 35.5) | `expectedDealBudget` = `activeEraMean` 35.1 × `activeYearFrequency` 0.75; no dated lumps without evidenced timing | | Downside deal cash | none (rate 0) | 0 for three years, then 14.0 a year (rate 1) | `quietRunYears` 3; `recordMean2021to2025` 14.0 | | Upside deal cash | 35.7 growing 15% | 35.7, 44.2, 44.2, 58.4, 58.4, then 95.0 | stepped 15% | | Record measures | — | mean 14.0 (2021–25), 26.3 (TSS era), 35.1 (active era); best year 2.55x the mean; quiet run 3 years | `tssEraMean`, `activeEraMean`, `recordBestToMean` | A calibration with year-by-year values equal to the smooth cap reproduced the returns exactly. | PLN94, 3 Nov 2026 | Downside | Base | Upside | Weighted | | --- | --- | --- | --- | --- | | 5 years, before → after | −2.08% → −1.83% | 12.28% → 11.71% | 16.18% → 16.19% | 9.65% → 9.39% | | 10 years, before → after | 1.63% → 2.12% | 11.40% → 10.90% | 15.11% → 15.13% | 10.07% → 9.94% | The base first used dated quiet years (years 3 and 8: 11.86%, 9.47%). It was corrected the same day to the expected budget, because the timing of quiet years is not evidenced. ## 10 November 2026: roll to the full Q3 2026 report The nine-month report (sgn-q326, Polish original, published 10 November) replaces the 2 November provisional bridge. | Input | Before | After | Named step or source | | --- | --- | --- | --- | | Trailing window | June 2026 (FY2025 + H1 2026 − H1 2025) plus a preliminary-profit Q3 increment | reported September 2026 (FY2025 + 9M 2026 − 9M 2025) | the `…Ttm` steps; sgn-q326 p.11 | | Normalized owner cash, current-tax definition | 94.58 (June) + 11.45 provisional Q3 = 106.03 | 97.42 reported | `normalizedTrailingOwnerCash` | | Current tax on excluded working-capital timing | not separated | +9.27 (deferred-tax asset on deferred income +5.07; contract-revenue liability −4.20, Sep 2025 to Sep 2026) | `advanceTaxTiming`; sgn-q326 p.20, sgn-q325 note 12.2 | | September earning power | 106.03 provisional | 106.69 | `preliminaryUpdatedOwnerCash` | | Q3 KPO wave | judged equal-weight mean, 4.31 pretax (trailing wave 6.06 owner cash) | measured residual, 9.46 pretax (10.23 owner cash) | `kpoWavePretaxTrailing` now uses `kpoWavePretaxQ3Residual`; parent Q3 reported (+4.53 ex estimated dividends, against an assumed +6.96) | | KPO wave revenue | 18.5 (Q1 contracts 22.757 gross) | 65.5 (H1 contracts 80.544 gross, same nine hospitals) | `kpoWaveRevenue`; contradiction between the Q1 and Q3 reports recorded | | Anchor run-rate owner cash | 113.28 | 109.49 | `runRateOwnerCash` | | Starting revenue / margin | 492.2 / 23.0% | 441.4 / 24.8% | `startingRevenue`, `startingMargin` | | Downside margin from year 2 / upside from year 3 | fixed 21.5% / 23.5% | starting margin −1.5 / +0.5 points (23.3% / 25.3%) | `downsideLaterMargin`, `upsideLaterMargin` | | Opening cash | June 225.1 − g.on 31.5 − reserve 71.1 = 122.5 | September 281.1 − g.on 31.5 − reserve 73.9 = 175.7 | `investableCash`; g.on closed 2 Oct (note 35) | | Year-one claims | 6.51 (June payables) | 9.22 payables + 24.45 advance-build reversal = 33.67 | `knownCashClaims`, `advanceReversalClaim` | | Retained claims and assets | June values | September: retirement 1.563, litigation 0.538, bond 0.100, JV 0.983, receivables 0.451 | sgn-q326 pp.9, 27–28 | Why each change: - **Tax timing.** On the frozen definition the Q3 bridge missed the provisional estimate by 8.6. Pretax cash before working capital rose 15.4, in line with the 14.2 profit increase. All of the miss was current tax: 13.98 in Q3 against 1.80 a year earlier, while deferred tax credited 7.0. Note 12.2 shows tax charged on customer advances. The bridge excludes the 48.4 working-capital inflow those advances created, so keeping the tax on them was inconsistent. The method requires separating cash tax from deferred-tax timing (docs/method/cash-flow-and-reinvestment.md). - **Advance reversal.** September cash includes the 2026 advance build: contract liabilities were 25.6 in September 2025, 41.3 in December, 55.8 in June and 85.8 in September, during the KPO deadline wave. Deploying that cash as if it were permanent would count it twice once the advances unwind. The after-tax nine-month working-capital inflow (30.18 × 0.81) is a year-one claim. If the advances turn out to be recurring maintenance prepayments, the claim should be released. - **KPO wave.** The frozen debate rule moved the Q3 wave to the measured residual once the reported subsidiary increase (9.65) exceeded 6.5. The parent's Q3 dividends are not split; half of the parent's quarterly other operating income is treated as dividends (`parentQ3DividendShare`). Bounds of 0 and 1 move the five-year weighted figure by +0.06 and −0.07 points. - **Relative margins.** The downside and upside margins were absolute levels set against a 23.0% start. Removing the lower-margin KPO revenue raises the starting margin, and fixed levels would then have added 3.3 points of downside compression and turned the upside gain into a loss, purely mechanically. The stated reasons describe a 1.5-point compression and a half-point mix gain, so they are now encoded relative to the start. With the old start the encoding gives the old values (23.02% − 1.5 = 21.52%). Return effect at PLN97, 10 November 2026, exits fixed (old model and new model at the same quote and date): | Horizon | Downside | Base | Upside | Weighted | | --- | --- | --- | --- | --- | | 5 years, before → after | −2.45% → −2.68% | 11.05% → 10.30% | 15.52% → 14.69% | 8.74% → 8.08% | | 10 years, before → after | 1.80% → 1.66% | 10.57% → 10.20% | 14.80% → 14.39% | 9.62% → 9.27% | One-at-a-time reversals of the five-year weighted figure: measured KPO wave −0.59, advance-reversal claim −0.17 and the tax-timing add-back +1.22 points (against the strict definition). The KPO revenue estimate is neutral once margins are relative (+0.01). Weights are unchanged (11-scenario-probabilities.md, 10 November). # Underwriting plan - 27 October 2026 This plan records the business logic and intended assumptions before calculating shareholder IRRs. The work is new underwriting from issuer evidence, not a Harvia-style source-return reproduction. ## Opportunity ledger Added 3 November 2026, after the original 27 October plan below and after returns had been calculated. It tests the existing scenario inputs and changes none of them. Money in PLN millions; all figures are judgments built from the cited facts. | Opportunity | Starting point | Customer need and why Sygnity wins | Reachable scale in 5 years | Investment and cash cost | Timing and limits | Evidence against | | --- | --- | --- | --- | --- | --- | --- | | Price on installed systems (indexation and repricing at renewal) | servicing 144.2 in 2025; NBP up to about 27 a year; six Ministry of Family 48-month contracts from 2024 | continuity of statutory and central-bank systems; incumbent code and staff | inflation-linked: roughly 3–5% a year on the recurring base, about 25–40 of revenue | none beyond staff already in costs | at each renewal (NBP end-2027; ministries 2028) | public buyers retender on price; the 2022–2025 repricing may already have taken the slack | | Volume: regulatory and functional change requests on own systems | implementation 189.0, much of it change work on installed systems | legal deadlines (energy market, labour-market reform, NBP needs) leave customers no time to switch | follows legislation; we assume 0–3% a year net of fading programmes | delivery staff at about 0.19 per FTE a year; roughly 6 people per 2.9 of revenue | lumpy; Solidarity Shield and CSIRE work is fading | utilities growth of +34% since FY2022 is partly temporary | | Share: new tenders and new public systems | public-sector projects 83 in 2024 | domain references; price decides at tender | not underwritten as growth; replaces lost or expiring work | bid costs, guarantees (13.1 outstanding), fixed-price risk | 2026–2029 EU and KPO cycle | e-Taxes failure; Asseco's scale | | Acquired niches: organic growth | about 51 of revenue (Sagra, Edrana, DocLogix, Mednow, g.on) | vertical software with maintenance | at 5–8% a year, +15–25 of revenue | small; working capital in SaaS is low | already in the base | no post-acquisition results disclosed | | Acquisitions | 34.3 (2024), 35.7 (2025), about 31.5 so far in 2026 | sellers (founders, Comarch, TSS) get a permanent CSU-family owner | at about 35–52 a year (base cap growing 10%), about 220 deployed over 5 years at 6.61x, adding about 33 of owner cash | cash only; no debt or equity | each deal contributes from the year after purchase; integration capacity of one or two deals a year | Sygnity's own record is two measurable deals; related-party supply (g.on) | | KPO hospital wave (Mednow OPTIMED NXT) | nine KPO hospital contracts of 22.8 gross (Q1 2026); subsidiary over-time revenue +35.8 in H1 2026 | hospitals had to spend recovery funds by the deadline | none after 2026: projects completed by 15 Jul 2026, spending eligible to 31 Aug 2026 | delivery staff and licences | ends Q3 2026; afterwards maintenance on the installed hospitals | removed from starting earnings (about 3.8 of owner cash); a Q4 2026 over-time level near Q2's would refute the wave reading | | Idle cash | 225.1 at June 2026; about 75 a year beyond deal spend | — | 600 or more by year five if not used | none | policy choice for TSS and the AGM | no dividend since at least 2021 | Scale check five years out (base): existing-business owner cash grows 5% to about 142. At constant conversion that is roughly 570 of existing-business revenue against 443.5 trailing September 2026 (provisional), plus about 150 of revenue from new deals at the 2024 deals' 1.15x price-to-sales, with their own staff. At today's 0.41 revenue per FTE plus 3% a year indexation, the existing business would need about 1,200 staff instead of roughly 860 now, about 7% a year of net hiring. That is feasible in Polish IT but adds to wage pressure. Management targets are not published, so there is no guidance to test against. The base sits below the company's recent revenue growth (+19.5% in 2025; +38% in the first nine months of 2026, provisional), which included repricing and acquisitions. ## Scenario mechanisms Each case is a coherent business outcome. Growth, margin, required capital, deal capacity, financing, dilution and the exit denominator move together and are recorded in `valuation/model.json` inputNotes. | | Downside | Base | Upside | | --- | --- | --- | --- | | Organic revenue growth (yearly path) | year 1 −2%, then 0% | year 1 9.5%, years 2–3 7%, years 4–10 5.5% (10-year mean 6.2%) | year 1 13%, years 2–3 11%, years 4–10 9% | | Organic owner-cash margin (yearly path) | 23.0% in year 1, then 21.5% (back to the 2024 level as wages outrun indexation and NBP and the ministries reprice) | 23.0% held (group EBIT 25.5% → 25.2%; repricing done at group level) | 23.0% in years 1–2, then 23.5% (mix shifts to the higher-margin parent) | | Required capital | none (negative working capital) | none beyond expensed staff | none | | Deal capacity and price | yearly budget: 0 for three years (the worst dry spell in the record), then 14.0 (2021–2025 mean) | yearly budget: 26.3 in year 1, growing 3.4% a year to 35.5 in year 10 (expected budget including quiet-year frequency: active-era pace 35.1 × 0.75); no lump or quiet year is placed in a specific year because no deal timing is evidenced; 1.06x price per zloty of revenue at a 16.2% margin; half a year in the purchase year; acquired revenue flat in years 1–2, then +3.4% | yearly budget: 35.7, 44.2, 44.2, 58.4, 58.4, then 95.0 (15% capacity growth); same price and margin | | Return on new capital | — | 15.1% entry cash yield | 15.1% | | Financing and dilution | no debt; awards bought back; zero net dilution | same | same | | Unplaced cash | retained, earning the measured 1.56% after-tax deposit yield, counted at face value | same | same | | Starting revenue and margin | revenue 441.4 (reported trailing 443.5, less the 65.5 KPO wave, rolled 0.57 years at 14%, plus g.on 29.8 and Mednow 3.3 estimates); owner cash 109.5; margin 24.8% (10 Nov 2026; was 492.2 / 113.3 / 23.0%) | same | same | | Exit denominator and multiple | 11.45x normalized after-financing owner cash (judged: 1/r at 10.45% averaged with the 13.3x 2022 trough) | 15.97x (Koller at r = 10.45%) | 15.97x: higher growth does not earn a higher multiple | Temporary under-use versus permanent impairment: idle cash is temporary under-use, valued at face and not destroyed. A permanent impairment would be a lost core system or a structural margin reset; those sit in the bear case (05-thesis.md) and the risks register, not in the downside path. Deals are funded from cash before their earnings are recognized. The capacity cap reflects one or two integrations a year, not the cash available. Outside view (growth-durability review, 3 November). Five-year organic owner-cash growth is 0%, 6.5% and 11.0%. Including deals, base owner cash grows from 115.7 to about 197 in year five, about 11% a year. 6.5% organic is in line with Sygnity's own 2020–2025 through-cycle record. It is twice the CSU-family organic mean (3.4%), which the parent's maintenance growth supports. The upside's 11% for a decade is above what most mature IT-services companies sustain and depends on the parent's recurring growth persisting; it carries a 20–25% weight. The KPO wave is a funding deadline, not demand that recurs, so it is removed rather than extrapolated. Diagnostic at PLN94 on 3 November, base with no deals: 2% growth at 18x gives 7.9%, 3% gives 8.8% and 4% gives 9.7%. The market price implies about 3.3% organic cash growth at an 18x exit for a 9% return. Author calibration (3 November): the CSU-family cash P/E table implies 134.7 of calendar-2027 owner cash. The base now gives 132.6 (−1.6%), against 118.7 before. ## Business and definition Regulated banking, public administration and utility customers buy implementations and ongoing servicing of business-critical software. Servicing is a repeat-business anchor, but projects and public procurement remain substantial. Sygnity is not a pure subscription vendor. Product development, implementation staff, subcontractors and incentive compensation remain economic costs. Use the shared after-financing cash-equity adapter. Normalize trailing consolidated operating cash flow for working-capital and tax-payment timing; deduct sustaining cash investment, cash leases, financing paid outside operating cash and the employee-share addback. Remove income on cash/financial assets valued separately. Do not deduct acquisition amortization again or charge an arbitrary industrial organic ROIIC on top of software costs already expensed. Use FY2025 + H1 2026 - H1 2025. Keep current taxes, investment cash, employee awards and deposit income separately identifiable. Preserve the actual published CFO bridge before normalization. Employee-share buybacks fund awards; charging the economic award cost means no second future award-buyback charge or free share-count accretion. ## Current portfolio and capital Start from June cash. Subtract the initial EUR7.2m g.on price using the transaction-date NBP EUR/PLN reference rate as an estimated cash payment. Retain an operating reserve equal to one quarter of observed cash operating costs because contract delivery and advance receipts create funding obligations. Reserve cash remains an asset at face value, outside the acquisition budget. No interest on it is capitalized into operating earnings. Treat the June deferred acquisition payable as a year-one cash claim. Carry reported pension and litigation provisions, the small bond balance and the JV/other financial assets explicitly. Cash leases are charged in earnings; lease principal is not deducted again as a separate terminal debt. No new credit draw is assumed. The PLN80m facility includes a PLN30m guarantee sub-limit and a PLN50m overdraft; a facility limit is neither cash nor a shareholder asset. The October g.on release discloses a price but not stand-alone owner cash. Mednow is in only seven months of the trailing window. Estimate missing mature cash using a disclosed acquisition-cash proxy, not a one-month December profit annualization. Mark these additions as estimates, not reported earning power or guaranteed synergies. Hold the estimates constant across scenarios until better target evidence is available. ## Acquisition economics DocLogix provides a small, imperfect empirical anchor: a six-month post-purchase net profit contribution, the purchase price, acquired cash, transaction costs and the purchase-price intangible asset balance. Annualize the six-month contribution only as a rough proxy, with explicit seasonality limitations. The lower cash-multiple estimate reverses after-tax acquisition amortization using an estimated eight-year useful life within the disclosed 5-11 year range. The upper estimate gives no cash credit for that amortization. Use the midpoint of these alternative cash definitions for a base purchase multiple. This is not an observed portfolio hurdle or a proof that future deals will earn it. Set the base deployment fraction so the first modeled year's purchase budget matches FY2025 actual acquisition cash spending, given the current investable cash plus estimated first-year owner cash, after known cash claims. It is a calibration to observed deployment pace, not to an IRR. Keep that fraction fixed thereafter and inspect the resulting annual deal sizes. Faster deployment requires more actual deal capacity, not just available financing. New purchases occur at year-end and earn cash the following year; no early mature earnings is recognized. ## Coherent scenarios | Case | Existing-business cash growth | New purchases | Exit valuation | Economic interpretation | | --- | --- | --- | --- | --- | | Downside | 0% | None | 14x after-financing owner cash | Project demand and the acquisition platform stall; cash accumulates, with no unsupported extra permanent margin haircut. | | Base | 5% | Fraction calibrated to FY2025 cash deployment; midpoint purchase-cash proxy | 18x | Maintenance, regulatory work and implementations grow more slowly than recent reported results; acquisition discipline and deployment stay near observed capacity. | | Upside | 8% | Twice the base initial deployment fraction, same purchase-cash multiple | 18x | Stronger product/servicing demand and a broader deal pipeline; no extra terminal re-rating or lower purchase price. | Growth, deployment and terminal valuation are explicit estimates. Recent FY2025 revenue excluding newly acquired companies grew roughly 14%, but that is not a perpetual cash-growth forecast. A 5% base requires continuing product investment already charged to cash earnings. The 18x multiple is a valuation judgment reflecting installed mission-critical software, balanced against project exposure, public-sector procurement and controlling-shareholder risk. It is not copied from Harvia, Lumine or CSU, and not solved from a target return. The 14x downside reflects weaker durability and minority-marketability concerns. Probabilities are not assigned. ## Tests before acceptance Check the published cash balance identity, consolidated versus parent-only accounts, all ownership claims and cash-tax timing. Change one financial observation and confirm the calculation trail changes. Change only the live quote and confirm the operating forecast does not change. Test zero deployment, slower cash growth, an acquisition-price increase, a delay in acquisition contribution, a lower exit multiple, and a separate guarantee-loss stress. No scenario probability or author endorsement is implied by these tests. Sources: issuer FY2025 consolidated statements; H1 2026 consolidated statements; April 2026 investor presentation; October acquisition release; September facility notice; October reporting-calendar notice; NBP transaction-date FX; competitor Asseco H1 2026 release. Customer contract notices corroborate demand mechanisms, not forecast revenue recognition. ## Yearly paths (revenue build, 3 November 2026) Owner cash is now revenue × margin. Each assumption is stated per year range with its reason and sources in `valuation/model.json` (`cases.*.inputs.yearly`); the starting point is in named steps `startingRevenue` and `startingMargin`. **Starting point.** Trailing October 2025 to September 2026 revenue was 443.5 (FY2025 audited plus the reported 9M 2026, less 9M 2025; the 10 November report confirmed the preliminary figure). The KPO hospital wave is removed as revenue and margin, not only as cash: 65.5 net of VAT (the nine Mednow hospital contracts, 80.5 gross per the Q3 report; the Q1 report had said 22.8 for the same hospitals), carrying 10.2 of owner cash, a 15.6% margin (`kpoWaveRevenue`, `kpoWaveMargin`). The rest is rolled 0.57 years at 14% to the anchor. Estimates are then added for g.on (29.8) and Mednow's two pre-consolidation months (3.3), both at Sygnity's own 1.06x price-to-sales. That gives starting revenue of 441.4 against owner cash of 109.5, a 24.8% margin. Before the 10 November roll: 18.5 KPO revenue at 6.06 owner cash, 492.2 revenue, 113.3 owner cash, 23.0%. | Path | Base (years: value) | Why | | --- | --- | --- | | Organic growth | 1: 9.5% · 2–3: 7% · 4–10: 5.5% | Year 1 weights the parent (69% of ex-KPO revenue, organic 13.4–16.8% since 2024, maintenance +17–23%, ministry renewals) at 13% against subsidiaries at 2% (point-in-time revenue flat in H1 2026). Years 2–3 fade under the EU deficit procedure's net-expenditure path (4.0% in 2027, 3.5% in 2028; defence exempt) and the NBP renewal due in December 2027. Years 4–10 are long-run growth just below the 6.51% through-cycle rate. The 10-year mean is 6.2%. | | Organic margin | 1–10: starting margin, 24.8% since 10 Nov (was 23.0%) | Group EBIT margin was flat at 25.5% (2025) and 25.2% (H1 2026). The parent is still rising (core 26.1% → 32.6%), but low-margin acquired units offset it. Wage growth (+5.3% per FTE) is passed into renewals. | | Price per zloty of acquired revenue | 1–10: 1.06x | Own deals net of cash: Edrana 0.63x, Sagra 1.43x, DocLogix 0.69x. At the acquired margin this is 6.6x cash, the CSU-family aggregate. | | Acquired margin | 1–10: 16.2% | Edrana 15.8% and Sagra 16.4% first-year net margins; DocLogix 7.9%; subsidiaries about 14% after tax in H1 2026, against the parent's 32.6% core margin. | | Deal budget | 1: 26.3, growing 3.4% a year to 35.5 in year 10 | Expected budget including quiet-year frequency. Deal cash was 0, 0, 0, 34.3 and 35.7 (2021–25), plus 35.1 so far in 2026. The active-era pace (35.1) is multiplied by the TSS-era active-year frequency (0.75) and grown at the 3.4% verdict. No deal has been announced, signed or financed, so no lump is put in a specific year. g.on was paid before the anchor. | | Acquired growth | 1–2: 0% · 3–10: 3.4% | Subsidiaries' point-in-time revenue was flat in H1 2026; after that the CSU-family organic mean. | Base rows, years 1–5 (revenue / organic / acquired / margin / owner cash): 555.8 / 538.9 / 16.9 / 22.8% / 128.7; 627.9 / 576.6 / 51.3 / 22.5% / 144.3; 704.9 / 617.0 / 87.9 / 22.2% / 161.2; 778.0 / 650.9 / 127.1 / 21.9% / 177.2; 855.5 / 686.7 / 168.7 / 21.7% / 194.3. The blended margin falls because acquired revenue earns 16.2%. Calibration: constant paths equal to the previous scalar inputs (the margin at 23.0%, acquired margin at 23.0%, price-to-sales at 6.61 × 23.0%) reproduced the debate model exactly: five-year −0.59/11.86/16.06%, 9.69% weighted. # Method self-evaluation - 27 October 2026 The economic plan was saved before the first return calculation. The analysis uses Sygnity's own consolidated financials, delivery/servicing mix and incentive/funding structure. ## Source and accounting checks The FY2025 and H1 cash statements reconcile from opening to closing cash, including investing, financing and translation. The model uses consolidated cash, rather than the parent-only cash statement containing subsidiary dividends. The original Polish table confirms the Mednow purchase allocation is in PLN thousand despite English prose saying EUR. Statutory revenue tables take precedence over presentation sub-bars that do not sum to their headline. The latest Mednow allocation is used, with the original provisional value left in the source record. Acquisition-year timing is respected: DocLogix is effectively fully consolidated in the current trailing window, Mednow has seven months, and g.on is absent. Known-target missing cash is estimated separately rather than treating a closing-month profit as a full-year run-rate. The owner-cash bridge retains employee incentive cost, actual capital purchases, leases and tax. Changing the compensation observation reduces cash through the calculation trail without reducing the share count for free. Operating reserves, legacy taxes/payables, financial assets, claims and contingent guarantees have separate treatment. Facility limits are not assumed cash. ## Economic tests at one fixed price/date All results below use PLN90.90, 27 October 2026, and a 27 October 2031 exit. They are single-variable diagnostics around the recorded base. They are not probabilities, confidence limits or source-author forecasts. | Counterfactual | Estimated five-year XIRR | What it tests | | --- | ---: | --- | | Recorded base | 12.18% | Current-portfolio cash grows 5%; purchases start near observed cash spending. | | No future acquisitions | 8.29% | Cash alone does not guarantee an opportunity set. | | No organic cash growth | 8.05% | The base depends on existing products and delivery economics too. | | New purchases cost 50% more per cash euro | 10.24% | Available money is not the same as attractive purchase returns. | | New purchases contribute one additional year later | 10.98% | No free mature cash before the investment contributes. | | Exit multiple 14x rather than 18x | 7.56% | Terminal judgment is material. | | Full PLN13.078m guarantee cash call in year one | 12.00% | Contingent cash risk, without pretending it measures consequential business damage. | | Remove both estimated current-target additions | 10.94% | Unreported target earning power is not silently treated as known. | Changing only the share price preserves forecast rows. Missing compensation evidence, source references and model inputs are validated. The company uses the existing shared cash-equity and XIRR engines; no company-specific executable valuation code was added. ## Limits and next evidence The purchase-cash proxy rests on a six-month DocLogix result and an estimated amortization life. Tax timing, current-target profitability, future project cash needs and deployment capacity can all differ. The forecast gives actual first-/last-year purchase budgets so that sourcing requirements are visible: base approximately PLN35.7m to PLN88.9m, upside PLN72.4m to PLN156.4m. A finite history does not prove that expansion can continue. The downside is a coherent stalled-growth/deployment case, not a bound on loss. A permanent software impairment, dilution, unfair related-party deal or collapsing project margin can be worse and needs separate evidence. Base and upside retain the same purchase and terminal multiples, preventing an unsupported extra re-rating from manufacturing the favorable result. Issuer releases, the financial statements, business/product pages and a competitor release were read. Some PGE/customer pages timed out, and no comprehensive current customer-budget survey or g.on stand-alone filing review is claimed. Q3 results on 10 November are the next confirmed event to replace current proxies. # Capital productivity and reinvestment Money is in PLN millions. These calculations use the researched scenarios and the shared forecast ledger. They are forward diagnostics, not independently measured historical ROIIC or realized acquisition IRRs. Local sensitivities may differ. | Period | Scenario | Capital return / proxy | Period-weighted reinvestment | New growth capital | Matching earnings | | --- | --- | ---: | ---: | ---: | ---: | | 5 years | downside | 15.3% | 5.2% | 28.031 | 542.912 | | 5 years | base | 15.3% | 18.4% | 140.724 | 765.707 | | 5 years | upside | 15.3% | 27.2% | 240.936 | 884.568 | | 10 years | downside | 15.3% | 8.4% | 98.109 | 1162.796 | | 10 years | base | 15.3% | 15.8% | 307.055 | 1940.120 | | 10 years | upside | 15.3% | 27.8% | 715.936 | 2578.506 | Updated 10 Nov 2026 after the Q3 roll (anchor run rate 109.5 after the measured 10.2 KPO wave). Deal budgets are unchanged, so new growth capital is unchanged; matching earnings are about 3% lower. Deal cash is a stated yearly budget (4 Nov 2026). The base uses the expected budget (26.3 growing 3.4%), which prices quiet years in without dating them; the downside is dry for three years, then deploys 14.0 a year. Revenue build (3 Nov 2026): the deal return proxy is now the acquired margin (16.2%) divided by price-to-sales (1.06x), 15.3%, and base deal capacity grows 3.4% a year (debate judgment). Return definition: Acquisition ROIIC · mature cash proxy. Numerator 33.023 / denominator 218.172. Earnings basis: After-financing common-owner cash, after ongoing operating investment. The forecast runs from 2026-10-27 to 2031-10-27; a ten-year selection uses its own summed ledger. No historical lagged ROIIC is claimed; the CSU-family 2020–2025 aggregate acquisition yield that sets the purchase multiple is in `valuation/calculations.json` (aggregate steps). Average period reinvestment is total new growth investment / total matching earnings, not the arithmetic mean of annual rates. The base available-cash deployment control is 14.9%; annual purchases are also capped by the observed deal capacity, growing at the scenario's capacity growth, so the 27.1% five-year weighted rate is lower than an uncapped forecast. Matching earnings now include interest on retained cash. Old acquisition claims, dividends, debt repayment and financial stakes do not acquire a second increment of operating profit. Five-year base funding: cash generated before new investment 787.510, less distributions 0.000, less old claims 6.507, plus opening cash drawn / (cash built) -640.280, plus net borrowing / (deleveraging) 0.000. This funds growth capital 140.724 and separately funded financial stakes 0.000. Reconciliation error is below 0.000001. New-cohort earnings at exit / period investment: 15.9% (cohorts are flat for two years, then grow 3.4% a year). Limits: - A mature acquisition cash yield is a model proxy, not measured cohort ROIIC or a discounted project IRR. Previously funded acquisition claims and opening portfolio ramp are not new investments. - Organic product/development investment is embedded in expenses and not separated. The rate shown measures external acquisition reinvestment; it is not total economic reinvestment. - These are period-specific forecast diagnostics, not independently measured historical ROIIC or a project XIRR. A matched multi-year lagged capital/earnings history remains a separate research requirement. Definitions and original methodology links are in `docs/method/cash-flow-and-reinvestment.md`; source locators are in this company’s registry. Probabilities are separate judgments recorded in `research/11-scenario-probabilities.md`. # Scenario probabilities Assessment date: 2026-10-27. These are initial subjective judgments, with low confidence. Financial sources support the mechanisms, not measured numerical odds. The original authors did not provide or endorse these weights. | Horizon | Downside | Base | Upside | | --- | ---: | ---: | ---: | | 5 years | 30% | 50% | 20% | | 10 years | 35% | 40% | 25% | 5-year reasoning: Installed software and servicing contracts support cash durability, but project concentration, control/related-party issues and sparse target cash data limit confidence in both organic growth and acquisition deployment. 10-year reasoning: Longer customer/product transitions and a larger deployment opportunity set widen outcomes. Lower central weight reflects weaker visibility into a decade of purchases and minority-holder economics. Downside: Tender timing, weaker project conversion or governance concerns can stall growth and acquisitions. Cash retention limits this representative loss path, but severe contract or capital-allocation damage can be worse. Base: Existing servicing/development demand supports 5% after-spending cash growth. Base purchases begin near the reported annual deployment amount rather than assuming every available zloty finds a suitable target. Upside: Stronger product growth and a more productive acquisition pipeline support the 8% growth/twice-deployment case. The limited transaction sample and related-party safeguards limit its current weight. Review triggers: - Servicing recurrence, tender pipeline and project cash conversion. - Reported g.on/Mednow/DocLogix cash and updated purchase allocations. - Independent deal terms, minority rights and actual deployment. Each path represents a broad outcome region, not the probability of hitting an exact share price or IRR. Paths are a coarse approximation of a continuous joint distribution; the displayed downside and upside are not bounds. Joint driver dependencies must be considered, rather than multiplying marginal driver guesses. Scenario weights sum to 100% within this approximation. Severe outcomes remain possible outside the modeled paths. Price movement alone does not change these weights. A changed causal premise, acquisition capacity, financial report or scenario definition does. Local assumption sensitivities suspend the researched weights until the full distribution is reassessed. No weighted-average IRR is presented as an expected shareholder return. Sources: sgn-fy25, sgn-h126, sgn-gon, sgn-asseco-h126. Method: [Damodaran, chapter 33](https://pages.stern.nyu.edu/~adamodar/pdfiles/val3ed/c33.pdf), printed pp.894-897. See `valuation/model.json` for machine-readable weights and `docs/method/cash-flow-and-reinvestment.md` for the capital definitions. ## 2 November 2026 review Weights unchanged. Stronger preliminary Q3 profit raises the provisional earning-power base in every path, but it doesn't settle cash conversion, implementation lumpiness, target economics or controlling-shareholder risk. A four-number release can't support a finer numerical odds revision. Reassess after the 10 November statements. ## 3 November 2026 review Scenario definitions changed: deals are priced at the CSU-family aggregate and capped at the observed pace, with capacity growth of 0%/10%/15%. The upside no longer deploys a doubled resource fraction without limit. Weights are reassessed and kept, because the change constrains every path rather than shifting their relative likelihood. ## 3 November 2026 pipeline review Weights unchanged at 30/50/20 (5 years) and 35/40/25 (10 years). The five-year filings support the mechanisms behind each path: - **Downside:** cash accumulates without deployment, the pattern of 2021–2023. - **Base:** deals at the observed flat pace, with a held margin. - **Upside:** a faster pace, which has not yet been observed. They don't justify a finer numerical change. The bear mechanism in 05-thesis.md (margin back to about 21%, no deals, 12x; −4.5% over 5 years) lies below the downside path. It is a tail inside the downside region's 30% weight, not a fourth weighted path. Price changes alone do not move the weights. ## 3 November 2026: author calibration Scenario definitions changed in every path: - the starting point is the anchor run rate; - deals earn half a year when bought; - retained cash earns 1.56% after tax; - acquired businesses grow 3.4%; - the downside exit is 12.2x. These are measurement and timing corrections common to all paths. They don't shift the relative likelihood of the business outcomes, so the weights were reassessed and kept at 30/50/20 (5 years) and 35/40/25 (10 years). The Q3 report on 10 November is the next trigger. ## 3 November 2026: growth-durability review Weights kept at 30/50/20 and 35/40/25. The review split H1 2026 growth into the parent's durable organic growth, which supports a higher base and upside, and a KPO deadline wave inside Mednow, which is removed from every path. Neither finding changes which business outcome is more likely relative to the others; both change the level of every path. The upside's 11% organic growth sits above outside-view base rates, which argues against raising its weight. The 10 November Q3 report is the next trigger. ## 3 November 2026: bull–bear debate judgment Weights kept at 30/50/20 (5 years) and 35/40/25 (10 years). The judgment (12-debate.md) changed the scenario definitions: - the base and upside exit is 16.0x, and the downside 11.5x, at a 10.45% market cost of equity; - base deal-capacity growth is 3.4%; - the KPO wave removed from every path is 6.06. These change the level of the paths more than their relative likelihood, so the weights were reassessed rather than carried over. - **Bull's 25/50/25 and 30/45/25: rejected.** The attacks that survived rebuttal (cost of equity, KPO size, a floor-price tender) argue against more upside weight. The 11% upside already exceeds outside-view base rates. - **Bear's 40/45/15: withdrawn** by the bear as an unsupported judgment. Measured parent organic growth argues against more downside weight. - **The unresolved tender tail** is the reason the downside weight is not lowered. Next trigger: the 10 November Q3 report. ## 3 November 2026: revenue build Weights kept at 30/50/20 and 35/40/25. The paths now state revenue and margin per year. The downside gained an explicit margin decline (the bear's mechanism), and the base fades its growth. These restate the same three business outcomes in more detail; they do not change their relative likelihood. ## 4 November 2026: lumpy deal budget Weights kept. The yearly deal budget states the record's lumpiness inside each path: quiet years in the base, and a three-year dry spell in the downside. It does not change the relative likelihood of the three outcomes. ## 10 November 2026: Q3 report Weights kept at 30/50/20 and 35/40/25, reviewed against the nine-month report. The roll changed the level of every path (measured KPO wave, reported cash and claims), not their relative likelihood. Evidence that could move the weights, in both directions: - **Against the upside:** the parent's Q3 revenue grew 9.1%, against about 16.8% in H1 (nine months +14.0%). The upside's year-one 13% organic growth leans on the recent parent pace. One quarter after a deadline-driven H1 is not a trend, so the weight is not cut yet. A second quarter below 10% (FY2026 report) would be the trigger. - **Against the downside:** pretax cash matched profit, no new contingent liability appeared (the 8.2 public-client contingency was settled for 0.5), and the main lawsuit is in mediation with both proceedings suspended. - **Unresolved:** whether the 85.8 advance balance is KPO timing or recurring prepayment. This is held in the paths as a claim, not in the weights. No numerical odds come from the report. These remain subjective judgments. Next trigger: the FY2026 annual report. # Bull–bear debate (3 November 2026, PLN94) Run under `prompts/debate.md` after the valuation stage. The bull and bear advocates wrote their cases separately, then cross-examined each other. A judge who did not build the base case decided each contested point. The advocates' working papers (`sygnity-bull.md`, `sygnity-bear.md` and their rebuttals) are local scratch files and are not tracked. This page summarizes them. All returns come from the shared engine at PLN94 on 3 November 2026, with exits fixed at 27 Oct 2031 and 2036. Before the debate the model gave 0.62/14.73/18.81% at five years (12.18% weighted) and 2.95/13.24/17.03% at ten (11.83% weighted). ## Bull case The bull steelmanned the bear first: the 2022–2025 margin step was a one-off, wages rise about 8% a year, two-thirds of H1 2026 growth was acquired, and a controller hoards cash. His opening case then had five claims: 1. **Organic growth of 8.5%, not 6.51%.** The 2020–2025 window starts in the creditor restructuring. The parent company, which holds no acquisitions, grew 13.4% in 2025 and 16.5–16.8% in each first half since 2024. Its maintenance grew 17.4% and 23% (`parentRevenue*`, `parentMaintenance*`; sgn-h126 p.44, FY2025 JSF p.28). NBP raised the ZSK ceiling by 25 gross (sgn-nbp-contract). There were 37 public awards worth 195 gross (mimira-awards). The mechanism is legislation-forced change requests on systems the customer can't switch. 2. **A 20x exit.** NBP cut rates to 3.75%. Sygnity's own ex-cash multiples were 21.6–24.0x in 2023–2025 (`peExCash2023`–`2025`), and Asseco trades at 24.2x (asseco-pe-aggregator). 3. **A 40% payout.** The AGM reserved profit for "dividends and investment", and Topicus has paid a special dividend. 4. **A 2.8% after-tax return on cash** instead of the measured 1.56%. 5. **Weights of 20/55/25** at five years, because the record shows a deal every year since 2024 and parent growth every period. Combined, these gave a 5-year weighted return of 16.9%. The bull conceded up front the flat deal pace, the 6.61x deal price, a held (not expanding) margin, no take-private premium and removal of the KPO wave. ## Bear case The bear steelmanned the bull's installed-base franchise and said permanent loss is hard to construct with 194 of net cash. His case was that several optimistic choices stack: 1. **Q3 was the KPO peak, not a tail.** The preliminary Q3 shows revenue up 38.2 and pretax up 14.2 (`pbtQ3Change`; sgn-9m26-preliminary). Projects had to be complete by 15 July (kpo-d112-schedule). Taking out the parent's H1 pace and a Mednow base leaves about 6 of Q3 pretax, against the 1.6 assumed. Wave: 7.4, not 3.85. 2. **Roll the run rate at 6.51%, not 14%** (`runRateUplift`): −4.4 of owner cash. 3. **The EU deficit procedure caps public spending growth** at 4.4%, 4.0% and 3.5% for 2026–28. Base growth: 3.5%. 4. **An 18x exit needs a cost of equity below Poland's.** 6.47% 10-year yield (15 Sep 2026) plus Damodaran's 5.33% ERP gives about 11.8%, and the thesis's own Koller formula then gives 12.8–14.5x. 5. **Deals.** The pace is flat (34.3, 35.7, about 31.5), so capacity growth is 0%. g.on (a related-party purchase) has no stand-alone financials. 9.04x is the upper bound of the DocLogix price. 6. **Cash trapped under a controller:** a 25% haircut, and a tender at the art.79 floor (about PLN80–84) returning −6.6% a year. 7. **Weights of 40/45/15.** Combined (1+2+3+4+5a), these gave 6.67% base and 6.08% weighted over five years. ## Rebuttals Cross-examination narrowed both cases. - **Bull concessions.** The 20x exit was withdrawn once the long rate was shown to have risen. The 40% payout was withdrawn: no dividend since 2021, two AGMs without one. Growth was cut to 7.5% (a 12/9/7/5/4 fade from the parent's 16.8%). Weights were cut to 25/50/25 and 30/45/25. The KPO wave was accepted at the midpoint, 5.6. Deployment growth was accepted at 5%. - **Bull rebuttals.** The roll fills a past window, so observed growth belongs there, not the forward rate. The EU cap applies to total expenditure, and defence has been exempt under the national escape clause since July 2025. The bear's 11.8% counts country risk twice. The tender is voluntary: squeeze-out needs 95% (art.82), and Barca's 9.01% blocks it. - **Bear concessions.** Growth rose to 5.0%, since parent maintenance is measured evidence. The cash haircut and the 40/45/15 weights were dropped as unsupported judgments. The 9.04x deal price was not defended. - **Bear rebuttals.** The parent window coincides with temporary utility programmes (06-risks.md item 7). The NBP annex ends in December 2027. The policy rate is the wrong duration for valuing equity. Asseco's multiple is unreconciled and moved from 17.7x to 24.2x in one quarter. Flat means 0%, not 5%. - **Where they ended.** The revised bull case gave 15.16% base and 13.17% weighted at five years. The revised bear case gave 7.85% base and 6.69% weighted, or 8.01% weighted at a 16x exit. ## Judgment **Fact check.** Both advocates used a stale shared fact. The Polish 10-year yield was **6.31% on 3 November 2026**, not 6.47% (te-pl10y). The remaining checks: - **Damodaran (January 2026), Poland:** total ERP 5.33%, made of a 4.23% mature premium and a 1.10% country premium. The sovereign default spread is 0.72% (damodaran-ctryprem). His local-currency method subtracts the default spread from the local bond yield (damodaran-riskfree). So the bull is right that the bear double-counted, but only by the 0.72 points of default spread, not the whole premium. - **Asseco:** aggregators range from 9x to 24x depending on the denominator. The figure is unusable as an anchor for either side. - **EU deficit procedure:** net-expenditure path 4.4/4.0/3.5% for 2026–28 (pl-mof-mtfsp). The procedure is open. The defence escape clause was adopted on 8 July 2025 (council-nec-2025). Both are verified. - **Public Offering Act:** arts. 73, 79, 82 and 91 as recorded (polish-public-offering-act). - **Parent growth:** the facts reproduce +13.4%, +16.5% and +16.8% for revenue, and +17.4% and +23.3% for maintenance. - **NBP reference rate:** 3.75%, held through September 2026. | # | Claim | Bull | Bear | Evidence | Verdict | Reason | | --- | --- | --- | --- | --- | --- | --- | | 1 | Base organic growth | 7.5% (fade from 16.8%) | 5.0% (EDP ceiling) | Parent organic growth is measured. The fade path is asserted. The engine applies one rate to both horizons. The EU cap is an aggregate ceiling that exempts defence | **Split: keep 6.51%** | The bull's own path averages 7.4% over five years and 5.7% over ten; 6.51% sits between. The bear's cap covers a third of revenue and didn't stop public revenue growing 28% under the same procedure. Balanced evidence keeps the outside view, which is Sygnity's own through-cycle rate | | 2 | Exit multiple | 18x | 14.5–16x (r ≈ 11.8%) | Build-up r = 6.31 − 0.72 + 5.33 = 10.92% (Koller 14.7x). The market paid 17.5x on 3 Nov, which implies r = 9.98%. Own history 13.3–24x, but 2023–25 came with 20–35% growth. Asseco unusable | **Split, leaning bear: 16.0x** (`baseExitMultiple`) | 18x needs r ≈ 9.8%, below both estimates. The two estimates of the buyer's rate are weighted equally, giving r = 10.45%. That r also applies to the downside: 1/r = 9.6x, so the downside exit is 11.5x (was 12.2x) | | 3 | KPO wave in starting earnings | 5.6 | 7.4 | The Q3 subsidiary residual after the parent's H1 pace, the new Mednow PPA charge and the Mednow base is 7.0 pretax, against a 1.6 tail. The deadline makes Q3 a peak. The known contracts (22.8 gross, about 18.5 net) bound the total. The parent's Q3 pace is assumed | **Split: 6.06** (`kpoWavePretaxQ3Judged`) | The evidence contradicts a half-size tail (bear). The residual is an upper estimate, since it also absorbs subsidiary organic growth and any parent acceleration, and 9.1 pretax would be a 49% margin on the known contracts. The two estimates are weighted equally until the 10 Nov notes | | 5 | Base deployment growth | 5% | 0% | Three flat years: 34.3, 35.7, about 31.5. The family rate is 10%. More platforms is an asserted mechanism | **Split, mechanism-based: 3.4%** (`acquiredBusinessGrowth`) | The record supports a flat deal count, not 10% growth. A zero nominal rate for ten years would shrink the real pace. With a constant count and multiple, cash per deal grows with target earnings (the family's 3.4% organic rate). Upside keeps 15% | | 6 | EU deficit cap and defence exemption | the cap doesn't bind Sygnity | a 3.5% ceiling | Verified as above | **Bull on the mechanism; no input** | The ceiling applies to total net expenditure and exempts defence. It still limits civil budget growth to about 4% through 2028, which argues against going above the through-cycle rate. Recorded as a reason for verdict 1 | | 7 | Tender floor versus refusal and squeeze-out | refusal blocks it | −6.6% path | Arts. 73, 79, 82, 91. TSS already holds more than 50%, so any tender is voluntary. Delisting needs 9/10 of votes cast with Barca present | **Unresolved; no input** | Not a forced outcome while Barca holds 9.01%. It is still a coercive tail: a tender at the floor after a drawdown, with a delisting threat. It stays inside the downside weight, which is therefore not lowered | | 8 | Weights | 25/50/25; 30/45/25 | recorded (40/45/15 withdrawn) | Several attacks survive (see Conviction) | **Keep 30/50/20; 35/40/25** | The surviving attacks argue against raising the upside weight. Parent evidence argues against raising the downside weight | | 9 | Cash return | 2.8% | measured 1.56% | H1 2026 deposits yielded 1.92% pre-tax | **Bear** | Measured, not hypothetical. +0.14 points if changed | | 10 | Deal price 9.04x; g.on value; payout | — | — | — | **6.61x kept; g.on unresolved; payout withdrawn** | 9.04x is a definitional bound, not a price. g.on lacks data rather than having contrary data. A payout is a prospective check at the June 2027 AGM | **Model effect** (PLN94, 3 Nov 2026, exits fixed; five-year weighted change in brackets): | Step | 5-year d/b/u, weighted | 10-year d/b/u, weighted | | --- | --- | --- | | Before | 0.62 / 14.73 / 18.81, 12.18 | 2.95 / 13.24 / 17.03, 11.83 | | KPO wave 3.85 → 6.06 (baseline 115.7 → 113.3) | 0.24 / 14.33 / 18.38, 11.78 (−0.40) | 2.75 / 13.06 / 16.83, 11.64 | | Base deployment growth 10% → 3.4% | 0.24 / 14.03 / 18.38, 11.62 (−0.16) | 2.75 / 12.50 / 16.83, 11.40 | | Base and upside exit 18x → 16.0x | 0.24 / 11.86 / 16.06, 9.86 (−1.76) | 2.75 / 11.53 / 15.75, 10.56 | | Downside exit 12.2x → 11.5x (after) | **−0.59 / 11.86 / 16.06, 9.69** (−0.17) | **2.43 / 11.53 / 15.75, 10.50** | Single-claim sensitivities around the judged model (5-year weighted 9.69%): - exit 14.7x (build-up r only): 8.53%; - exit 17.5x (market-implied r only): 11.02%; - growth 5.0%: 9.03%; - growth 7.5%: 10.13%; - KPO wave 7.4: 9.45%; - roll at 6.51%: 8.96%; - the bull's weights: 10.46%. Even the bull's weights would leave it below 12%. Named steps are in `valuation/calculations.json`, inputs and reasons in `valuation/model.json` inputNotes, and the record in `updates/2026-11-03-debate-judgment.md`. ## Conviction Attacks that survived rebuttal: 1. **Exit multiple and cost of equity (bear).** The verified 6.31% PLN 10-year yield and a correctly built 10.9% cost of equity put the Koller value at 14.7x. Only the market's own 17.5x supports more. This is the largest single driver: about 1.8 points of five-year weighted return. 2. **The size of the KPO wave (bear, partly).** Q3 profit contradicts a half-size tail. The true wave is 4–7 of owner cash and is unknown until 10 November. 3. **Deal capacity (bear, partly).** Three flat years; the family's 10% growth has not been shown at Sygnity. 4. **A controller tender at the floor (bear, as a tail).** Not forced, but available to TSS after a drawdown. It caps the minority's upside at a negotiated premium. 5. **Near-term organic momentum (bull).** Parent growth of 13–17%, measured and organic, survived the bear's attacks on its duration but not on its level. It is the main reason the base is not lower. **Conviction: low to moderate.** The business case survives: an installed-base franchise, measured organic growth, net cash and no plausible permanent loss in the representative paths. The investment case does not survive at PLN94. The judged 5-year weighted return is 9.7% (10.5% at ten years), below a 12% hurdle. The exit-multiple attack was rebutted only by pointing to today's price, which can't be the evidence that the price is cheap. **What would change it:** the 10 November Q3 report. Parent nine-month organic growth of at least 15%, with subsidiaries' Q3 profit showing the KPO wave was near the 3.85 tail, would restore about 0.4 points and support the bull on growth. Even then the cost-of-equity verdict leaves the weighted return near 10%. On the same model, the five-year weighted return reaches 12% at about PLN85 and 13.3% at PLN80, which is around the art.79 floor area. A durable fall in Polish long yields towards 5% would lower the build-up cost of equity. Either would make the case investable; neither would raise conviction in the business itself. ## After the judgment: revenue build (3 November 2026) The judged verdicts are kept, and the model now projects revenue and margin year by year (08-underwriting-plan.md, Yearly paths). A calibration with constant paths reproduced the judged figures exactly: 5-year −0.59/11.86/16.06%, 9.69% weighted. The revenue build then exposed three things the scalar engine hid: - **Organic growth fades instead of staying constant.** The base runs 9.5% / 7% / 5.5%, with a 10-year mean of 6.2% against the judged 6.51%. Verdict 1's through-cycle anchor is kept as the long-run level. The deficit-procedure ceiling (claim 6) is the stated reason for the years 2–3 fade. - **Acquired businesses earn 16.2%, not the group's 23%.** At 1.06x revenue that is 6.6x cash, so the judged deal price stands. - **The downside now carries a margin decline to 21.5% from year 2.** This is the bear's margin-regression mechanism, stated explicitly. New returns at PLN94: 5-year −2.08 / 12.28 / 16.18%, 9.65% weighted; 10-year 1.63 / 11.40 / 15.11%, 10.07% weighted. The conviction rating above is unchanged. ## After the judgment: lumpy deal budget (4 November 2026) The engine now takes a stated yearly deal budget. Measured record: 0, 0, 0, 34.3 and 35.7 (2021–25), and 35.1 so far in 2026. The 2021–25 mean is 14.0, and the best year is 2.55 times the mean. The TSS-era mean is 26.3, the active-era mean 35.1, and the longest quiet run three years. The budget refines the 3.4% capacity-growth verdict rather than replacing it. The base uses the expected budget including quiet-year frequency: the active-era pace (35.1) × the TSS-era active-year frequency (0.75) = 26.3, grown at the 3.4% verdict. No lump or quiet year is placed in a specific year, because no deal timing is evidenced (rule in prompts/stages/08-forecast.md). An earlier draft put quiet years in years 3 and 8 and gave 11.86% / 9.47%; that draft was replaced. The downside repeats the three-year dry spell, then deploys the record mean. A calibration with the smooth path reproduced the earlier figures exactly. New returns at PLN94: 5-year −1.83 / 11.71 / 16.19%, 9.39% weighted; 10-year 2.12 / 10.90 / 15.13%, 9.94% weighted. ## Public Primary & Secondary Sources - `sgn-fy25` (primary): Sygnity FY2025 consolidated financial statements — https://www.sygnity.pl/wp-content/uploads/2026/04/Sprawozdanie-FY25.zip (published 2026-03-24) - `sgn-h126` (primary): Sygnity H1 2026 consolidated and separate statements — https://www.sygnity.pl/wp-content/uploads/2026/09/SSF-i-JSF-SGN-H126-EN.pdf (published 2026-09-03) - `sgn-h126-pl` (primary): Sygnity H1 2026 original Polish statements — https://www.sygnity.pl/wp-content/uploads/2026/09/SSF-i-JSF-SGN-H126.pdf (published 2026-09-03) - `sgn-presentation` (primary): Sygnity FY2025 investor presentation — https://www.sygnity.pl/wp-content/uploads/2026/04/Financial-results-of-the-Sygnity-Group-for-the-financial-year-2025.pdf (published 2026-04-22) - `sgn-gon` (primary): g.on acquisition from controlling shareholder — https://www.sygnity.pl/wp-content/uploads/2026/10/CR_27_2026-Notice-regarding-the-acquisition-of-shares-in-a-foreign-entity-by-a-subsidiary-of-Sygnity.pdf (published 2026-10-02) - `sgn-facility` (primary): Sygnity credit-facility amendment — https://www.sygnity.pl/wp-content/uploads/2026/09/RB-26_2026-EN.pdf (published 2026-09-07) - `sgn-calendar` (primary): Revised Q3 2026 publication date — https://www.sygnity.pl/wp-content/uploads/2026/10/RB-28_2026-EN.pdf (published 2026-10-13) - `sgn-ir` (primary): Sygnity issuer reports and announcements — https://www.sygnity.pl/en/for-investors/ - `sgn-business` (primary): Sygnity business, history and management — https://www.sygnity.pl/en/about-us/ - `sgn-nbp-contract` (primary): National Bank of Poland maintenance/development contract amendment — https://www.sygnity.pl/wp-content/uploads/2026/06/CR-21_2026-Update-on-the-agreement-with-the-National-Bank-of-Poland.pdf (published 2026-06-12) - `sgn-asseco-h126` (primary): Competitor Asseco first-half 2026 results — https://inwestor.asseco.com/en/about-asseco/press-releases/asseco-summarizes-the-first-half-of-2026-5747/ (published 2026-08-27) - `sgn-fx` (primary): NBP EUR/PLN table A on acquisition date — https://api.nbp.pl/api/exchangerates/rates/a/eur/2026-10-02/?format=json (published 2026-10-02) - `sgn-quote` (market data): SGN.WA delayed quote observed for underwriting — https://finance.yahoo.com/quote/SGN.WA/ - `capital-method-roiic` (methodology): Mauboussin and Callahan: Return on Invested Capital — https://www.morganstanley.com/im/publication/insights/articles/article_returnoninvestedcapital.pdf (published 2022-10-06) - `capital-method-cash` (methodology): Dan Shuart, Eagle Point Capital: A Framework For Spotting Value Traps — https://eaglepointcapital.substack.com/p/spotting-value-traps-the-anti-compounders (published 2023-08-26) - `scenario-method` (methodology): Aswath Damodaran: Probabilistic Approaches in Valuation, chapter 33 — https://pages.stern.nyu.edu/~adamodar/pdfiles/val3ed/c33.pdf - `sgn-9m26-preliminary` (primary preliminary issuer release): Sygnity current report 29/2026: preliminary January–September 2026 consolidated results — https://www.sygnity.pl/wp-content/uploads/2026/11/RB_29_2026-Informacja-nt.-wstepnych-wynikow-finansowych-za-III-kwartaly-roku-obrotowego-2026-.pdf (published 2026-11-02) - `sgn-q325` (primary): Sygnity nine-month 2025 consolidated and separate statements — https://www.sygnity.pl/wp-content/uploads/2025/11/SSF-i-JSF-SGN-Q325_EN.pdf (published 2025-11-12) - `sgn-ir-pl` (primary): Sygnity original-language financial report index — https://www.sygnity.pl/dla-inwestorow/ - `csi-fy2021-mda` (primary): CSU FY2021 management's discussion and analysis — https://www.csisoftware.com/wp-content/uploads/2026/04/CSI-MDA-Q4-2021-Final.pdf - `csi-fy2022` (primary): CSU FY2022 shareholder report — https://www.csisoftware.com/wp-content/uploads/2026/04/Q4-2022-Shareholder-Report.pdf - `csi-fy2023` (primary): CSU FY2023 shareholder report — https://www.csisoftware.com/wp-content/uploads/2026/04/Q4-2023-Shareholder-Report.pdf - `csi-fy2025` (primary): CSU FY2025 shareholder report and audited financial statements — https://www.csisoftware.com/wp-content/uploads/2026/04/Q4-2025-Shareholder-Report.pdf - `lumine-fy2023-mda` (primary): Lumine FY2023 management's discussion and analysis — https://www.luminegroup.com/wp-content/uploads/2025/02/Lumine-MDA-2023-Q4.pdf - `csi-q2-2026` (primary): CSU Q2 2026 shareholder report and interim financial statements — https://www.csisoftware.com/wp-content/uploads/2026/08/Q2-2026-Shareholder-Report.pdf (published 2026-08-11) - `sgn-fy21` (primary): Sygnity annual report FY Oct 2020-Sep 2021 (consolidated statements and activity report) — https://www.sygnity.pl/wp-content/uploads/2022/02/Annual-report-separate-and-consolidated-data-FY21-1.zip - `sgn-fy21-report` (primary): Sygnity activity report for FY Oct 2020-Sep 2021 — https://www.sygnity.pl/wp-content/uploads/2022/02/Annual-report-separate-and-consolidated-data-FY21-1.zip - `sgn-fy22` (primary): Sygnity annual report FY Oct 2021-Sep 2022 (consolidated statements) — https://www.sygnity.pl/wp-content/uploads/2023/02/Raport-roczny-2022-EN.zip - `sgn-fy22-report` (primary): Sygnity activity report FY Oct 2021-Sep 2022 — https://www.sygnity.pl/wp-content/uploads/2023/02/Raport-roczny-2022-EN.zip - `sgn-fy23` (primary): Sygnity annual report for the 15-month year Oct 2022-Dec 2023 — https://www.sygnity.pl/wp-content/uploads/2024/04/Annual-Report-FY24-1.zip (published 2024-04) - `sgn-fy23-report` (primary): Sygnity activity report for the 15-month year Oct 2022-Dec 2023 — https://www.sygnity.pl/wp-content/uploads/2024/04/Annual-Report-FY24-1.zip (published 2024-04) - `sgn-fy24` (primary): Sygnity annual report calendar 2024 (consolidated statements) — https://www.sygnity.pl/wp-content/uploads/2025/05/Annual-Report-FY24.zip (published 2025-05) - `sgn-fy24-report` (primary): Sygnity report on operations, calendar 2024 — https://www.sygnity.pl/wp-content/uploads/2025/05/Annual-Report-FY24.zip (published 2025-05) - `sgn-q121` (primary): Sygnity interim statements, 3 months to 31 Dec 2020 — https://www.sygnity.pl/wp-content/uploads/2021/03/SSF_GK_SGN_Q121_EN.pdf - `sgn-q122` (primary): Sygnity interim statements, 3 months to 31 Dec 2021 — https://www.sygnity.pl/wp-content/uploads/2022/03/Interim-Condensed-Consolidated-And-Separate-Financial-Statement-for-the-period-of-3-months-ended-on-31-December-2021.pdf - `sgn-rem25` (primary): Sygnity supervisory board remuneration report for 2025 — https://www.sygnity.pl/wp-content/uploads/2026/05/Report-of-the-Supervisory-Board-on-remuneration-for-the-financial-year-2025.pdf (published 2026-05-06) - `sgn-tss-call` (primary): Sygnity management board position on the TSS Europe tender offer — https://www.sygnity.pl/wp-content/uploads/2023/09/Stanowisko-Zarzadu-Sygnity-S.A.-dot.-wezwania-do-zapisywania-sie-na-sprzedaz-akcji-Sygn_EN.pdf (published 2022-04-06) - `sgn-mbank-2024` (secondary): mBank brokerage research note on Sygnity (sell, target PLN39) — https://biznes.pap.pl/download/attachment/49778285/DOC.20241010.49778285.Sygnity_20241010.pdf (published 2024-10-10) - `sgn-price-history` (secondary): SGN.WA daily price and volume history — https://query1.finance.yahoo.com/v8/finance/chart/SGN.WA?range=5y&interval=1d - `asseco-pe-aggregator` (secondary): Asseco Poland trailing P/E (aggregator) — https://multiples.vc/public-comps/asseco-poland-valuation-multiples - `sgn-q126` (primary): Sygnity Q1 2026 interim statements — https://www.sygnity.pl/wp-content/uploads/2026/06/SSF-i-JSF-SGN-Q126_EN.pdf - `sgn-h125` (primary): Sygnity H1 2025 interim statements and management report — https://www.sygnity.pl/wp-content/uploads/2025/09/H1-FY25-1.zip - `sgn-q125` (primary): Sygnity Q1 2025 interim statements — https://www.sygnity.pl/wp-content/uploads/2025/05/SSF_GK_SGN_Q125__EN.pdf - `sgn-calendar-2026` (primary): Sygnity calendar of corporate events 2026 — https://www.sygnity.pl/wp-content/uploads/2026/10/Calendar-of-corporate-events-2026.pdf - `kpo-d112-schedule` (primary): Polish Ministry of Health: updated schedule for KPO investment D1.1.2 (digital transformation of healthcare) — https://www.gov.pl/attachment/8e024326-0f21-4b31-a8cc-b0b87b2657b9 (published 2026-03-11) - `topicus-asseco` (primary): CSU and Topicus joint press release: completion of Asseco Poland treasury-share purchase — https://www.csisoftware.com/joint-press-release-of-constellation-software-inc-and-topicus-com-inc-topicus-com-inc-completes-agreement-to-acquire-14-84-of-treasury-shares-in-the-capital-of-asseco-poland-s-a/ - `polish-public-offering-act` (primary): Act on Public Offering (Dz.U.2025.592 consolidated): arts. 73, 79, 82, 83, 91 — https://lexlege.pb.pl/ustawa-o-ofercie-publicznej-i-warunkach-wprowadzania-instrumentow-finansowych-do-zorganizowanego-systemu-obrotu-oraz-o-spolkach-publicznych/art-91/ - `mimira-awards` (secondary): Mimira tender aggregator: contracts won by Sygnity S.A. — https://mimira.eu/przetargi/wykonawca/sygnity-s-a - `sii-2q26` (secondary): SII: Sygnity after Q2 2026, highest revenue in nearly ten years with an unexplained structure — https://www.sii.org.pl/19349/analizy/analizy-i-komentarze/sygnity-po-2q-2026-najwyzsze-przychody-od-blisko-10-lat-a-ich-struktury-spolka-nie-tlumaczy-analiza.html - `itwiz-comarch-his` (secondary): ITwiz: Comarch sells its HIS segment to Sygnity — https://itwiz.pl/comarch-sprzedaje-segment-his-firmie-sygnity/ (published 2025-10-03) - `tip-podcast-csu-spinoffs` (secondary): The Investors Podcast TIVP066: valuing Constellation spinoffs (Topicus, Lumine, Sygnity, Asseco) — https://www.theinvestorspodcast.com/intrinsic-value-podcast/valuing-constellation-softwares-spinoffs-topicus-lumine-and-co-w-daniel-mahncke-shawn-omalley/ (published 2026-04-05) - `te-pl10y` (secondary market data): Trading Economics: Poland 10-year government bond yield — https://tradingeconomics.com/poland/government-bond-yield (published 2026-11-03) - `damodaran-ctryprem` (secondary academic dataset): Damodaran Online: country default spreads and risk premiums (January 2026 update) — https://pages.stern.nyu.edu/~adamodar/New_Home_Page/datafile/ctryprem.html (published 2026-01-05) - `damodaran-riskfree` (secondary method): Damodaran, "What is the riskfree rate? A search for the basic building block" (NYU Stern working paper, December 2008) — https://pages.stern.nyu.edu/~adamodar/pdfiles/papers/riskfreerate.pdf (published 2008-12-01) - `pl-mof-mtfsp` (primary): Polish Ministry of Finance: medium-term fiscal-structural plan 2025-2028 and the Council corrective net-expenditure path — https://www.gov.pl/web/finance/medium-term-fiscal-structural-plan (published 2025-01-21) - `council-nec-2025` (primary): Council of the EU: coordinated activation of the national escape clause for defence (Poland among 15 member states) — https://data.consilium.europa.eu/doc/document/ST-10475-2025-INIT/en/pdf (published 2025-07-08) - `sgn-q326` (primary): Sygnity nine-month 2026 consolidated and separate statements (Polish original) — https://www.sygnity.pl/wp-content/uploads/2026/11/SSF-i-JSF-SGN-Q326.pdf (published 2026-11-10) --- # TerraVest (TVK.TO) - Legal Name: TerraVest Industries Inc. - Exchange: Toronto (Canada) - Sector: Industrial manufacturing - Currency: CAD - Reviewed: 2026-11-04 - Model Framework: Acquisition and restructuring compounder - Financial Period: TTM to June 2026; June bank debt/shares; March claim estimates; July SPV purchase (anchored 2026-10-26) - Research Status: 8 of 11 research checks complete. Still to complete: Security and evidence base; Opportunities and funded scenarios; Variant view and catalysts - Static Report URL: https://longview-investing.vercel.app/research/terravest.html - Interactive Model URL: https://longview-investing.vercel.app/#/company/terravest - JSON API: https://longview-investing.vercel.app/api/companies/terravest ## Executive Summary Buy industrial businesses, improve their cash earnings and redeploy within debt capacity. ## Key Operational Watch Items - Deal cost and NOPAT after restructuring, including all claims - Cash generation, debt capacity and refinancing headroom - Replacement demand versus structural decline and cyclical orders ## Valuation Scenario Assumptions ### Downside Case Borrowing capacity falls with earnings, stopping deals while debt exceeds the ceiling. The original 13× EV/NOPAT multiple prices the reduced earning power. Revenue and cash margins are now explicit yearly paths with matched purchased sales; current KBK missing months and all-in costs replace the old half-year shortcut. - Key Drivers: Exit EV / normalized cash NOPAT: 13.00× | Net debt / existing-business EBITDA: 2.40× - Organic revenue growth: 0.0%, 0.0%, 0.0%, 0.0%, 0.0% - NOPAT margin on organic revenue: 4.8%, 6.0%, 7.2%, 7.2%, 7.2% - EBITDA margin on organic revenue (debt capacity): 9.0%, 11.3%, 13.5%, 13.5%, 13.5% - Price paid per unit of acquired revenue: 0.88×, 0.88×, 0.88×, 0.88×, 0.88× - Recovered NOPAT margin of acquired businesses: 3.7%, 4.6%, 5.5%, 5.5%, 5.5% ### Base Case Pay interest, dividends and identified vendor/earnout claims before deploying capacity. Existing-business net debt capacity remains 2.4× EBITDA; new deals are small (the deployment budget cannot buy another Entrans) and are priced at the sellers’ trailing earnings on Highland and KBK’s issuer multiples (7.40× mature NOPAT), reaching that level over the three-year ramp with no restructuring uplift. Value terminal cash NOPAT at the original 17× EV multiple. Revenue and cash margins are now explicit yearly paths with matched purchased sales; current KBK missing months and all-in costs replace the old half-year shortcut. - Key Drivers: Exit EV / normalized cash NOPAT: 17.00× | Net debt / existing-business EBITDA: 2.40× - Organic revenue growth: 0.0%, 0.0%, 0.0%, 0.0%, 0.0% - NOPAT margin on organic revenue: 9.6%, 9.6%, 9.6%, 9.6%, 9.6% - EBITDA margin on organic revenue (debt capacity): 18.0%, 18.0%, 18.0%, 18.0%, 18.0% - Price paid per unit of acquired revenue: 0.88×, 0.88×, 0.88×, 0.88×, 0.88× - Recovered NOPAT margin of acquired businesses: 11.9%, 11.9%, 11.9%, 11.9%, 11.9% ### Upside Case Keep zero organic growth, the three-year ramp, the 2.4× existing-business debt ceiling and the 17× exit. This isolates better purchase economics; it adds no leverage, free growth or rerating. Revenue and cash margins are now explicit yearly paths with matched purchased sales; current KBK missing months and all-in costs replace the old half-year shortcut. - Key Drivers: Exit EV / normalized cash NOPAT: 17.00× | Net debt / existing-business EBITDA: 2.40× - Organic revenue growth: 0.0%, 0.0%, 0.0%, 0.0%, 0.0% - NOPAT margin on organic revenue: 9.6%, 9.6%, 9.6%, 9.6%, 9.6% - EBITDA margin on organic revenue (debt capacity): 18.0%, 18.0%, 18.0%, 18.0%, 18.0% - Price paid per unit of acquired revenue: 0.88×, 0.88×, 0.88×, 0.88×, 0.88× - Recovered NOPAT margin of acquired businesses: 18.9%, 18.9%, 18.9%, 18.9%, 18.9% ## Research Notebook # Business model TerraVest buys small and mid-sized North American manufacturers of steel and fibreglass tanks, pressure vessels, tank trailers, heating appliances and Western Canadian energy equipment and services. It then restructures purchasing, plants and incentives and leaves the subsidiaries to run with a very small head office. Money is C$ millions unless stated. Fiscal years end 30 September. TTM means the twelve months to 30 June 2026 (FY2025 + 9M FY2026 − 9M FY2025). ## Value-creation mechanism The mechanism is acquire cheaply, restructure, then redeploy the cash within a leverage ceiling. The levers are concrete and mostly on the cost side. Steel is roughly half of a tank's cost of goods, so mill-direct purchasing at group volume, consolidating nearby plants, sharing welders and a profit-not-revenue mindset raise margins on the same price. The second step, buying, matters more than it used to. The deals are now far larger (Entrans cost C$779.6m in March 2025, against C$240m for all eighteen deals from 2014 to 2022), so the purchase multiple and the cycle at purchase now drive more of the value than plant-level restructuring does. Re-underwriting today (`research/07-assumption-register.md`, 3 November 2026) found the FY2024–TTM cohort earning 16.6× its capital, against 9.9× at the sellers' trailing earnings. That gap is the trailer downturn at Entrans, not a failed restructuring yet. The value step to watch is therefore no longer only whether TerraVest can restructure, but whether it can buy at cycle-normal earnings when the deals are large enough to matter. ## Business-model profile Comparable firms are chosen by these traits, not by the industry label. **Growth vehicle: Acquisitions.** Revenue rose from 307.5 in FY2021 to 1,371.2 in FY2025. Of the 1,071.1 invested for growth in FY2024–FY2025, net acquisitions took 999.5 (93%); organic working capital and growth capex took the remaining 7%. Acquisition cash was 34.6% of revenue over FY2023–FY2025, Entrans dominating FY2025. The model funds acquisitions from cash and borrowing capacity each year, with a quiet year where debt is above the limit. **Capital intensity: 5.3% of sales, capital-intensive.** Capital spending, including the rental fleet, is about 5.3% of revenue: plants for steel and fibreglass tanks, pressure vessels and trailers, and rental equipment. Tangible capital including leases was 892.3 at September 2025, mostly working capital and plant. The capital that drives growth is acquisition cash (34.6% of revenue over FY2023–FY2025). **Comparables.** A comparable firm buys small and mid-sized manufacturers in fragmented industrial niches, funds deals from cash flow and debt capacity, improves purchasing and plants after purchase, and grows organically only with its end markets. Its revenue follows energy, agriculture and housing cycles. Not the label: "Industrial machinery" suggests an organic manufacturer; TerraVest grows by acquisition and is constrained by debt capacity. Declared in `valuation/model.json` (`businessProfile`). `npm run check` measures deal and capital intensity from the reported facts, and the outside view grades the forecast only against firms that share them. ## Economic snapshot Definitions. Returns use TTM June 2026 earnings against the 30 September 2025 balance sheet, the latest audited one (`terravest-fy2025-fs`). The mismatch matters a little: KBK (C$125m, January 2026) is half in TTM earnings but not in September capital. Earnings are normalized cash NOPAT on the baseline definition: adjusted EBITDA less issuer maintenance capex (in place of depreciation), lease principal and lease interest, taxed at 25%. Before minorities this is 178.2 (`preMinorityNopat`, `valuation/calculations.json`); after the GES minorities it is 166.0. | Capital at 30 Sep 2025 | C$m | | --- | ---: | | PP&E (incl. rental fleet) | 302.8 | | Right-of-use assets | 223.0 | | Operating working capital excl. cash (receivables 210.4 + inventory 351.6 + other 37.6 − payables 172.5 − customer deposits 58.2 − tax 2.6) | 366.4 | | **Tangible operating capital, leases capitalized** | **892.3** | | Acquired intangibles 390.1 + goodwill 461.1 | 851.1 | | Bank and term debt 754.0; lease liabilities 237.0; cash 13.9 (no excess cash) | | | Investments at fair value (non-operating, credited separately) 63.1 | | | Return (after tax, TTM) | Numerator | Capital | Return | | --- | ---: | ---: | ---: | | ROTC: leases capitalized (NOPAT before minorities + after-tax lease interest 13.2) | 191.5 | 892.3 | 21.5% | | ROCE: leases expensed, tangible (PP&E + working capital) | 178.2 | 669.3 | 26.6% | | ROCE including acquired goodwill and intangibles | 178.2 | 1,520.4 | 11.7% | | Same, adding the H1 FY2026 deal capital (132.9) | 178.2 | 1,653.3 | 10.8% | Tangible returns have since risen; returns including purchase premiums have fallen to about 11%, because FY2025 bought C$960m of businesses, mostly goodwill and intangibles. The operating businesses are good on their own tangible capital. The question is what TerraVest earns on the price it pays for them. | Margins and turns | FY2021 | FY2022 | FY2023 | FY2024 | FY2025 | TTM Jun-26 | | --- | ---: | ---: | ---: | ---: | ---: | ---: | | Revenue | 307.5 | 576.7 | 678.4 | 911.8 | 1,371.2 | 1,731.2 | | Gross margin | 26.2% | 21.6% | 24.5% | 28.9% | 27.8% | 25.2% (9M) | | Adjusted EBITDA margin | 21.4% | 15.0% | 17.9% | 20.8% | 19.3% | 18.0% | | Cash NOPAT / revenue | | | | 11.3% | 10.9% | 9.6% | FY2021 EBITDA includes C$18.1m of COVID subsidies; without them the margin was 15.5% (`covidSubsidiesFY2021`). Inventory was 119 days of cost of sales at FY2024 and 130 days at FY2025; the year-end figure overstates the run rate because it holds all of Entrans's inventory against 6.5 months of its costs. Receivables were 51 and 56 days. TerraVest deliberately carries steel ahead of its peak quarters (Q1 and Q4), so a 120–140 day inventory is the business model, not slippage. The incremental EBITDA margin from FY2023 to TTM was 18.1% (190.6 on 1,052.8 of added sales), almost all acquired. Normalization. TTM maintenance capex (41.2) now roughly equals FY2025 PP&E depreciation (41.3), so the baseline charges about 100% of depreciation. Q3 FY2026 maintenance rose to 14.4 from 4.6 a year earlier, mainly in Service; that is used rather than smoothed away. Minorities (12.2 TTM) are mostly the 33% outside stake in GES (Service). The Entrans earnout reversal (48.3) and bargain-purchase gains (5.3 in 9M FY2026) are excluded from earning power. ## Economic segmentation Reported segments mix economics. Entrans tank trailers sit in Compressed Gas Equipment next to propane storage tanks. Data-centre fuel tanks are built in HVAC and Containment but partly in other segments' plants. Water-management services in Service are 67% owned. Segment adjusted EBITDA (pre-lease, before corporate) for 9M FY2026 was HVAC 118.9, Compressed Gas 58.2, Processing 7.6 and Service 52.8 (Q3 MD&A pp.7–10). Adding an estimated Q4 FY2025 (FY2025 segment net income + D&A + financing less 9M FY2025 adjusted EBITDA) gives an approximate TTM of HVAC 143, Compressed Gas 101, Processing 11 and Service 69. That totals about 325, against a group TTM of 312 after corporate costs. Re-cut by how profit is made (estimates; the HVAC heating/tank split and the Compressed Gas trailer/tank split are not disclosed): | Economic unit | Approx. TTM EBITDA | Share | Demand driver | Character | | --- | ---: | ---: | --- | --- | | Stationary tanks and pressure vessels: propane/NH3 bulk and residential tanks, refined-fuel and data-centre generator tanks (Highland, KBK, Simplex, Granby tanks) | 130–155 | 40–48% | Replacement of an installed base; c-store and fuel storage rules; data-centre build | Replacement core plus a project boom | | Transport equipment: tank trailers, bobtails (Entrans, LBT, Tankcon, Advance, MTC, SPV) | 60–75 | 18–23% | Fleet replacement by chemical, fuel, food-grade and LPG carriers | Cyclical; at a trough | | Heating appliances: cast-iron boilers and furnaces (ECR, Granby) | 30–45 | 10–14% | Replacement in older North-eastern homes | Slowly declining, cash-generative | | Western Canadian energy: water management, flowback, rentals, processing equipment (GES/Fraction, Colter, NWP) | about 80 | about 25% | Western Canadian drilling and completion activity | Cyclical; one-third owned by minorities | The trailer unit is why the TTM is below purchase-time earnings. Compressed Gas's adjusted margin fell from 16.5% in FY2024 to 9.6% for 9M FY2026 and 8.7% in Q3. HVAC's margin (25.7% for 9M FY2026, against 24.5% a year earlier and 25.5% in FY2024) did not jump. The data-centre boom is therefore volume at normal margins, not a price windfall. HVAC base-portfolio sales rose 66 in 9M FY2026, roughly 22 of annualized EBITDA, or about 7% of group EBITDA. If that project volume fades, it is a cyclical give-back of that size, not a margin collapse. ## Unit economics The unit through which TerraVest adds value is the acquired company. Three units, in absolute money: Gross profit per vessel goes from about C$70k to C$125k, and gross margin from 20% to 34%. Almost all of the gain is buying power and labour productivity on an unchanged product. **Highland Tank (November 2023), a good recent deal.** Price 106.5 (including 6.9 in shares; `highlandConsideration`), at about 4.0× the seller's TTM EBITDA of about 26.6 (issuer release). On the price paid that is a 19.7% after-tax yield (5.1×), against a 9–10% cost of capital. Highland then became the platform for the data-centre tank volume and for Simplex (April 2025, EV about US$28m). **Entrans (March 2025), the large deal at the wrong point in the cycle.** | Entrans | At purchase | TTM June 2026 (estimate) | | --- | ---: | ---: | | Capital: cash consideration (earnout not paid) | 779.6 | 779.6 | | EBITDA after rent | about 111 (7.0× issuer multiple) | about 56 (pre-rent IFRS, acquired trailer businesses incl. LBT and Tankcon) | | NOPAT (group conversion 0.64 after rent; 0.53 pre-rent) | about 71 | about 30 | | After-tax yield on capital | 9.1% | about 3.8% | The up-to-US$46m earnout depended on Entrans's first twelve months. It was reversed as a C$48.3m gain in Q1 FY2026 (`entransEarnoutGain9mFY2026`), so the seller's own targets were missed. Compressed Gas Q3 FY2026 EBITDA was 17.3 against 29.7 a year earlier, as US tank-trailer demand fell (Q3 MD&A p.8). A unit earning 3.8% on its price destroys value at a 9–10% hurdle until it recovers. On a two-to-three-year view, Entrans is genuinely different from the bolt-on pattern. It was bought at near-peak trailer earnings, already had scale purchasing, and its shortfall is volume, not cost. Its recovery is upside to existing earnings, not evidence about future bolt-on prices. Entrans already had scale purchasing as the industry leader, so the 15% steel saving on a mom-and-pop does not transfer automatically. **Failures, with reasons.** - Iowa Steel Fabricators (bought 2019 for about 11.6): closed in FY2026 (Q3 MD&A p.8). - The biogas activity in Processing: terminated early in FY2026 after low-margin jobs. These exits are evidence that the team cuts losers rather than funding them. They also show the lever works only where TerraVest's purchasing scale is large relative to the supplier. ## Independent checks 2. **Peers' audited results.** Wabash's Q3 2026 release (`wabash-q3-2026`) reports an adjusted operating loss on weak trailer demand, and ACT reports depressed trailer orders (`act-trailers-october-2026`). The Entrans shortfall is industry-wide, so it does not by itself show an integration failure. A failed integration would show as Compressed Gas margins below peers when the cycle turns. That cannot be tested yet. 4. **Adverse allocation.** Charge all corporate costs (about 12.7 TTM EBITDA, roughly 6.8 after tax) to the acquired cohort, and value KBK at only its TTM months. The FY2024–TTM achieved multiple rises from 16.6× to about 18.8× (964 / 51.4). The conclusion survives: at today's trough earnings, recent deals do not cover the cost of capital. Only a recovery in trailer earnings makes the cohort economic. This study's evidence says the recent cohort is not yet earning them. The observation that would settle it is Compressed Gas segment EBITDA in FY2027 relative to its FY2025 level (98.7 IFRS basis). A return above about 80 would point to the base case; staying near 60–70 would point to the downside. # History, market & cycle The 2024 case rested on the management team's record of buying and improving industrial businesses. The source valuation starts with FY2023 economics, adds Highland and LV after restructuring, and estimates further capital deployment through FY2026. It is a forecast vintage, not an FY2026 result. The portfolio has since changed. Revenue grew from 678.4 (FY2023) to a TTM 1,731.2 at June 2026, and adjusted EBITDA from 121.6 to 312.2. Base-portfolio sales growth (issuer definition, excluding the year's acquisitions) was about +1% in FY2023, +4% in FY2024, +1% in FY2025 and +8% in 9M FY2026. The rest of the growth was bought. ## Demand drivers and cycle **Who pays and who decides.** - Propane and NH3 tanks: the payers are propane retailers and fertilizer distributors, who own the tanks at customers' sites. They decide on delivery time and price. - Data-centre generator fuel tanks: the developer pays, and the EPC or genset supplier specifies them. - Tank trailers: carriers of fuel, chemicals, food-grade and cryogenic products pay and decide, mainly on fleet age and freight economics. - Western Canadian services: exploration and production companies pay and decide, against their drilling and completion budgets. **Demand split (judgment, by economic unit from `01-business.md`).** - Growth and regulation-driven demand comes from data-centre backup fuel storage, c-store and agricultural fiberglass tanks (KBK), and fuel-system rules. - Discretionary or cyclical demand comes from trailer fleet additions and Western Canadian activity. **Cycle measured, not described.** - The current cycle is in trailers. Compressed Gas adjusted EBITDA was 17.3 in Q3 FY2026 against 29.7 a year earlier, on base-portfolio sales down 13% (Q3 MD&A pp.7–8). ACT's September 2026 trailer orders and Wabash's Q3 2026 results show an industry-wide downturn in new trailer demand, with tariff pull-forward clouding the timing (`act-trailers-october-2026`, `wabash-q3-2026`). - The offset is data centres. HVAC base-portfolio sales rose 61% in Q3 and 23% for nine months, and management expects the strength to continue near term (Q3 MD&A pp.5, 7). - Western Canadian services grew 11–17% excluding Colter in 9M FY2026. - The group's total base-portfolio growth of +8% therefore nets a trailer contraction against two booms. **Misperception and its test.** The market's view since June 2026 appears to be that the deal engine is impaired: by governance (the AMF allegations, `03-management.md`) and by an expensive cyclical deal (Entrans). The series that tests the second belief is Compressed Gas segment EBITDA through FY2027, read against ACT and FTR tank-trailer orders. A recovery to the FY2025 run rate (about 99 on an IFRS basis) with orders recovering would show Entrans was bought at a cyclical high rather than overpaid structurally. No series yet tests whether data-centre volume is durable. HVAC base-portfolio sales growth is the observation to watch. ## Competition Competition happens in three different places: The number of competitors is falling as family owners retire. TerraVest buys many of them (Advance, LBT, Tankcon, SPV and Jet Peinture in two years), which is both the acquisition runway and the consolidation trend. The advantage in one sentence: the lowest steel and labour cost per vessel among regional fabricators, from group-scale purchasing and shared plants, plus a buyer of last resort for retiring owners who want continuity rather than a top price. It is durable where TerraVest's volume is large relative to its suppliers (steel plate, valves) and weak where a single supplier controls a component (ECR's heat exchangers). The competitor who could hurt it: private equity for larger deals, and an integrated trailer maker cutting price in a downturn. The Entrans multiple (7.0× after-rent EBITDA, about 9.9× NOPAT) shows TerraVest now pays closer to auction prices when it competes for scale assets. ## Counterparty incentives - **Sellers (founders and families).** They want continuity of name, staff and location and a fair, certain price. The deal releases quote sellers wanting TerraVest to "grow the legacy" (`terravest-deal-releases`). TerraVest offers decentralization and keeps the brands. The sellers' outside option is private equity, which pays more but usually restructures and resells. The arrangement would change if TerraVest stopped honouring autonomy or if a governance scandal made it a less trusted buyer. The June 2026 allegations put that reputation at risk. SPV's owner still sold in July 2026, a month after they became public; B&R and Jet Peinture closed days before. One deal is too few to say the deal flow is unaffected. - **Earnout sellers.** Entrans's sellers had up to US$46m tied to twelve-month results, and the target was missed. An earnout transfers cyclical risk back to the seller only up to its size. Here it covered about 8% of the price. - **Steel suppliers.** Mills want volume and steady offtake. TerraVest concentrates purchasing to win discounts, so its dependence runs both ways. Where its share of a supplier's sales is small (ECR's foundry), the lever fails. - **Customers: propane retailers, carriers and data-centre EPCs.** Large propane retailers have buying power, but they value delivery time and local service. The data-centre EPCs are concentrated and project-based: a concentrated relationship that is currently a source of volume, and could leave. - **Lenders.** The syndicate holds a C$815m revolver plus term loans of 100 (March 2027) and 46 (March 2028) (Q3 MD&A pp.12–13). They want covenant compliance, and their leverage tolerance sets the deployment ceiling. TerraVest was in compliance in June 2026. - **Minority partners in GES.** About a third of the water-management unit's earnings go to them (minorities 12.2 TTM). Deals made through GES (LV, Aureus, Colter) share their returns. - **Regulators.** The Quebec AMF's investigation concerns the Executive Chairman's personal conduct, not the company's accounts (`amf-allegation-reports-2026`). The company says the AMF had not contacted it. The regulator wants deterrence. A charge would bear on board composition and on the company's reputation with sellers, not directly on cash generation. # Management & incentives Gilbert sources and negotiates deals, Haw leads diligence and the operating changes, and Pellerin is involved throughout; subsidiary presidents supply operating knowledge. An investment in TerraVest is largely a bet on how these few people allocate free cash. The deals are now far larger. And in June 2026 the Executive Chairman became the subject of a securities-regulator investigation. Evidence limits: the 2026 management information circular was not retrieved. Current holdings come from an aggregator transcription of filings (`insider-holdings-aggregator-2026`). Both are labeled. ## Ownership and incentives | Person | Role | Shares | Options | Cash pay | Open-market record | Background | | --- | --- | ---: | --- | --- | --- | --- | | Charles Pellerin | Executive Chairman since 2014 (CEO 2015–17) | 3,356,500 (15.5%), Feb 2026; 3.5m in Jan 2024 | 350k (Jan 2024) | about C$0.2m (2024 report); aggregators show about C$0.44m | Buyer: about C$9.8m net cash input, including 53k at C$16 (Jan 2021) and 10k at C$23 (Jul 2022) | Partner of a Quebec accounting firm; co-owned Granby before TerraVest bought it; Clarke director | | Dustin Haw | CEO since 2017; director since 2014 | 175,889 (0.8%), Jan 2026 | 500k (Jan 2024), likely most of the 2022 grant | averaged C$0.81m in 2021–23 | Small buyer at C$7–9 (2016–17) | PhD physics, CFA; ex-Clarke VP | | Mitch Gilbert | CIO since 2013 | 117k (Jan 2024) | 333k (Jan 2024) | about C$0.42m | Seller: 147k at C$16 (May 2020), 17k at C$27 (2023) | 14 years in bank M&A | | Mitch DeBelser | President, Compressed Gas | 405k (Jan 2024) | — | about C$0.58m | — | Came with NWP (2014) | | Pierre Fournier | President, HVAC | 79k (Jan 2024) | — | about C$0.5m | Small seller | Granby since 2006 | | Guillaume Cloutier | CFO, joined 2025 | n/a | n/a | n/a | n/a | Ex-CFO of an aerospace company | | Directors Cook, Laniuk, Rossi, MacBean | Board | Laniuk 2.1m (2024), others small | — | C$44–54k each (2024) | Laniuk and MacBean sellers | Cook chairs audit and sits on Clarke's board | Options outstanding at 30 June 2026: 616,500 from the February 2022 grant (strikes C$24.49–34.49) and 133,333 granted in February 2026 at C$142.61. That is 3.5% of shares (Q3 MD&A p.13). At C$107.85 the 2022 tranche is worth about C$49m of intrinsic value; the 2026 tranche is out of the money. **Whose interests dominate.** At 10% a year for ten years the share price goes from C$107.85 to about C$280. Pellerin's stake would gain about C$577m against cash pay under C$0.5m a year. Haw's shares and (if he still holds about 500k of the 2022 options) options would gain about C$115m against roughly C$1m a year of pay. Share ownership, not salary, drives both. Pellerin's stake is more than ten times Haw's, which makes the chairman's position, and its possible sale, the more important variable. The board "does not establish quantitative return on capital criteria for management" (Q3 MD&A p.13). There is no return-weighted metric to back out; alignment rests on ownership. ## Track record and references The peak was C$175.20 (`yahoo-tvk-trading-2026-11`). **Per-share earning power.** Cash NOPAT per share (baseline definition) was C$5.28 in FY2024 (102.996 / 19.50m shares) and C$6.88 in FY2025 (149.300 / 21.69m). It was about C$7.75 for the TTM to June 2026 (166.0 / 21.43m). That is +47% per share in under two years, while the share count rose 9.9% from September 2024 to June 2026. Per-share value has grown even after equity issuance. Including acquired goodwill and intangibles it is now about 11%, because recent deals are larger and priced nearer market multiples. **Decisions that failed or have not yet worked.** - Iowa Steel (2019): closed in FY2026. - Biogas jobs: ended. - Entrans (2025, 80% of recent deal capital): earns about half its purchase EBITDA and missed its earnout. It judged the bench shallow and middle management weaker (pp.28–29). No new references were collected for this study (no fieldwork claimed). Since then, the long-serving CFO has been replaced (2025) and the Executive Chairman is under investigation. Bench depth matters more than it did. **What trades reveal.** In May and June 2025, near the share-price peak, 266,500 options were exercised cashlessly and 100,000 options from the January 2020 grant were settled for C$15.7m cash at C$170.05 (FY2025 FS note 19.2). The holders are not named in the statements. The company issued 2.0m shares at C$160.30 the same month. Executives monetized near the high, and the company sold stock near the high. Both are price-aware, the second in shareholders' favor. ## Capital allocation principles 2. **Fund strategies, not projects.** Met. Deals build platforms: Highland, then Simplex; Entrans, then LBT, Tankcon and SPV; GES, then LV, Aureus and Colter. Each platform's buying and plant network makes the next bolt-on cheaper to restructure. 4. **No tolerance for bad growth.** Met at the unit level (closures and disposals above). Not yet testable for Entrans, which is cyclical rather than proven bad. **Governing objective.** The MD&A states the aim of maximizing returns to shareholders through opportunistic acquisitions. Behaviour (per-share earning power, issuing dear and buying cheaper, closures) is consistent with long-term value per share. The Entrans scale, and the leverage taken to fund it, show a higher risk appetite than the 2014–2022 record. ## Integrity and governance **The AMF investigation.** On 5 June 2026, Le Journal de Montréal reported that Quebec's securities regulator obtained warrants to search Executive Chairman Charles Pellerin's home and other locations (11 and 13 February 2026). The AMF suspects he passed non-public information about the Entrans acquisition, which he allegedly knew of by 12 December 2024, to family members and acquaintances. Nine connected people allegedly made about C$6.8m (`amf-allegation-reports-2026`). No charges had been filed in the reports located. The company's release said the board, excluding Mr Pellerin, is reviewing the allegations with legal advisers and will cooperate with regulators if approached (`terravest-pellerin-update-2026`). The shares fell about 30% that day. No outcome of the board review or change to his role was found by 3 November 2026. The 19 June 2026 Faguy authorization motion has now been read for its allegations, parties and requested relief (`terravest-faguy-authorization-2026-06-19`); it names TerraVest, Pellerin and the certifying officers. It is a claimant pleading, not a finding or quantified award. The current court docket, exhibits and any authorization decision remain unverified. These are allegations, not findings. Even so, they concern the person with the largest stake and the board's chair. The company's response so far has been minimal. **Other evidence.** - Negative: an insider-dominated board; two directors share a board with Clarke; the 2024 compensation committee was the whole board. The board sets no return criteria for management. **The three agency costs** (Mauboussin and Callahan pp.47–48): - *Control of resources:* low because pay is small. However, each larger deal raises the operational and governance load on a very small team. - *Risk tolerance:* a 15% holder might be expected to take less risk than diversified owners want. Yet the Entrans purchase took leverage above the stated ceiling, so concentration did not produce caution. - *Time horizon:* long, with ten-year options and large holdings. Buybacks in FY2026 (259,915 shares) exceeded share-based issuance (none in FY2026; 165,908 in FY2025). They are not merely offsetting compensation. **Agency cost and how the valuation treats it.** The material agency cost is governance and key-person risk, not pay. Several outcomes are possible: a charge, a forced change of chairman, a large block sale (3.36m shares is about 59 days of the 3-month median TSX volume), reputational damage with sellers, or litigation costs. None changes the cash economics of the plants. Several could slow deal flow or raise the cost of capital. The model includes no fine, settlement or block-sale discount, because none is quantifiable from public evidence. The risk is recorded as a monitored trigger in `06-risks.md` and `valuation/prospective-checks.json`; the scenario weights were not changed for it (see `11-scenario-probabilities.md`). Method: Mauboussin and Callahan, Capital Allocation (2025) pp.47–55, as cited in `prompts/stages/05-management.md`. # Capital allocation The value driver is the cash management can redeploy after keeping the existing businesses healthy, and what that cash buys. This note measures what the last five years of cash actually did. Money is C$ millions. Cash-flow statements were reviewed for FY2024 and FY2025 (`terravest-fy2025-fs`). For FY2021–FY2023 only the issuer's Cash Available for Distribution reconciliations were reviewed (`terravest-fy2022`, `terravest-fy2023`, `terravest-fy2024`), not the full statements, so those years lack investing and financing lines. 9M FY2026 totals come from the Q3 MD&A. ## Sources and uses of cash | | FY2021 | FY2022 | FY2023 | FY2024 | FY2025 | 9M FY2026 | | --- | ---: | ---: | ---: | ---: | ---: | ---: | | Operating cash before working capital (after interest and tax paid) | 46.9 | 68.3 | 96.6 | 138.8 | 173.6 | 154.4 | | Working capital (absorbed −) | −23.9 | −38.3 | −17.4 | 17.6 | −62.5 | 31.7 | | **Operating cash flow** | 23.1 | 29.9 | 79.2 | 156.5 | 111.1 | 186.0 | | Maintenance capex (issuer) | −5.8 | −8.0 | −14.9 | −26.1 | −26.8 | −34.4 | | Lease principal | −4.4 | −5.8 | −5.8 | −8.3 | −12.5 | −11.9 | | Gross PP&E capex incl. rental fleet | | | | −55.8 | −66.7 | | | PP&E disposal proceeds | | | | 31.0 | 11.7 | | | Acquisitions, net of cash acquired | | | | −149.1 | −857.6 | | | Businesses sold | | | | 4.3 | 3.0 | | | Financial investments, NCI buyouts, intangibles | | | | −5.7 | −45.1 | | | Net bank and term debt | | | | −45.8 | 574.7 | | | Shares issued (bought deals) | | | | 91.9 | 307.0 | | | Options settled in cash, net | | | | −4.6 | −26.0 | | | Buybacks | | | | 0 | 0 | −30.0 | | Dividends | −7.3 | −7.1 | −8.5 | −10.6 | −13.5 | −12.5 | | Net change in cash | | | | 3.8 | −14.0 | 5.5 | | Closing cash (after FX) | | | 24.6 | 28.4 | 13.9 | 22.0 | The cash balances reconcile: 24.6 + 3.8 + 0.0 FX = 28.4; 28.4 − 14.0 − 0.5 = 13.9; 13.9 + 5.5 + 2.5 = 22.0. - Operating cash before working capital and share compensation: 287. - Working capital: −61. - Net capex: −84, against 126 of D&A. - M&A: −117. Dividends: −37. - Debt: +81. Equity: −29. Other: −16. **What the table shows.** - Until FY2023, TerraVest funded small deals and a modest dividend from internal cash and some debt, and kept capex below depreciation. - FY2024–FY2025 changed the scale. 1,006.7 of acquisitions was 3.2× the 312.4 of operating cash generated before working capital. It was funded by 528.9 of net new debt and 398.9 of new equity. The share count rose from 17.91m to 21.69m. - In FY2026 the priority reversed: cash went to deleveraging, small bolt-ons and the first buybacks in years. - Dividends stayed small, 10–12% of Cash Available for Distribution. - Option settlements (30.5 over two years) were a real financing cost that the issuer's CAD measure does not show. ## Investment rates per sales change Window FY2023 → FY2025 (Δsales 692.8), the two years with full statements; named steps in `valuation/calculations.json`: | Rate | Numerator | Rate per Δsales | Step | | --- | ---: | ---: | --- | | Incremental M&A rate (acquisitions net of divestitures) | 999.5 | 144% | `incrementalMaRate` | | Fixed capital rate (net capex − PP&E depreciation) | 6.7 | 1.0% | `fixedCapitalRate` | | Fixed capital, gross (capex − disposals, before depreciation) | 79.7 | 11.5% | `netCapexFY2024To2025` | | Internally generated intangibles (no R&D disclosed; purchased intangibles) | 2.3 | 0.3% | `intangibleRate` | | Working capital, organic cash-flow basis | 44.9 | 6.5% | `organicWcRate` | | Working capital, balance sheet incl. acquired, FY2025 only | 137.1 on Δsales 459.3 | 29.9% | `balanceSheetWcRate` | TerraVest invests almost entirely through acquisitions. Each dollar of added sales cost about C$1.44 of purchase price, and organic investment barely exceeded replacement. That is the inflection, and it came in FY2025 with Entrans. The same team was in charge (Haw as CEO, Gilbert as CIO, Pellerin as Executive Chairman); the CFO changed in 2025. The balance-sheet working-capital rate (30%) overstates the run rate, because it holds Entrans's full year-end balances against 6.5 months of sales. On a cash basis organic working capital was 6.5% of the sales change, consistent with the 120–130 inventory days in `01-business.md`. FY2026 has released 31.7 so far, mainly customer deposits on data-centre work, which will reverse as projects ship. ## Reinvestment and returns **Owner earning power** (cash NOPAT, baseline definition): - FY2024: 103.0. FY2025: 149.3 (`cashNopatFY2024`, `cashNopatFY2025`). - TTM June 2026: 166.0. **Reinvestment rate, FY2024–FY2025.** - Growth investment: 999.5 of net acquisitions + 26.7 of growth capex beyond maintenance and disposals + 44.9 of organic working capital = 1,071.1. - Against 252.3 of matching cash NOPAT, the reinvestment rate is 425%. **Incremental return, lagged.** - Cash NOPAT rose 63.0 from FY2024 to TTM June 2026, against about 1,092 of capital deployed in FY2025 and H1 FY2026 (869.0 of FY2025 deals net of cash, business sales and the 48.3 earnout reversal; 132.9 in H1 FY2026; 28.1 of FY2025 growth capex; 62.5 of working capital; Q3 deals excluded). That is about 5.8%. - The matched deal-by-deal measurement (`research/07-assumption-register.md`, 3 November) finds 16.6× achieved (6.0%), or 12.05× (8.3%) after allowing for deals still in their restructuring ramp. - Both are below a 9–10% cost of capital. A rolling three-year lagged ROIIC is not yet meaningful: the capital is less than two years old and 80% of it is one cyclical deal. Judging Entrans fifteen months after closing would confuse a trailer trough with permanent capital inefficiency (Mauboussin and Callahan pp.46–47). **Identity check.** FY2025 NOPAT growth was 45% (103.0 → 149.3). FY2025 growth investment of about 945 equals 9.2× the prior year's NOPAT. Growth ≈ reinvestment rate × return implies about 4.9% earned within the year on new capital. That is consistent with Entrans contributing 6.5 months at trough-ish trailer earnings, so the definitions hold. Operating leverage was not the source: incremental cash NOPAT per sales dollar in FY2025 was 10.1% (`operatingLeverageFY2025`), below the 11.3% FY2024 margin. The acquired businesses diluted margins. Durable capital efficiency, not margin, is the open question. **Mix and returns by use.** - Acquisitions: 93% of growth investment, earning 6–8% so far (above). - Organic growth capex (fibreglass capacity, rental fleet) and working capital: about 7%, earning on the order of 13.1 of extra EBITDA at constant margin from FY2024 to TTM. Organic EBITDA growth is a margin-held estimate; unit-level capex returns are not disclosed. - Payouts: dividends about 2% of uses; FY2026 buybacks at C$115 average. ## Acquisition record | Date | Business | Stake | Price (C$m) | Multiple and later result | | --- | --- | ---: | ---: | --- | | Oct 2023 | LV Energy (via GES) | 67% | 25.0 | Undisclosed; sister company resold for 4.3 | | Apr 2024 | Advance Engineered Products | 100% | 33.6 | Undisclosed; tank trailers | | Jan 2025 | Aureus (via GES), assets | 67% | in FY2025 total | Hot-oiling line resold for 3.0 | | Mar 2025 | Entrans | 100% | 779.6 cash (+ up to US$46m earnout, reversed) | About 7.0× after-rent TTM EBITDA (about 9.9× NOPAT); now about half its purchase EBITDA; missed its earnout | | Apr–May 2025 | LBT (EV ~US$16m), Simplex (EV ~US$28m), Tankcon (EV ~C$27.8m) | 100% | in FY2025 total 960.5 | Simplex feeds the data-centre platform; LBT and Tankcon add to trailers | | Sep 2025 | Wave (via GES) | | small | Integrated | | Jan 2026 | KBK Industries | 100% | 125.2 | About 5.6× after-rent TTM EBITDA | | H1 FY2026 | Three small deals, one 60% owned | | 9.1 | | | May–Jul 2026 | Colter (via GES; bargain-purchase gain about 5.0), B&R, Jet Peinture, SPV (23.4) | | Colter/B&R/Jet undisclosed | SPV debt-funded; outside the June TTM | **Aggregate.** Over FY2024 to TTM June 2026 (matched to ownership months), capital of 964.2 earned 58.2 of TTM NOPAT: 16.6× achieved, 12.05× after allowing for the restructuring ramp (`achievedMultiple`, `matureMultiple`). - Margin bounds: acquired EBITDA is the change in group EBITDA less organic growth at the FY2024 margin. The base portfolio has roughly 975 of sales at the window's end, so one margin point is about 9.8 of EBITDA. If its margin was one point higher at the end (plausible with data-centre volume), the achieved multiple is about 18.2×; one point lower (the base trailer business) gives about 15.2×. **Deployment pace.** Net acquisition capital was 158.8 in FY2024, 869.0 in FY2025 (with a C$307m equity raise) and 132.9 in H1 FY2026, plus SPV's 23.4 and undisclosed Q3 deals. The model deploys nothing in year 1 while debt falls to the ceiling, then 110–167 a year in years 2–5 (base). That is within the observed non-equity pace and implies no new equity. ## Current funding position Starting net bank debt is 860.3 (June 2026 bank debt less cash of 836.9, plus the July SPV purchase of 23.4). At TTM EBITDA of 312.2 the 2.4× ceiling allows 749.2, so TerraVest needs to delever before debt-funded deployment resumes; the base makes no new acquisition in year 1. Interest, dividends and the carried earnout and vendor claims (8.6) are paid before the acquisition budget. New deals start at 63.08% of mature earnings and ramp to 100% over three years, and future deal EBITDA does not raise the borrowing ceiling. The pension deficit (4.3) is a terminal claim. Fair-value investments (56.5, March 2026) are credited once. Term loans of 100 (March 2027) and 46.4 (March 2028) fall due inside the forecast. In June 2026, C$120.1m of the revolver was undrawn. Capital-efficiency diagnostics: `research/10-capital-efficiency.md`. Revenue, operating cash and FY2024–FY2025 deal cash reconcile to our facts. - FY2014–FY2024 (eleven years, from the first deal year to the year before Entrans): mean 0.88×, median 0.92×. Five years exceeded 1.0× (FY2017 1.43, FY2018 2.05, FY2021 1.67, FY2022 1.26), funded by debt. Including FY2012–FY2013, which had no deal cash, the mean is 0.75×. - Quiet years (under a quarter of the mean) were FY2016 (0.06×) and FY2020 (0.19×), never consecutive; FY2023 was close at 0.23×. FY2025 (Entrans, 7.72×) was funded by debt and equity and sits outside the budget the model can deploy (stage 6 perimeter rule). Comparison with the model: the base deploys 0, then 0.64× rising to 0.85× of cash earnings, and 0.81–0.95× of free cash flow after interest, up to the 2.4× ceiling on existing-business EBITDA. That is within the record (0.75–0.88× mean, 0.92× median) on the nearest matching definitions. The binding constraint is the leverage ceiling, not deal supply. Year one is quiet because leverage is above the ceiling, which is evidenced timing. The downside's four-year dry spell exceeds the record's longest quiet run of one year. The base already sits at the expected budget, so no input changed. If future acquired EBITDA were allowed to raise borrowing capacity, the base would exceed the record, and it would then need a yearly budget path stating the expected deal supply. # Thesis and valuation Current assessment:5November2026, after source reconciliation of the revenue-to-owner-cash bridge. TerraVest buys niche industrial manufacturers, improves their operations and reallocates cash into further purchases. Tank-trailer recovery can offset data-centre/heating/energy cyclicality, so the accepted source framework uses zero net organic sales growth. Profitable growth comes from funded acquisition sales and their explicit cash margins, with debt capacity tied to the existing business. ## Current valuation Saved quote CAD 106.84, market time 2026-11-04T21:00:00.000Z; fixed valuation date 5 November 2026 and original model exits. These are forward estimates. | Horizon | Bear | Base | Bull | Weighted cash-flow IRR | | --- | ---: | ---: | ---: | ---: | | 5 years | -11.20% | 7.98% | 12.24% | 4.66% | | 10 years | -3.03% | 10.19% | 15.65% | 9.09% | Reported trailing revenue of CAD 1,731.218m now drives annual NOPAT and EBITDA margin paths. The prior earnings-factor forecast was first reproduced mechanically before updating estimates. Highland’s 11 owned months and KBK’s issuer nine-month pro-forma sales supply a labelled annual-sales proxy; purchase capital includes transaction fees. KBK adds only CAD 46.416m sales missing from the trailing baseline, with seller-derived cash still recovering over three years. SPV sales remain a separately labelled small-deal price/sales estimate. ## Pre-mortem Highland flat-month annualization and KBK pro-forma annualization are estimates with seasonality limits. SPV sales are not reported by the issuer and are estimated from its funded price and matched small-business economics. Margin recovery, target-specific maintenance/rent, bank refinancing, governance and limited small-deal supply remain material uncertainties. The 2.4× modeled borrowing ceiling is not an actual loan covenant. The central estimate can fail even when a useful product or good acquisition history survives. Customer demand, supplier pricing, the capital needed for delivery and ownership/financing claims must all work together. The adverse case is a representative failure mechanism, not the largest possible loss. No new report or positive share-price move alone resolves these risks. ## Exit multiple Current17x central/upside and 13× adverse EV/cash-NOPAT are separate current judgments, with current bank debt, claims, non-operating assets and option proceeds accounted once. The revenue bridge does not supply independent proof of terminal value or remove the survival risks identified in the written debate. Shorter duration, poorer cash conversion and higher required returns remain relevant stresses. Historical source reproduction and current underwriting remain distinct. ## Bear case Highland flat-month annualization and KBK pro-forma annualization are estimates with seasonality limits. SPV sales are not reported by the issuer and are estimated from its funded price and matched small-business economics. Margin recovery, target-specific maintenance/rent, bank refinancing, governance and limited small-deal supply remain material uncertainties. The 2.4× modeled borrowing ceiling is not an actual loan covenant. The existing adverse operating and capital-allocation paths remain the coherent modeled stress. Matching sales to cash prevents a missing denominator from hiding those assumptions; it does not turn a proxy margin or assumed refinancing into a reported fact. Debt, deferred claims, operating liquidity and distributions remain funded in the same cash ledger. ## Variant view The market's belief appears to be: a proven serial acquirer, temporarily discounted for a chairman's personal legal problem and a trailer downturn. This study broadly agrees, with one difference. One large, peak-cycle purchase (Entrans, 80% of recent capital) is earning 4–9% and drags the measured cohort to 12–17× NOPAT (`04-capital-allocation.md`). At about 19× cash NOPAT, today's price capitalizes TerraVest as a compounder that keeps buying small deals well. On that base the expected return is about 8–10% a year, near the hurdle. **No mispricing in our favor:** the June fall removed a premium rather than creating a clear discount. Upside needs restructuring on top (12.0% / 15.5%) or an Entrans recovery, which is not modeled and would add to existing earnings. A repeat of a large peak-cycle deal, or a governance shock that stops deal flow, would move the outcome toward the downside. There is no hard catalyst. The source bridge correction changes the central estimate by only a small amount at the same price/date. It does not establish a new market mispricing or justify an unsupported advocate sensitivity. The [independent claim judgment](12-debate.md) still sets the evidence burden; the current numerical returns are the table above. ## Why the opportunity exists The following observations retain the measurement dates in their cited sources; no new holder, liquidity or short-interest survey is claimed. Neglect is no longer the explanation: - Market capitalization is about C$2.31bn (21.43m shares at C$107.85). Three-month median TSX volume is about 56,900 shares, or C$6.5m a day (`yahoo-tvk-trading-2026-11`). - Short interest and retail ownership were not checked (gap). The sellers had specific reasons: - 5 June 2026: the AMF allegations, with the stock down about 30% in a day on about 14× normal volume. - A visible trailer downturn: Compressed Gas Q3 EBITDA −42%. - The Entrans earnout reversal, showing the big deal missed its own targets. The discount is largely justified by business evidence, not just by fear. A buyer today pays about 19× for a company whose marginal capital earns about 6–8%. ## What would prove us wrong The next FY2026 issuer report and subsequent debt disclosures: actual KBK/SPV revenue and cash contributions, Entrans margins, maintenance/leases, deposits, all-in prices and 2027/2028 refinancing requirements. The next results date was not confirmed in checked evidence. Compare future observations with the same ownership, cost and reporting periods used in this model. Preserve earlier frozen checks and score them as their reports arrive; do not edit their forecasts to match the result. Unavailable deal disclosure needs a bounded estimate and a revision trigger, not a claim of observed target profitability. [Source bridge and same-price/date changes](../updates/2026-11-05-revenue-bridge-closure.md). The earlier initial completion audit’s three revenue gaps are now closed with reported observations and explicit estimates; the remaining limits above are ordinary underwriting uncertainty, not missing forecast outputs. # Risks & disconfirming evidence ## Paths to permanent loss 1. **Reinvesting below the cost of capital.** This is the main one. The FY2024–TTM deal cohort earns 12–17× NOPAT (6–8% after tax) against a 9–10% hurdle (`04-capital-allocation.md`). If future deals price like Entrans and earnings stay at today's cycle point, every dollar redeployed destroys value while reported EBITDA grows. This is a value trap, and it would not show in issuer Cash Available for Distribution or adjusted EBITDA growth. 3. **Governance and key person.** The Executive Chairman (15.5% holder) is under AMF investigation for alleged tipping before the Entrans acquisition. Possible outcomes are a charge, his removal, a block sale (about 59 days of median volume), reduced trust among founder-sellers, and litigation (the 19 June 2026 Faguy authorization motion names the issuer and officers; its allegations and requested relief are now read, but no later court decision or quantified issuer liability is established). None is in the model. 4. **Structural decline in legacy lines.** Electrification pressure on propane and heating oil, and cast-iron boilers losing out to high-efficiency appliances, could shrink the replacement core faster than its low single-digit drift. 5. **Data-centre concentration.** About 22 of annualized EBITDA comes from recent project volume with a few concentrated EPC customers. If it fades, that is a give-back of about 7% of EBITDA. Customer deposits (part of the 31.7 working-capital release in 9M FY2026) would reverse at the same time. ## Unverified inputs (largest first) - **Base deal multiple, 7.40×.** It rests on two issuer-stated small-deal multiples (Highland 4.0×, rent treatment unstated; KBK 5.6× after rent). It assumes future deals are bolt-ons whose earnings hold at the sellers' trailing level. A return to Entrans-scale or Entrans-priced deals would invalidate it (9.87–12.05×). - **Exit multiple, 17×.** It sits inside the Koller range for small-deal base economics (15.6–21.0×) and at the stock's own recent low (`05-thesis.md`). At 14× the five-year base falls to 2.8% (weighted 1.7%). - **Segment economics.** The split of HVAC between tanks and heating, and of Compressed Gas between trailers and stationary tanks, is estimated, not disclosed. - **Claims and assets.** Carried March 2026 contingent claims (8.6), pension (4.3) and financial investments (56.5). - **KBK and SPV cohort NOPAT** (7.14 at KBK's own multiple; 3.16 at the small-deal multiple). These are estimates, not issuer disclosures. - **Lease interest** (17.6, Q2 × 4) and the 25% normalized tax rate. - **Annual history.** Full FY2021–FY2023 cash-flow statements were not retrieved, so five-year sources and uses are partial for those years. ## Disconfirming evidence to watch | Evidence | Source and timing | What it would mean | | --- | --- | --- | | Compressed Gas adjusted EBITDA staying below about 70 a year through FY2027 | Quarterly MD&A | Trough is structural; the downside deal multiple becomes the base | | A new large deal at ≥ 7× after-rent EBITDA, funded by debt above 2.5× | Acquisition releases, MD&A | Reinvestment risk rising; value trap path | | Equity issued below C$100 | Issuer releases | Permanent dilution in a downturn | | AMF charges or the board review's outcome; chairman change | AMF, issuer releases | Governance path; possible deal-flow and cost-of-capital effects | | HVAC base-portfolio sales falling year over year | Quarterly MD&A | Data-centre fade | | Maintenance capex above about 50 a year | CAD reconciliation | Earning power overstated | Earlier risk notes remain valid. Larger or more competitive deals can raise purchase prices. Restructuring can take longer or yield less. Inventory and customer deposits move reported cash without changing durable earning power. Issuer CAD is after financing costs and cannot be put into an EV/NOPAT multiple. Accounting gains (the 48.3 earnout reversal, bargain purchases) are not recurring cash. ## Surviving debate attacks — 5 November 2026 Low; small-deal economics conditional, refinancing/governance unresolved and revenue-to-owner-cash forecast incomplete. Reconstructed the independent isolated stress at the same quote/date/horizons; do not treat stress size as a probability or a measured current fact. Reconstructed the independent isolated stress at the same quote/date/horizons; do not treat stress size as a probability or a measured current fact. Reconstructed the independent isolated stress at the same quote/date/horizons; do not treat stress size as a probability or a measured current fact. Discriminating evidence: Next FY2026 report/debt maturity disclosures: Entrans earnings, maintenance capital and deposits, all-in small deal prices, March2027/2028 refinancing. The [complete debate](12-debate.md) sizes these risks using the shared forecast; a sensitivity is neither a measured fact nor a calibrated probability. ## Revenue bridge closure and remaining uncertainty — 5 November 2026 The previously missing sales/cash perimeter is now modeled. Highland flat-month annualization and KBK pro-forma annualization are estimates with seasonality limits. SPV sales are not reported by the issuer and are estimated from its funded price and matched small-business economics. Margin recovery, target-specific maintenance/rent, bank refinancing, governance and limited small-deal supply remain material uncertainties. The 2.4× modeled borrowing ceiling is not an actual loan covenant. The next FY2026 issuer report and subsequent debt disclosures: actual KBK/SPV revenue and cash contributions, Entrans margins, maintenance/leases, deposits, all-in prices and 2027/2028 refinancing requirements. The next results date was not confirmed in checked evidence. This closes the missing forecast output rather than claiming that the target economics or refinancing outcomes are certain. Earlier source-gap statements in the dated debate describe its initial information set; current inputs are in 07-assumption-register.md. # Current assumption register Assessment:5November2026. Reported facts, issuer pro-forma estimates, normalization and forward judgments remain separately classified. ## Reported starting bridge Reported trailing revenue of CAD 1,731.218m now drives annual NOPAT and EBITDA margin paths. The prior earnings-factor forecast was first reproduced mechanically before updating estimates. Highland’s 11 owned months and KBK’s issuer nine-month pro-forma sales supply a labelled annual-sales proxy; purchase capital includes transaction fees. KBK adds only CAD 46.416m sales missing from the trailing baseline, with seller-derived cash still recovering over three years. SPV sales remain a separately labelled small-deal price/sales estimate. | Input | Value | Classification | Definition/evidence | | --- | ---: | --- | --- | | revenue | 1731.218 | derived reported | FY2025 plus9M2026 less9M2025:1371.151+1311.809−951.742=1731.218 CADm. Acquired sales already consolidated remain inside this baseline; only missing contribution is added. Sources: terravest-fy2025, terravest-q3-2026 | ## Known and committed cohort bridge Revenue is consolidated gross; cash is on the model’s common-owner or cash-NOPAT perimeter. A reported sales observation is not a reported cash margin. Missing months exclude sales already in the starting baseline. Capital already paid or reserved is not purchased again. | Cohort | Missing annual sales | Missing/mature cash | Timing and basis | | --- | ---: | ---: | --- | ## Forward scenario mechanisms TerraVest buys niche industrial manufacturers, improves their operations and reallocates cash into further purchases. Tank-trailer recovery can offset data-centre/heating/energy cyclicality, so the accepted source framework uses zero net organic sales growth. Profitable growth comes from funded acquisition sales and their explicit cash margins, with debt capacity tied to the existing business. Current17x central/upside and 13× adverse EV/cash-NOPAT are separate current judgments, with current bank debt, claims, non-operating assets and option proceeds accounted once. ### base Pay interest, dividends and identified vendor/earnout claims before deploying capacity. Existing-business net debt capacity remains 2.4× EBITDA; new deals are small (the deployment budget cannot buy another Entrans) and are priced at the sellers’ trailing earnings on Highland and KBK’s issuer multiples (7.40× mature NOPAT), reaching that level over the three-year ramp with no restructuring uplift. Value terminal cash NOPAT at the original 17× EV multiple. Revenue and cash margins are now explicit yearly paths with matched purchased sales; current KBK missing months and all-in costs replace the old half-year shortcut. | Driver | Value/path | Reason and sources | | --- | --- | --- | ### downside Borrowing capacity falls with earnings, stopping deals while debt exceeds the ceiling. The original 13× EV/NOPAT multiple prices the reduced earning power. Revenue and cash margins are now explicit yearly paths with matched purchased sales; current KBK missing months and all-in costs replace the old half-year shortcut. | Driver | Value/path | Reason and sources | | --- | --- | --- | ### upside Keep zero organic growth, the three-year ramp, the 2.4× existing-business debt ceiling and the 17× exit. This isolates better purchase economics; it adds no leverage, free growth or rerating. Revenue and cash margins are now explicit yearly paths with matched purchased sales; current KBK missing months and all-in costs replace the old half-year shortcut. | Driver | Value/path | Reason and sources | | --- | --- | --- | ## Estimation limits and revision triggers Highland flat-month annualization and KBK pro-forma annualization are estimates with seasonality limits. SPV sales are not reported by the issuer and are estimated from its funded price and matched small-business economics. Margin recovery, target-specific maintenance/rent, bank refinancing, governance and limited small-deal supply remain material uncertainties. The 2.4× modeled borrowing ceiling is not an actual loan covenant. The next FY2026 issuer report and subsequent debt disclosures: actual KBK/SPV revenue and cash contributions, Entrans margins, maintenance/leases, deposits, all-in prices and 2027/2028 refinancing requirements. The next results date was not confirmed in checked evidence. Named arithmetic, stable fact IDs and source links are retained in valuation/calculations.json. Inputs were selected before inspecting the revised IRR; the earlier source result and prior models are not overwritten. # Underwriting plan Written from the evidence in `01-business.md` through `04-capital-allocation.md`. The review moved the base and upside deal multiples to the deal size the model can actually deploy into. Money is C$ millions; the forecast vintage starts 26 October 2026. ## Opportunity ledger | Opportunity | Starting point | Customer need and why TerraVest wins | Reachable scale in five years | Investment and cash cost | Timing and limits | Evidence against | | --- | --- | --- | --- | --- | --- | --- | | Tank-trailer **recovery** (Entrans, LBT, Tankcon, Advance, SPV) | Acquired trailer EBITDA about 56 TTM against about 111 at purchase; Compressed Gas Q3 FY2026 17.3 vs 29.7 | Carriers must replace aging fuel, chemical and food-grade fleets; Entrans brands lead in tank trailers | Recovery to purchase-level EBITDA adds about 55 EBITDA, about 29 NOPAT (+18% on the baseline) | None beyond working capital as volume returns (about 6.5% of the sales change) | Trailer cycles have run two to four years; ACT shows depressed September 2026 orders | Tariff pull-forward may have borrowed demand; the earnout miss shows the sellers' own targets failed | | Data-centre fuel storage and fuel systems (Highland, Simplex, KBK, other plants) | HVAC base-portfolio sales +61% in Q3 and +23% for 9M FY2026; about 22 of annualized EBITDA from the increase | Backup generators need on-site fuel tanks; capacity and steel-buying scale let TerraVest deliver large projects | Holding today's level is the base; growth would need new fibreglass and steel capacity | Q3 FY2026 growth capex 9.4, mainly fibreglass capacity; customer deposits fund part of the working capital | Project-driven; management expects strength "in the near term" only | Hyperscale capex cycles; concentrated EPC customers; deposits reverse as projects ship | | Heating appliances (ECR, Granby) | About 10–14% of profit | Installer loyalty; replacement in older homes | Declining low single digits a year | Minimal | Ongoing | Electrification; single-source heat exchangers | | Western Canadian services (GES/Fraction, Colter) | Service 9M FY2026 EBITDA 52.8, +12%; one-third minority-owned | Water management and flowback for large producers; integration of LV, Aureus and Colter | Share gains possible; activity follows oil and gas prices | Service capex near its D&A; maintenance rose to 14.4 in Q3 FY2026 | Seasonal (first half), cyclical | Commodity-driven activity; minorities take a third | | **Acquisitions** | Observed net capital 158.8 (FY2024), 869.0 (FY2025, equity-funded), 132.9 (H1 FY2026) plus SPV 23.4 | Retiring founders want continuity; TerraVest's purchasing and plant network raise acquired margins | Base deploys about 559 over five years (0, 110, 133, 149, 167) at 7.40× → about 76 of mature NOPAT | Internally generated cash after interest, dividends and claims, with net debt pinned at 2.4× existing EBITDA (749); no new equity | Year 1 deleverages from 2.76×; term loans of 100 (Mar 2027) and 46 (Mar 2028) | Recent cohort earns 12–17×; governance reputation risk with sellers; private equity competition on larger targets | | Buybacks (ASPP from 19 Oct 2026) | 30.0 spent at C$115 average in 9M FY2026 | — | Not modeled; would compete with deleveraging and deals | Cash at the share price | Discretionary | Accretive only below value; the base values the stock near today's price | **Overlap.** The base holds the TTM earning power flat (zero organic growth). That implicitly nets the trailer recovery (+29 NOPAT if it comes) against data-centre fade, heating decline and Western Canadian cyclicality, without assuming any of them. Bought earnings enter only through the acquisition engine at the stated multiple, and known cohorts (KBK's missing half-year, SPV) are added once. The TTM is not obviously inflated. Q3, usually the weakest quarter, produced the highest adjusted EBITDA of the last eight (87.0), though with Compressed Gas at its lowest. **Absolute scale, base, year 5 (October 2031).** NOPAT 218.8 against 166.0 today. Deals add about 76 of mature NOPAT on about 559 of capital, partly still ramping at exit. At the group's 0.53 NOPAT/EBITDA conversion and 18% EBITDA margin, that is about 140 of EBITDA and about 790 of revenue. Group revenue would be about 2,500, almost half above today, with no organic growth. Deployment of 110–170 a year is roughly one KBK or Highland-sized deal plus bolt-ons each year. That matches the FY2024 pace without equity, and is the deal size the 7.40× multiple was measured on. Net debt stays at 749 and leverage falls as EBITDA grows. Nothing here needs more staff at head office than TerraVest has used before, but a deal the size of Entrans is outside the base. The base's 7.40× (13.5% mature yield, no restructuring) sits just below that expectation, near the 2014–2022 record (7.2×). ## Scenario mechanisms | | Downside | Base | Upside | | --- | --- | --- | --- | | Existing earning power | Halves in year 1 (83.7), recovers to 75% by year 3 (131.0) and stays there | Flat at TTM 166.0 plus KBK/SPV ramps | Flat, as base | | Organic growth | 0 | 0 | 0 | | Deal ramp | 63% → 100% over three years | same | same | | Deployment | Stops: debt above the falling ceiling; 15.8 by year 5 | 0 in year 1, then 110–167 a year | 0 in year 1, then 110–195 a year | | Leverage ceiling | 2.4× of a lower EBITDA | 2.4× | 2.4× | | Financing and dilution | No equity; existing options only | same | same | | Exit (EV / cash NOPAT) | 13× | 17× | 17× | | 5y / 10y XIRR at C$107.85 (3 Nov 2026) | −11.4% / −3.1% | 7.7% / 10.1% | 12.0% / 15.5% | **Mechanisms.** - **Base.** Today's earning power holds, with cyclical gains and losses netting to zero. TerraVest keeps buying bolt-ons at the prices it paid for Highland and KBK, and bought earnings hold at the sellers' trailing level with nothing from restructuring. The base needs no trailer recovery. Entrans's own recovery would be upside to existing earnings and is not modeled. Another Entrans-scale, peak-cycle purchase would break the base (see `05-thesis.md`, what would prove us wrong). **Funding.** Every deal is funded from cash after interest at 6%, dividends of C$0.80 a share and the carried claims, before any contribution, with a one-year delay. Debt never exceeds the ceiling, so the engine cannot repeat FY2025's equity-funded jump. The model omits the cost of refinancing the 2027 and 2028 term loans; it assumes they roll at 6%. **Adapter.** `capital-acquirer`: debt-capacity-limited redeployment at a stated mature multiple, with a restructuring ramp and an EV/NOPAT exit. Its cost is that organic growth is fixed at zero and buybacks are not modeled. Each input's classification, sources and reasoning are in the `inputNotes` of `valuation/model.json`. ## Revenue bridge closure — 5 November 2026 Reported trailing revenue of CAD 1,731.218m now drives annual NOPAT and EBITDA margin paths. The prior earnings-factor forecast was first reproduced mechanically before updating estimates. Highland’s 11 owned months and KBK’s issuer nine-month pro-forma sales supply a labelled annual-sales proxy; purchase capital includes transaction fees. KBK adds only CAD 46.416m sales missing from the trailing baseline, with seller-derived cash still recovering over three years. SPV sales remain a separately labelled small-deal price/sales estimate. 5-year base revenue 2250.119, operating owner cash 219.198 (CADm). 10-year base revenue 3474.947, operating owner cash 356.554 (CADm). The original customer/opportunity and acquisition-budget reasoning remains; this correction does not infer new demand from a matched denominator. Highland flat-month annualization and KBK pro-forma annualization are estimates with seasonality limits. SPV sales are not reported by the issuer and are estimated from its funded price and matched small-business economics. Margin recovery, target-specific maintenance/rent, bank refinancing, governance and limited small-deal supply remain material uncertainties. The 2.4× modeled borrowing ceiling is not an actual loan covenant. [Source arithmetic and old/new inputs](../updates/2026-11-05-revenue-bridge-closure.json). # Capital productivity and reinvestment Assessment:5November2026. These are shared-engine diagnostics from the same owner-cash projection shown in the dashboard, not a new historical measurement. Money is in CAD millions. | Horizon | Scenario | Return/proxy definition | Rate | Period-weighted reinvestment | New growth capital | Matching earnings | Funding residual | | --- | --- | --- | ---: | ---: | ---: | ---: | ---: | | 5 years | downside | Acquisition ROIIC · mature cash proxy | 6.22% | 2.80% | 16.407 | 586.809 | 1.71e-13 | | 5 years | base | Acquisition ROIIC · mature cash proxy | 13.53% | 59.30% | 560.118 | 944.485 | 1.14e-13 | | 5 years | upside | Acquisition ROIIC · mature cash proxy | 21.44% | 61.34% | 609.774 | 994.141 | 1.14e-13 | | 10 years | downside | Acquisition ROIIC · mature cash proxy | 6.22% | 39.45% | 513.122 | 1300.603 | 2.27e-13 | | 10 years | base | Acquisition ROIIC · mature cash proxy | 13.53% | 73.40% | 1776.236 | 2419.822 | 0.00e+0 | | 10 years | upside | Acquisition ROIIC · mature cash proxy | 21.44% | 78.44% | 2341.271 | 2984.857 | 0.00e+0 | Reported trailing revenue of CAD 1,731.218m now drives annual NOPAT and EBITDA margin paths. The prior earnings-factor forecast was first reproduced mechanically before updating estimates. Highland’s 11 owned months and KBK’s issuer nine-month pro-forma sales supply a labelled annual-sales proxy; purchase capital includes transaction fees. KBK adds only CAD 46.416m sales missing from the trailing baseline, with seller-derived cash still recovering over three years. SPV sales remain a separately labelled small-deal price/sales estimate. Reinvestment is total new growth investment divided by total matching earnings, not an average of annual percentages. Previously funded cohort contributions are not new investment returns. Acquisition cash-yield proxies and forecast-implied productivity are not measured historical ROIIC or project IRRs. All dividends, existing claims, liquidity and financing are reconciled before new purchase capital. Highland flat-month annualization and KBK pro-forma annualization are estimates with seasonality limits. SPV sales are not reported by the issuer and are estimated from its funded price and matched small-business economics. Margin recovery, target-specific maintenance/rent, bank refinancing, governance and limited small-deal supply remain material uncertainties. The 2.4× modeled borrowing ceiling is not an actual loan covenant. Historical measurements and definitions remain in 04-capital-allocation.md and valuation/calculations.json. [Dated source bridge and funding ledger](../updates/2026-11-05-revenue-bridge-closure.json). # Scenario probabilities Assessment date: 2026-10-27. These are initial subjective judgments, with low confidence. Financial sources support the mechanisms, not measured numerical odds. The original authors did not provide or endorse these weights. | Horizon | Downside | Base | Upside | | --- | ---: | ---: | ---: | | 5 years | 30% | 50% | 20% | | 10 years | 35% | 40% | 25% | 5-year reasoning: The acquisition/restructuring framework and cash generation support the central path, while opening leverage above the stated acquisition ceiling leaves meaningful cyclical and financing sensitivity. A recovery of acquired earnings to purchase-date levels is possible but not yet evidenced. 10-year reasoning: Longer industrial cycles and a growing cumulative purchase budget widen the distribution. Give less weight to uninterrupted central execution, without assuming rising leverage or a larger exit multiple. Downside: A cyclical earnings shock contracts both cash generation and debt capacity. The accepted source stress is a representative adverse path, not a 30% issuer default probability or a maximum loss. Base: Normalized operating cash, deleveraging and small bolt-on purchases at the sellers’ trailing earnings (7.40×), with no restructuring uplift, support base. Zero further organic growth and fixed existing-business debt capacity limit reliance on free expansion. Review triggers: - Normalized segment cash earnings and industrial demand. - Net debt versus the existing-business debt ceiling. - All-in deal cost, restructuring spending and realized cohort NOPAT. Each path represents a broad outcome region, not the probability of hitting an exact share price or IRR. Paths are a coarse approximation of a continuous joint distribution; the displayed downside and upside are not bounds. Joint driver dependencies must be considered, rather than multiplying marginal driver guesses. Scenario weights sum to 100% within this approximation. Severe outcomes remain possible outside the modeled paths. Price movement alone does not change these weights. A changed causal premise, acquisition capacity, financial report or scenario definition does. Local assumption sensitivities suspend the researched weights until the full distribution is reassessed. No weighted-average IRR is presented as an expected shareholder return. Method: [Damodaran, chapter 33](https://pages.stern.nyu.edu/~adamodar/pdfiles/val3ed/c33.pdf), printed pp.894-897. See `valuation/model.json` for machine-readable weights and `docs/method/cash-flow-and-reinvestment.md` for the capital definitions. ## 3 November 2026: scenario definitions after deal re-underwriting The paths keep their causal roles, but the purchase economics are redefined. Base buys at 9.87×, the issuer's purchase prices on the sellers' trailing EBITDA, with acquired earnings recovering to that level and no restructuring uplift. Downside buys at the measured FY2024–TTM June 2026 mature-equivalent 12.05×, which embeds today's trailer trough. Weights are unchanged at 30/50/20 (five years) and 35/40/25 (ten years). The new evidence lowers the purchase yield in every path. It does not show that the cyclical downside or a trailer recovery is more or less likely; the Q3 Compressed Gas decline is already in the TTM baseline. Calculation and sources: `07-assumption-register.md` and `valuation/calculations.json`. ## 3 November 2026: pipeline review of the weights Reviewed after completing the eleven-stage research pipeline, including the governance evidence that the 27 October assessment did not record. On 5 June 2026 the Executive Chairman became the subject of a reported AMF investigation (`03-management.md`). Weights are unchanged at 30/50/20 (five years) and 35/40/25 (ten years). Reasons: - The allegation concerns personal trading, not the company's cash, accounts or plants. Its plausible effects (slower deal flow, a block sale, a higher cost of capital, litigation cost) act through deployment and valuation more than through the operating path that separates the three scenarios. The downside already stops deployment. - The new pipeline evidence points both ways. The trailer trough and the below-hurdle cohort returns are already priced into the downside and base deal multiples. Q3 FY2026 produced the highest adjusted EBITDA of the last eight quarters, and Service and data-centre volume are growing. Nothing in the new work is a causal change to the probability of a joint industrial and energy recession. - The exit-multiple finding (base 17× is at the top of what base returns justify) is a valuation sensitivity inside the base, not evidence for a different path. It is reported separately (`05-thesis.md`), not folded into the weights. Review triggers added: an AMF charge or change of chairman; Compressed Gas adjusted EBITDA below about 70 for FY2027; a refinancing of the March 2027 term loan materially above 6%. ## 3 November 2026: small-deal review (scenario definitions changed, weights reassessed) The causal roles are unchanged: purchase economics only, with the same operating paths, leverage, exit multiples and zero organic growth. Weights stay at 30/50/20 and 35/40/25: - ECR, and the absence of disclosed post-deal results for KBK, keep it below the base. - The downside's cyclical shock is unaffected by deal size. ## Source bridge review — 5 November 2026 The revenue-to-cash coverage gap is closed. Reported trailing revenue of CAD 1,731.218m now drives annual NOPAT and EBITDA margin paths. The prior earnings-factor forecast was first reproduced mechanically before updating estimates. Highland’s 11 owned months and KBK’s issuer nine-month pro-forma sales supply a labelled annual-sales proxy; purchase capital includes transaction fees. KBK adds only CAD 46.416m sales missing from the trailing baseline, with seller-derived cash still recovering over three years. SPV sales remain a separately labelled small-deal price/sales estimate. Existing horizon-specific weights were reassessed and retained: the core joint demand/capital scenarios have not been replaced and exact numerical frequency evidence has not appeared. Highland flat-month annualization and KBK pro-forma annualization are estimates with seasonality limits. SPV sales are not reported by the issuer and are estimated from its funded price and matched small-business economics. Margin recovery, target-specific maintenance/rent, bank refinancing, governance and limited small-deal supply remain material uncertainties. The 2.4× modeled borrowing ceiling is not an actual loan covenant. These are subjective estimates, not measured odds. # Bull–bear review — TerraVest, 5 November 2026 Subsequent source closure: the missing revenue bridge identified in this debate has now been completed. [Current source-derived estimates, corrections and same-price/date effects](../updates/2026-11-05-revenue-bridge-closure.md). The original independent argument record below is retained with its information set; it is not a claim that all estimated target margins have become reported facts. Independent advocates prepared opening cases without seeing the other draft, then exchanged point-by-point cross-examinations. A separate judge who did not author the prior model assessed the competing claims. Source reconstructions retain their stated provenance; inaccessible private reports were not newly read. Quoted sensitivities test arguments and are not automatically adopted forecasts. Fixed quote CAD 106.84, market time 2026-11-04T21:00:00.000Z; valuation date 2026-11-05. Original forecast exits retained. | Horizon | Before bear/base/bull | After bear/base/bull | Weighted before → after | | --- | --- | --- | --- | | 5 years | -11.24% / 7.94% / 12.19% | -11.24% / 7.94% / 12.19% | 4.62% → 4.62% | | 10 years | -3.05% / 10.16% / 15.62% | -3.05% / 10.16% / 15.62% | 9.06% → 9.06% | ## Bull case As of 5 November 2026. No opposing advocate draft was read. Existing company research, fact/calculation ledgers and exact registered source locators were examined; original issuer sources additionally checked live: Lumine Q3, CSU Q2, Diploma Q3 transcript, NVIDIA CFO commentary, TerraVest Highland/KBK, WOSG FY26. Harvia Q3 PDF retrieval failed; its registered extraction remains the evidence limit. Private source interpretations below are the repository's documented reconstruction, not a new claim to have read an unavailable private original. The strongest bear case is that Entrans consumed most recent capital at peak-cycle earnings, leaving leverage and concentrated industrial exposures precisely when trailer demand weakens. Chairman-related allegations add governance and funding uncertainty; neither a low entry multiple nor a good earlier cohort resolves that. Test the existing upside 4.6686× NOPAT, reflecting successful restructuring, without raising leverage. The restructuring uplift remains conditional; two deal quotes are not an unlimited addressable supply. **TVK-B2 — Entrans recovery is genuine uncredited optionality.** Sellers' reported pre-acquisition earnings and the larger installed manufacturing platform make some recovery economically possible. Test `recoveredEarningsFactor`1.10 versus 1.00 (10% group recovery, not a forecast of target EBIT), retaining initial factor 1.00 and the three-year recovery mechanism. Sources `terravest-entrans-2025`, acquisition price/7× EBITDA; `terravest-q3-2026-mda`, pp.7–8 segment losses and earnout reversal; `terravest-fy2025-fs`, note 4 pp.18–24. There is no source-derived10% quantum, so this sizes a hypothesis rather than promotes an input. Independent orders/segment results must establish the recovery. **TVK-B3 —17× can survive without perpetual peak growth.** A finite diagnostic needs about 14 post-exit years at 7% growth, 13.5% marginal return and 8% required return for 17× cash NOPAT. Proprietary niche equipment, restructuring and cross-selling are the mechanism; energy/heating substitution, governance and larger expensive deals limit it. Test13× to expose duration dependence. Today is a separate judgment. No revenue is fabricated for NOPAT-only funded cohorts. ### Isolated shared-engine effects Saved quote 106.84; market timestamp 2026-11-04T21:00:00.000Z. Fixed valuation date 2026-11-05; model 20a71f6ffc6c; exits remain the original model vintage. These are sensitivities, not joint forecasts or probability revisions. Baseline base IRR 7.94% / 10.16% (5/10 years). | Claim | Exact input/path | 5-year base | 10-year base | Change, pp (5/10) | | --- | --- | ---: | ---: | --- | | TVK-B1 | `cases.base.inputs.acquisitionMultiple` | 12.19% | 15.62% | +4.26 / +5.45 | | TVK-B2 | `cases.base.inputs.recoveredEarningsFactor` | 11.97% | 12.86% | +4.04 / +2.70 | | TVK-B3 | `cases.base.inputs.exitMultiple` | 1.07% | 6.94% | -6.87 / -3.22 | ## Bear case ### Strongest opposing case Highland and KBK provide evidence that reasonably priced smaller transactions still exist, while recurring tanks and heating replacement create cash through a trailer downturn. Original2024 source endpoints stay valid as historical calibration. ### Bear case The question is which deal population will receive today's much larger pool of capital. Entrans represents about 80% of recent deal capital and the current cohort's6–8% cash yield is below the 9–10% hurdle. Future base deals instead assume 7.40x mature NOPAT, a13.5% yield, from two smaller issuer-quoted multiples. This is a selection hypothesis, not a measured portfolio-wide result. Highland's rent treatment and future sellers' earnings durability are uncertain. Full committed capital and the trailing acquisition-year cycle point must stay in the denominator. [terravest-fy2025-fs note 4 printed pp.18–24; terravest-entrans-2025; terravest-highland-2023; terravest-kbk-2026; valuation/calculations.json small-deal and cohort bridge.] Leverage makes waiting costly. Net bank debt isCAD860.3m against 312.2m EBITDA, above the model 2.4x allocation ceiling. In the modeled recession leverage rises above 5x while the engine assumes refinancing rather than issuing equity. Financing at 6% and carrying principal through the March2027/2028 maturities are assumptions, not contractual proof. Measured debt reduction matters more than EBITDA growth purchased with fresh debt. [terravest-q3-2026-mda pp.11–13; terravest-fy2025-fs note 15.1 pp.32–33; research/06-risks.md.] The issuer's Q3 report itself attributes much of operating-cash growth to payables/customer deposits and notes higher maintenance capital. Deposits arrive with obligations; data-centre orders cannot be treated as permanent replacement demand. Current accounting gains, including contingent-consideration reversals, are already excluded appropriately. Governance allegations are unresolved and cannot be converted into an invented liability, but they reduce conviction in deal access and minority outcomes. [terravest-q3-2026, Business Performance/Cash Available for Distribution; terravest-faguy-authorization-2026-06-19 pp.1–12,18–19; application, not judgment.] The17x EV/NOPAT exit uses an 8% WACC and 3–5% perpetual growth at 13.5% RONIC. At9–10% capital cost or poorer deals the premium compresses. Do not pair a small-deal return indefinitely with capital supply at large-deal scale. ### Contested claims - TVK-B1 —Test12.05x future NOPAT purchases, the existing downside's measured adverse/cohort conversion, rather than selecting small-deal7.40x indefinitely. - TVK-B2 —Test14x exit versus 17x: still above no-growth12.5x at 8% WACC, and already documented as a sensitivity. - TVK-B3 —Test8% bank financing versus 6%, with all operating assumptions held fixed. This is an explicit funding stress, not a quote for the company's current facility. The Q3 issuer release was reopened; no subsequent court outcome, financing agreement or unpublished financial report was assumed. The missing revenue and cash detail of already-funded cohorts remains a limitation of the current forecast, not a claim that original historical source reproduction failed. ### Shared-engine effects Fixed saved quote CAD106.84; market timestamp 2026-11-04T21:00:00.000Z; valuation date 5 November 2026; fixed original exit calendar. Base before: 7.94% / 10.16% (5/10years). These are one-change sensitivities to the base, except named bundles; existing scenario probabilities are not revised or claimed as validated. | Claim | Change | 5-year base IRR | 10-year base IRR | 5-year effect | 10-year effect | | --- | --- | ---: | ---: | ---: | ---: | | TVK-B1 | Future deal multiple 12.0487x | 5.08% | 6.51% | -2.86pp | -3.65pp | | TVK-B2 | Exit14x | 2.97% | 7.83% | -4.97pp | -2.33pp | | TVK-B3 | Bank financing 8% vs 6% | 7.03% | 9.38% | -0.91pp | -0.78pp | ## Rebuttals ### Bull response 5 November 2026. Written after independently frozen openings were exchanged. Every numbered bear claim is addressed below; original bull openings and sensitivities remain unchanged. 1. **Bear12.05× future deals: rebut changing deal populations without matching budget.** Base annual purchases~100–150m cannot buy another Entrans; Highland/KBK are more representative in size and issuer-stated prices (`terravest-highland-2023`, price/multiple; `terravest-kbk-2026`, price/multiple). The aggregate dominated by Entrans correctly measures what recent capital earned, not necessarily what future small budgets can purchase. Still, two small transactions cannot establish indefinite seller supply; 7.40× is a conditional selection hypothesis and 4.67× successful-restructuring upside. Require new deal-by-deal observations, not aggregate success extrapolation. 3. **Bear8% bank financing: retain stress; no claim current facility repriced.** `terravest-fy2025-fs`, note 15.1 pp32–33 supplies actual loan structure and maturities; model 6% is an underwriting rate. Borrowing capacity falls with earnings, and cash first deleverages before deals. Financing failure or equity issuance can be worse than a higher-rate sensitivity. The next agreement must resolve maturity assumptions. 4. **Bull10% group recovery attacked: concede no quantified estimate.** Entrans purchase EBITDA and current trough segment results create an option, not evidence it adds exactly 10% group NOPAT. Keep it a sensitivity until plant orders/prices/utilisation and segment cash validate a bridge. Governance allegations are unresolved, not a fabricated provision. Missing funded-cohort revenue must remain disclosed. Conviction moderate in small-bolt-on mechanism; lower in forecasting consolidated cash through cycle/leverage. ### Bear response 5 November 2026, after independent openings were exchanged. 1. **Bull small-deal selection: partly conceded.** Planned100–150m annual budgets are a better match for Highland/KBK than another Entrans. Two smaller issuer-quoted multiples justify testing 7.40x NOPAT rather than forcing the Entrans-dominated aggregate onto every future purchase. I concede central 12.05x is not established by mismatched deal populations. The remaining challenge is whether enough such sellers exist at unchanged prices and durable EBITDA (`terravest-highland-2023`; `terravest-kbk-2026`; `terravest-fy2025-fs`, note 4).4.67x requires successful restructuring, not merely purchase at a low EBITDA quote. 2. **Bull10% group recovery: unquantified option conceded by advocate.** Current trough segment cash and acquired plants make recovery possible, but neither purchase EBITDA nor a higher equipment installed base establishes precisely 10% extra group NOPAT. Keep it a sensitivity until orders, utilization, prices and margins supply a bridge (`terravest-entrans-2025`; `terravest-q3-2026-mda`, pp.7–8). Do not recognize the earnout reversal as recurring cash. 4. **Bear8% funding stress: defended as stress, not current loan quote.** The actual debt structure and maturities must determine refinancing exposure; the engine's6% continuation remains an estimate (`terravest-fy2025-fs`, note 15.1 pp.32–33). Governance allegations, litigation and potential dilution are unresolved, not fabricated quantified claims. The issuer's current cash improvement partly reflects deposits/payables and cannot be promised annually (`terravest-q3-2026`, Cash Available for Distribution). No supported central financial-input revision was proved by these arguments. Keep the small-deal mechanism conditional, record the supply and terminal-duration weaknesses, freeze evidence that would distinguish them, and retain lower conviction on consolidated cash through leverage/cycle than on plant economics. ## Judgment Existing estimates receive no presumption merely because they are already in the model. Positive company evidence, the relevant outside view and the capital/funding bridge determine each verdict; neither arbitrary conservatism nor an unsupported favorable sensitivity replaces it. The evidence and reconstruction below explain numerical changes or explicit no-change conclusions claim by claim. | Contested claim | Verdict | Evidence/locator | Reason and model action | | --- | --- | --- | --- | Retain existing coarse horizon-specific weights after reassessment: scenarios already represent these adverse/favorable operating mechanisms. Numerical odds are subjective, not verified frequencies; unresolved gaps prevent high conviction. No tuning to return. ## Conviction Low; small-deal economics conditional, refinancing/governance unresolved and revenue-to-owner-cash forecast incomplete. These are forward estimates with surviving attacks, not confirmed outcomes. Neither numerical calibration, a completed checklist nor passing software tests establishes investment skill. Reported facts stay separate from forward judgments, and adverse paths are not worst-case bounds. **What would change the view:** Next FY2026 report/debt maturity disclosures: Entrans earnings, maintenance capital and deposits, all-in small deal prices, March2027/2028 refinancing. [Exact inputs, isolated sensitivities, model hashes and same-price/date before/after returns](../updates/2026-11-05-debate-review.json). ## Public Primary & Secondary Sources - `terravest-q3-2026` (primary): TerraVest Q3 FY2026 results — https://terravestindustries.com/terravest-announces-third-quarter-results-for-fiscal-2026-and-dividend-declaration/ (published 2026-08-13) - `terravest-fy2025` (primary): TerraVest FY2025 results — https://terravestindustries.com/terravest-announces-fourth-quarter-and-year-end-results-for-fiscal-2025-and-a-14-dividend-increase/ (published 2025-12-11) - `terravest-q3-2026-mda` (primary): TerraVest Q3 FY2026 issuer MD&A — https://cdn.financialreports.eu/financialreports/media/filings/47078/2026/RNS/47078_rns_2026-08-13_53c77f90-2953-4437-8ce0-88250e978c1a.pdf (published 2026-08-13) - `terravest-q2-2026-fs` (primary): TerraVest Q2 FY2026 interim financial statements — https://newsfile.futunn.com/public/NN-PersistNoticeAttachment/7781/20260514/SEDAR_PLUS/CSA_SEDAR_PLUS_NOTICE_RECORD_ID_2407071.pdf (published 2026-05-14) - `capital-method-roiic` (methodology): Mauboussin and Callahan: Return on Invested Capital — https://www.morganstanley.com/im/publication/insights/articles/article_returnoninvestedcapital.pdf (published 2022-10-06) - `capital-method-cash` (methodology): Dan Shuart, Eagle Point Capital: A Framework For Spotting Value Traps — https://eaglepointcapital.substack.com/p/spotting-value-traps-the-anti-compounders (published 2023-08-26) - `scenario-method` (methodology): Aswath Damodaran: Probabilistic Approaches in Valuation, chapter 33 — https://pages.stern.nyu.edu/~adamodar/pdfiles/val3ed/c33.pdf - `terravest-press-index` (primary): TerraVest issuer press releases — https://terravestindustries.com/investor-relations/press-releases/ - `terravest-auto-plan-2026` (primary): TerraVest automatic share-purchase plan — https://terravestindustries.com/terravest-announces-automatic-purchase-plan-for-common-shares/ (published 2026-10-15) - `terravest-ncib-2026` (primary): TerraVest normal-course issuer bid renewal — https://terravestindustries.com/terravest-announces-renewal-of-normal-course-issuer-bid-for-its-common-shares-2/ (published 2026-08-19) - `terravest-events` (primary): TerraVest events and presentations — https://terravestindustries.com/investor-relations/events-and-presentations/ - `act-trailers-october-2026` (primary): ACT Research September trailer orders and backlog — https://www.actresearch.net/resources/blog/state-of-industry-u.s.-trailers-blog (published 2026-10-23) - `ftr-trailers-reviewed-2026` (primary): FTR trailer orders: mixed-period page — https://www.ftrintel.com/trailer-orders - `aos-q2-2026` (primary): A. O. Smith Q2 2026 results — https://www.prnewswire.com/news-releases/a-o-smith-reports-second-quarter-2026-results-302838509.html (published 2026-07-30) - `aos-q3-2026-calendar` (primary): A. O. Smith Q3 publication notice — https://investor.aosmith.com/node/23166/pdf (published 2026-10-02) - `nucor-q3-2026` (primary): Nucor Q3 2026 results — https://investors.nucor.com/news/news-details/2026/Nucor-Reports-Results-for-the-Third-Quarter-of-2026/default.aspx (published 2026-10-26) - `wabash-q3-2026-calendar` (primary): Wabash Q3 2026 publication notice — https://ir.onewabash.com/news/news-details/2026/Wabash-Schedules-Third-Quarter-2026-Earnings-Conference-Call/default.aspx (published 2026-10-05) - `wabash-q3-2026` (primary): Wabash Q3 2026 results — https://www.globenewswire.com/news-release/2026/10/29/3391075/10640/en/wabash-announces-third-quarter-2026-results.html (published 2026-10-29) - `aos-q3-2026` (primary): A. O. Smith Q3 2026 results — https://investor.aosmith.com/news-releases/news-release-details/o-smith-reports-third-quarter-2026-results (published 2026-10-29) - `eia-propane-2026-10-28` (primary): EIA weekly propane/propylene stocks — 28 October release — https://www.eia.gov/dnav/pet/hist/LeafHandler.ashx?n=PET&s=WPRSTUS1&f=W (published 2026-10-28) - `terravest-fy2024-fs` (primary): TerraVest FY2024 audited consolidated financial statements (years ended 30 September 2024 and 2023) — https://newsfile.futunn.com/public/NN-PersistNoticeAttachment/7781/20241213/SEDAR_PLUS/CSA_SEDAR_PLUS_NOTICE_RECORD_ID_1738509.pdf (published 2024-12-13) - `terravest-fy2025-fs` (primary): TerraVest FY2025 audited financial statements — https://newsfile.futunn.com/public/NN-PersistNoticeAttachment/7781/20251211/SEDAR_PLUS/CSA_SEDAR_PLUS_NOTICE_RECORD_ID_2278262.pdf (published 2025-12-11) - `trinity-q3-2026` (primary): Trinity Industries Q3 2026 results — https://www.trin.net/investor/press-releases/news-details/2026/Trinity-Industries-Inc--Announces-Third-Quarter-2026-Results/default.aspx (published 2026-10-29) - `ism-manufacturing-oct-2026` (primary): ISM October 2026 Manufacturing PMI report — https://www.ismworld.org/supply-management-news-and-reports/reports/ism-pmi-reports/pmi/october/ (published 2026-11-02) - `terravest-fy2024` (primary): TerraVest FY2024 results — https://terravestindustries.com/terravest-announces-fourth-quarter-and-year-end-results-for-fiscal-2024and-a-17-dividend-increase/ (published 2024-12-13) - `terravest-fy2023` (primary): TerraVest FY2023 results — https://terravestindustries.com/terravest-announces-fourth-quarter-and-year-end-results-for-fiscal-2023-and-a-20-dividend-increase/ (published 2023-12-13) - `terravest-entrans-2025` (primary): TerraVest acquisition of Entrans International — https://terravestindustries.com/terravest-announces-the-acquisition-of-entrans-international-and-amended-credit-facility/ (published 2025-03-17) - `terravest-highland-2023` (primary): TerraVest acquisition of Highland Tank — https://terravestindustries.com/terravest-announces-the-acquisition-of-highland-tank/ (published 2023-11-01) - `terravest-kbk-2026` (primary): TerraVest acquisition of KBK Industries — https://terravestindustries.com/acquisition-of-kbk-industries-llc/ (published 2026-01-09) - `terravest-fy2022` (primary): TerraVest FY2022 results — https://terravestindustries.com/fourth-quarter-and-year-end-results-2022/ (published 2022-12-14) - `terravest-pellerin-update-2026` (primary): TerraVest provides update (Executive Chairman allegations) — https://www.miningstockeducation.com/2026/06/terravest-provides-update/ (published 2026-06-05) - `amf-allegation-reports-2026` (secondary): Press and analyst reports of the AMF investigation of TerraVest's Executive Chairman — https://yonatanbrunshtein.substack.com/p/coverage-update-terravest-industries (published 2026-06-25) - `insider-holdings-aggregator-2026` (secondary): MarketScreener insider holdings: Charles Pellerin and Dustin Haw — https://www.marketscreener.com/insider/CHARLES-PELLERIN-A16THX/ - `terravest-management-page` (primary): TerraVest management and directors — https://terravestindustries.com/company/management-directors/ - `terravest-deal-releases` (primary): TerraVest acquisition releases, 2022–2026 — https://terravestindustries.com/investor-relations/press-releases/ (published 2026-07-02) - `yahoo-tvk-trading-2026-11` (market data): Yahoo Finance TVK.TO daily price and volume, three months to 3 November 2026 — https://query1.finance.yahoo.com/v8/finance/chart/TVK.TO?range=3mo&interval=1d (published 2026-11-03) - `terravest-faguy-authorization-2026-06-19` (primary claimant pleading; allegations unproven): Faguy & Co: proposed TerraVest class-action authorization motion — https://cae7fc35-4bdf-4bf9-a8fd-692aa9958840.usrfiles.com/ugd/40615b_d1a93ee1415b420fa1a4589d9ecd3d7b.pdf (published 2026-06-19) - `dot-lng-proposal-2026-10-28` (primary government proposal announcement): US DOT announcement of proposed LNG facility safety revisions — https://www.transportation.gov/briefing-room/trumps-transportation-secretary-sean-p-duffy-modernizes-lng-regulations-secure (published 2026-10-28) - `eia-weekly-2026-11-04` (primary industry statistics): EIA weekly petroleum status summary, week ended 30 October 2026 — https://www.eia.gov/petroleum/supply/weekly/archive/2026/2026_11_04/download/summary.txt (published 2026-11-04) - `terravest-q3-2026-fs-bridge` (primary issuer filing (HTML mirror)): TerraVest Q3 2026 issuer financial statements, source transcription — https://financialfilings.com/filings/terravest-industries/interim-quarterly-report/2026/75316275/ (published 2026-08-13) - `terravest-fy2024-fs-bridge` (primary issuer filing): TerraVest FY2024 audited financial statements — Highland acquisition costs — https://www.sedarplus.ca/csa-party/records/document.html?id=b96ff8ace23bd0ea3b23527ab48ac881d07e384e8e58d2fa4e94f8d39078793c (published 2024-12-13) - `eia-gas-storage-2026-11-05` (primary public industry statistics): EIA Weekly Natural Gas Storage Report, week ended 30 October 2026 — https://ir.eia.gov/ngs/ngs.html (published 2026-11-05) - `act-commercial-vehicle-2026-11-11` (primary): ACT: November North American commercial vehicle outlook public update — https://www.actresearch.net/resources/blog/northamerica-commercialvehicle-forecast-blog (published 2026-11-11) --- # Watches of Switzerland (WOSG.L) - Legal Name: Watches of Switzerland Group PLC - Exchange: London Stock Exchange (United Kingdom) - Sector: Luxury watch and jewellery retail - Currency: GBP - Reviewed: 2026-11-15 - Model Framework: Unit economics and cash-funded acquisitions - Financial Period: FY26 P&L normalized52/53; May operating-capital proxy;1November reported net-debt funding anchor.10November trading update informs forward earning power; full interim accounts pending. (anchored 2026-10-27) - Research Status: 8 of 11 research checks complete. Still to complete: Security and evidence base; Opportunities and funded scenarios; Variant view and catalysts - Static Report URL: https://longview-investing.vercel.app/research/watches-of-switzerland.html - Interactive Model URL: https://longview-investing.vercel.app/#/company/watches-of-switzerland - JSON API: https://longview-investing.vercel.app/api/companies/watches-of-switzerland ## Executive Summary A luxury retailer whose returns depend on brand access, productive showrooms, inventory turns and cash conversion. A growing watch market does not guarantee the retailer keeps the economics. ## Key Operational Watch Items - Brand allocation, retailer margins and Rolex/Bucherer channel decisions - Inventory turns, CPO capital, customer deposits and actual cash conversion - Store productivity after full maintenance/refurbishment and cash rent - Reachable US/CPO/jewellery growth, bank covenants and minority purchase settlements ## Valuation Scenario Assumptions ### Downside Case An ordinary adverse outcome: near-term H1 evidence, then two years of weaker demand, another supplier margin cut and slower stock, followed by limited recovery and cost/project adaptation. No acquisitions; positive cash reduces debt. Agency withdrawal or unsupported refinancing is a separate unquantified tail, not an automatic common wipeout. With the deal budget at zero, cash the business does not need is returned after debt, reserve and claims under the stated capital-allocation policy, rather than parked without yield indefinitely. 15November reopened: owned acquisition growth is funded, dealer/refit economics are matched, ordinary downside is separated from catastrophic allocation loss, and actual future unit profit receives a finite scenario-specific continuation. No fit to any published IRR. - Key Drivers: Supplier / merchandise gross-margin change: -1.0% | Inventory days change: 15.0% | Post-organic surplus cash allocated to acquisitions: 100.0% | All-in purchase / mature owner cash: 15.10× | Normalized operating tax rate: 27.0% | Annual share dilution: 0.0% | Protected cash retained at par: 100.0% | Stress loss / contingent exposure: 0.0% - UK allocation-led retail volume opportunity before realization: 4.0%, 2.0%, 2.5%, 3.5%, 2.5% - Other UK retail volume opportunity before realization: 4.0%, 2.0%, 2.0%, 2.0%, 2.0% - US allocation-led retail volume opportunity before realization: 7.0%, 5.0%, 5.0%, 8.0%, 6.0% - Other US retail volume opportunity before realization: 5.0%, 3.0%, 2.5%, 2.0%, 2.0% - External US wholesale volume opportunity before realization: 8.0%, 4.0%, 3.5%, 3.0%, 3.0% - Net fixed cash-cost growth: 3.0%, 3.0%, 3.0%, 2.0%, 2.0% - New capacity required per unit of volume growth: 0.60×, 0.30×, 0.30×, 0.50×, 0.50× - Volume opportunity realized: 75.0%, 0.0%, 0.0%, 50.0%, 50.0% - Annual price growth: 1.5%, 1.5%, 2.0%, 2.0%, 2.0% - Expensed growth investment: 4.00, 2.00, 1.00, 1.00, 1.00 - Deal budget (capital available for acquisitions): 0.00, 0.00, 0.00, 0.00, 0.00 - Acquired cohort organic growth: 0.0%, 0.0%, 0.0%, 0.0%, 0.0% - Revenue growth: 7.6%, 1.5%, 2.0%, 3.9%, 3.6% - Margin: 8.2%, 8.0%, 7.7%, 7.8%, 7.9% ### Base Case Paid capacity, CPO/jewellery, supplier-approved consolidation and maintained gross margins support earning power. Fund continuing growth programmes, wear and stock; then buy profitable targets at source-derived all-in economics within the existing seller budget. 15November reopened: owned acquisition growth is funded, dealer/refit economics are matched, ordinary downside is separated from catastrophic allocation loss, and actual future unit profit receives a finite scenario-specific continuation. No fit to any published IRR. - Key Drivers: Supplier / merchandise gross-margin change: 0.0% | Inventory days change: 0.0% | Post-organic surplus cash allocated to acquisitions: 100.0% | All-in purchase / mature owner cash: 8.13× | Normalized operating tax rate: 27.0% | Annual share dilution: 0.0% | Protected cash retained at par: 100.0% | Stress loss / contingent exposure: 0.0% - UK allocation-led retail volume opportunity before realization: 4.0%, 2.0%, 2.5%, 3.5%, 2.5% - Other UK retail volume opportunity before realization: 4.0%, 4.5%, 4.7%, 5.0%, 3.6% - US allocation-led retail volume opportunity before realization: 7.0%, 5.0%, 5.0%, 8.0%, 6.0% - Other US retail volume opportunity before realization: 5.0%, 5.0%, 5.2%, 5.4%, 4.0% - External US wholesale volume opportunity before realization: 8.0%, 9.8%, 9.8%, 9.8%, 5.9% - Net fixed cash-cost growth: 1.5%, 3.0%, 3.0%, 3.0%, 3.0% - New capacity required per unit of volume growth: 0.60×, 0.80×, 1.00×, 1.00×, 1.00× - Volume opportunity realized: 100.0%, 100.0%, 100.0%, 100.0%, 100.0% - Annual price growth: 2.0%, 2.0%, 2.0%, 2.0%, 2.0% - Expensed growth investment: 4.00, 4.00, 4.00, 4.00, 4.00 - Deal budget (capital available for acquisitions): 38.27, 38.27, 49.14, 49.14, 49.14 - Acquired cohort organic growth: 5.0%, 5.0%, 5.0%, 5.0%, 5.0% - Revenue growth: 9.4%, 8.7%, 8.7%, 10.1%, 8.2% - Margin: 8.7%, 9.0%, 9.3%, 9.6%, 9.8% ### Upside Case Better throughput/conversion, modest supplier-margin recovery and stock efficiency support higher earning power. Spend more on the project pipeline and fund all capacity/stock and capped acquisitions; use the same terminal economics as base, not a growth-driven rerating. 15November reopened: owned acquisition growth is funded, dealer/refit economics are matched, ordinary downside is separated from catastrophic allocation loss, and actual future unit profit receives a finite scenario-specific continuation. No fit to any published IRR. Stronger operations plus conditional quality-market realization define the five-year upside; its weight is rejudged as a joint outcome. Ten-year valuation returns to the independently funded continuation. - Key Drivers: Supplier / merchandise gross-margin change: 0.5% | Inventory days change: -5.0% | Post-organic surplus cash allocated to acquisitions: 100.0% | All-in purchase / mature owner cash: 8.13× | Normalized operating tax rate: 27.0% | Annual share dilution: 0.0% | Protected cash retained at par: 100.0% | Stress loss / contingent exposure: 0.0% - UK allocation-led retail volume opportunity before realization: 4.0%, 2.0%, 2.5%, 3.5%, 2.5% - Other UK retail volume opportunity before realization: 4.0%, 5.4%, 6.0%, 6.6%, 4.3% - US allocation-led retail volume opportunity before realization: 7.0%, 5.0%, 5.0%, 8.0%, 6.0% - Other US retail volume opportunity before realization: 5.0%, 6.1%, 6.6%, 7.2%, 4.8% - External US wholesale volume opportunity before realization: 8.0%, 13.1%, 13.1%, 13.1%, 7.1% - Net fixed cash-cost growth: 1.5%, 3.0%, 3.0%, 3.0%, 3.0% - New capacity required per unit of volume growth: 0.60×, 1.00×, 1.00×, 1.00×, 1.00× - Volume opportunity realized: 130.0%, 130.0%, 130.0%, 130.0%, 130.0% - Annual price growth: 2.5%, 2.5%, 2.5%, 2.5%, 2.5% - Expensed growth investment: 5.00, 5.00, 5.00, 5.00, 5.00 - Deal budget (capital available for acquisitions): 54.67, 54.67, 72.34, 72.34, 72.34 - Acquired cohort organic growth: 8.0%, 8.0%, 8.0%, 8.0%, 8.0% - Revenue growth: 11.6%, 12.3%, 12.6%, 14.6%, 11.4% - Margin: 8.9%, 9.5%, 10.1%, 10.7%, 11.0% ## Research Notebook # Business economics Current15November interpretation: scarce-brand cash productivity and demand resilience are supported; the brand-level tax/capital/cost perimeter remains uncertain, not disproven. [Reopened underwriting](../updates/2026-11-15-reopened-underwriting.md) governs current forecasts; historical numerical examples below are calibrations, not current group measurements. Reorganised 3 November 2026 under the pipeline's gated questions. Earlier text (27 October) is kept in substance below its new headings; nothing was deleted that carried evidence. GBP millions unless stated. Facts cite `financials/facts.json` ids and `sources/registry.json` ids; estimates are labelled. ## Evidence plan What the filings answer: six years of revenue, adjusted EBIT/EBITDA, the issuer's APM cash-flow table, net debt, inventory and payables (wosg-results-2022 to wosg-results-2026), segment profit for UK, US retail and US wholesale from FY25 (wosg-results-2026 financial review), the acquisition notes (wosg-annual-2021/2023/2025/2026) and the capital-allocation framework (wosg-results-2026, Capital Allocation section). What they do not answer: brand-level sales, margins and capital (the issuer declines to disclose Rolex's share), store cohorts, allocation contracts and their duration, SKU ageing. No interviews, store visits or channel checks were done for this repository. ## Value-creation mechanism WOSG turns capital into owner cash by converting a supplier's scarce allocation into retail gross profit, and the step where most value is created is winning and keeping that allocation. Rolex, Patek Philippe and Audemars Piguet make fewer watches than clients want; they choose which authorised dealers get them and in which locations. WOSG earns that allocation by spending its own capital on showrooms built to the brand's standard (£65.9m of expansionary capex in FY26, guided £60–70m a year, wosg-results-2026), by carrying stock, training staff and running client lists and the Rolex Certified Pre-Owned programme. In return it gets products that are largely pre-sold, carry no markdown risk and turn far faster than other brands. The second engine, US consolidation, extends the same mechanism by buying family-owned Rolex agencies (Ben Bridge, Betteridge, Bernie Robbins, Deutsch & Deutsch) whose transfer the brand must approve, then refurbishing them. Roberto Coin wholesale (FY25) is a different mechanism: a brand-distribution business with its own receivables and design risk. ## Business-model profile Comparable firms are chosen by these traits, not by the industry label. **Growth vehicle: Mixed: organic and acquisitions.** Revenue rose from 905.1 in FY2021 to 1,827.9 in FY2026. Organic showroom investment and stock took 51% (£267.8m) of growth investment and acquisitions most of the rest. Acquisition cash was 3.8% of revenue over FY2024–FY2026, with deals in every year from FY2022 to FY2026. New showrooms and acquired showrooms both depend on the brands allocating supply, so the model forecasts organic units and a yearly deal budget separately. **Capital intensity: 4.4% of sales, asset-light.** Capital spending is about 4.4% of revenue: showroom fit-outs and refurbishments that the brands require. Most showrooms are leased (lease liabilities of 427.6 at May 2026), so capex understates the capital in the estate; inventory (142 days of cost of sales) is the other large use of capital. **Comparables.** A comparable firm is an authorized retailer of supply-constrained luxury brands: the brands allocate scarce products and approve showrooms, so growth needs brand permission, leased showrooms and inventory. Matching traits: supplier-controlled allocation, leased stores, mixed organic and acquired growth, discretionary-spending cycles. Not the label: "Specialty retail" implies the retailer controls its products and prices; here the brands control allocation and pricing, and Rolex owns a competing retailer. Declared in `valuation/model.json` (`businessProfile`). `npm run check` measures deal and capital intensity from the reported facts, and the outside view grades the forecast only against firms that share them. ## Economic snapshot Definitions: NOPAT is pre-IFRS16 adjusted EBIT (rent already charged as cash) taxed at 27%; the model's normalised figure also replaces accounting rent with cash rent and scales the 53-week year to 52 (calculations.json `annualOperatingProfit` = 142.95 pre-tax, 104.4 after tax). Capital is book capital at 3 May 2026. | Item | FY26 | Source | | --- | ---: | --- | | Revenue (53 weeks) | 1,827.9 | `revenue` | | Gross ("net") margin, UK / US retail / US wholesale | 35.0% / 34.9% / 42.2% | wosg-results-2026 financial review | | Adjusted EBIT margin (pre-IFRS16) | 8.5% (FY25 9.1%, FY23 10.7%) | `adjustedEbit`, `ebitFy23` | | Normalised cash-rent EBIT margin (52 weeks) | 8.0% | calculations `annualOperatingProfit` / `annualRevenue` | | Inventory days on cost of sales | 142 (458.1 / (1,827.9 − 651.3) × 365); FY25 155 | `inventory`, segment gross profit | | PP&E / right-of-use assets | 217.0 / 338.2 | `physicalAssets`, wosg-results-2026 balance sheet | | Operating working capital excl. cash | 271.6 (inventory 458.1 + receivables 48.5 − operating payables 235.0) | calculations `workingCapital` | | Goodwill and intangibles | 319.7 | `goodwillIntangibles` | | Bank debt / leases / cash (of which restricted or in transit) | 122.1 / 427.6 / 65.1 (43.0) | `termDebt`+`revolverDebt`, `leaseDebt`, `cash` | Returns after tax: on tangible capital with leases expensed, 104.4 / (217.0 + 271.6) = 21.4%; including goodwill (ROCE, leases expensed) 104.4 / 808.3 = 12.9%; capitalising leases (post-IFRS16 adjusted EBIT 178.8, taxed and scaled to 52 weeks, on 826.8 including right-of-use assets) about 15.5%. The issuer's pre-tax pre-IFRS16 ROCE is 18.0% (`roceFy26`), down from 27.9% in FY23 (`roceFy23`). Incremental margin: FY25 to FY26 adjusted EBIT rose 5.1 on 176.4 more revenue (2.9%); FY23 to FY26 it fell 10.3 while revenue rose 285.1. Since the FY23 peak, extra sales have not carried group margin. The causes the issuer names are brand margin-rate reductions after US tariffs and gold-price rises, product mix, US e-commerce investment and a one-off department-store debtor write-down (wosg-results-2026). Normalisations applied before these returns: cash rent in place of accounting rent, 52/53 weeks, capex charged in full rather than the £3m "maintenance" line, share compensation kept as a cost, no minorities left after the D&D put was booked as a liability (`07-assumption-register.md`). ## Economic segmentation Reported segments are UK (EBIT 65.9), US retail (71.1) and US wholesale (25.0), with about 7.2 of unallocated cost (wosg-results-2026). The economically meaningful cut is by how the profit is earned: | Economic unit | Sales (model, 52 wk) | Unit EBIT (model) | Share of unit EBIT | Demand driver | Direction | | --- | ---: | ---: | ---: | --- | --- | | Allocation-led watches (Rolex, Patek, AP; UK and US; incl. Rolex CPO) | 795 | 74.7 | 52% | Brand output and allocation; client lists exceed supply | Grows with brand capacity (Bulle from 2029) and new agencies; not cyclical in volume, exposed to dealer-margin terms | | Other retail (other watch brands, jewellery, services, e-commerce) | 884 | 47.6 | 33% | Discretionary luxury spending, footfall, gold price | Cyclical; UK estate shrinking (35 closures FY25–FY26), US growing | | Roberto Coin US wholesale | 114 | 20.7 | 15% | Department-store and independent jeweller orders, Roberto Coin brand | Growing (+22% cc FY26) but carries receivable and inventory risk | The issuer does not disclose it. The cut shows that about 44% of retail sales produce more than half the profit, and that the share of value is larger again because this unit is the least cyclical and earns the highest return on capital (Unit economics). Revenue share does not describe profit share here: other retail is the largest unit by sales and the smallest by return. ## Unit economics The unit through which WOSG adds value is a multi-brand showroom with a brand shop-in-shop. In summary: the Rolex section produces roughly two-thirds of the store's £12.6m sales and about 60% of its gross profit at a 33% gross margin against 45% for other brands. After staff, rent, sustaining capex, other costs and an allocated half of head-office cost, it makes about £1.2m of the store's £1.6m EBIT. Its tangible capital is about £2.1m (capex about £1,500 per sq ft against £700, but only about 40 days of stock against about 243). After tax that is roughly a 40% return on tangible capital for the Rolex section against about 10% for the rest. The comparison that matters is the same square footage with and without the allocation: the non-Rolex space earns about the cost of capital, the allocated space earns four times it. The issuer's own unit rule is consistent in direction: showroom projects must pay back (fixed assets plus inventory, divided by showroom EBITDA) in under three years (wosg-results-2026, Balance Sheet), a pre-tax cash yield above 33% on the project's tangible capital, roughly 24% after tax before rent escalation and refurbishment cycles. This is a hurdle, not an achieved return. Capacity example: the Rolex Old Bond Street flagship carried £4.2m of exceptional pre-opening rent in FY25 (wosg-results-2025) and is described as "one of the leading Rolex retail destinations globally" (wosg-results-2026). No audited outcome is published, so the project IRR through its dead-money year is not computable from public data. Failures where the levers did not work: - Europe. Nine mono-brand stores with no Rolex were divested to brand partners between 2024 and FY26, with exit costs of £0.7m in FY25 and a £0.4m gain in FY26 (wosg-results-2026 exceptional items). Without allocation the format did not earn its capital. - UK non-core estate. 35 showrooms closed in FY25–FY26 and showroom impairments were £44.5m (FY25) and £8.8m gross (FY26) as projects appraised before rate and inflation rises fell short (wosg-results-2026). These are allocation-light formats and some refurbishments. - Hodinkee (£10.7m, FY25): pro forma profit "not material" and its overheads were annualised in FY26 US retail costs (wosg-annual-2025 p.264; wosg-results-2026). ## Independent checks 1. **Profit follows allocation, not sales.** In FY24 Rolex reduced its UK allocation (December 2023) and the UK lost tax-free tourist sales. Group revenue was flat (1,537.9 against 1,542.8) and adjusted EBIT fell 18% from 165.1 to 134.7 (`revenueFy24`, `adjustedEbitFy24`). A 0.3% sales decline costing 30.4 of EBIT is what a model with a high-profit allocation-led unit predicts when mix shifts to other brands; a blended-margin retailer would have lost a few million. 3. **Format exit.** The only region without Rolex (Europe) was sold to brand partners. Group ROCE (18–28% pre-tax) sits between the unit model's two returns, as it should for a blend. Adverse allocation. Our version removes the margin advantage entirely: give both retail units the group's 8.0% normalised margin. Allocation-led sales of 795 then earn 63.6 pre-tax. Other retail's 884 of sales earn 70.7 on about 365 of stock plus 393 of fit-out, about 7% after tax. The return gap survives because it comes from capital turnover (stock that sells off a list), not from the disputed cost split. The observation that would settle it is brand-level EBIT, which WOSG declines to publish. The FY24 profit warning is the best public natural experiment, and it supports the profit-concentration view. ## Other economics retained from the 27 October study New and pre-owned watches and jewellery differ in sourcing, margin, ageing and working capital. CPO widens price points and repeat trade-ins but requires owned stock, inspection and warranties, and a CPO sale can replace a new-watch sale, so opportunities are not simply additive. Group pre-owned watch revenue passed 8% of luxury watch revenue in FY26 (wosg-results-2026). The cash account is inventory and receivables less supplier and customer float, plus fit-out and software capital and cash rent; all leases are charged as cash rent, so IFRS16 lease debt is excluded from the bank-debt bridge. Customer deposits and supplier credit serve sales; they are not acquisition funding. A customer registration of interest is not a firm order. Suppliers can and did change dealer margins as prices and tariffs moved: FY26 percentage margins fell while "retailer cash margins were maintained or improved" through RRP increases (wosg-results-2026, US review). # History, market and counterparties Reorganised 3 November 2026 under the stage-4 questions; the 27 October starting-point notes are kept at the end. GBP millions. ## Demand drivers and cycle The end user and payer are the same person, a high-income watch or jewellery buyer, but the volume decision for the profitable unit belongs to someone else. For allocation-led watches, Rolex (and Patek, AP) decides how many pieces each dealer receives and the dealer decides which client on its list gets each one. For other brands and jewellery the client decides, and WOSG carries the stock until they do. Demand therefore splits into: - **Discretionary demand** for other watch brands, jewellery (£238m, +18% cc in FY26) and services. This part is cyclical, sensitive to UK consumer confidence, tourist tax rules and the gold price. - **Replacement and secondary-market demand** through CPO and trade-ins (pre-owned above 8% of watch revenue in FY26, wosg-results-2026). It depends on resale prices. - **Regulation-driven** effects are price and tax, not volume: the UK's removal of tax-free tourist shopping (UK tourist sales remain "significantly below" the tax-free era, wosg-results-2026) and US tariffs on Swiss imports, which brands passed into retail prices while cutting dealer percentage margins. Measured cycle, from the issuer's own six years (`revenueFy21` … `revenue`, `ebitFy21` … `adjustedEbit`): | FY | Revenue | Adj. EBIT | EBIT margin | ROCE (pre-tax) | What happened | | --- | ---: | ---: | ---: | ---: | --- | | 2021 (53 wk) | 905.1 | 77.6 | 8.6% | 19.7% | COVID closures | | 2022 | 1,238.0 | 130.3 | 10.5% | 27.4% | Post-COVID boom, grey-market premiums | | 2023 | 1,542.8 | 165.1 | 10.7% | 27.9% | Peak | | 2024 | 1,537.9 | 134.7 | 8.8% | 19.5% | UK allocation cut, secondary prices unwind, Bucherer deal | | 2025 | 1,651.5 | 149.7 | 9.1% | 19.0% | US growth, Roberto Coin | | 2026 (53 wk) | 1,827.9 | 154.8 | 8.5% | 18.0% | Tariff-driven dealer-margin cuts, D&D | The downturn of FY24 cost revenue nothing and EBIT 18%; it was a mix and allocation event, not a collapse in end demand. The test series since then is WOSG's own US constant-currency growth (+16% FY25, +24% FY26) and the September 2026 statement that trading in the first 17 weeks was consistent with 5–10% FY27 growth (wosg-agm-2026). That favours normalisation over impairment. The open question has moved from volume to dealer margin: FY23's 10.7% EBIT margin has not returned, and brands cut percentage margins twice (after tariffs and after gold-price rises). Swiss export data (fh-september-2026) measure shipments, not sell-through, and the federation itself says so; they are context, not a test. ## Competition Competition happens through allocation and locally. A client buys a Rolex from a dealer within reach that has the model, so stores compete within a metropolitan area, and dealers compete for the brand's approval to open, relocate or acquire a door. Brands are consolidating to "fewer, higher-quality points of distribution" (wosg-results-2026 mid-term priorities), so the number of competitors per brand is falling and the remaining ones must invest more per door. In the UK it is the largest retailer for Rolex, Omega, Cartier, TAG Heuer and Breitling (wosg-results-2026). The advantage in one sentence: WOSG is the dealer brands trust to fund and run their best doors, shown by the Old Bond Street flagship, the first Rolex agency in Newcastle (FY26) and Rolex's consent to transfer four Texas agencies when WOSG bought Deutsch & Deutsch in January 2026. Its evidence of durability is that these brand decisions came after Rolex bought Bucherer in August 2023 (rolex-bucherer). The competitor that could hurt is Bucherer, now owned by Rolex, which bought Tourneau, the largest US watch retailer, with 28 stores in 2018 (bucherer-tourneau-2018) and then had four London stores (same source). It has not hurt WOSG yet because Rolex states that Bucherer operates independently and, on the evidence, has kept awarding WOSG agencies. If Rolex chose, it could route new US allocation through Bucherer doors instead. The second competitive force is the brands' own boutiques: Richemont reports growth in its directly operated stores (richemont-q1-fy27), and brand mono-boutiques are the partners' outside option for non-Rolex brands. WOSG's response is to run those mono-brand boutiques for the brands (75 at September 2026, wosg-agm-2026). ## Counterparty incentives | Counterparty | What it wants | What WOSG gives it | Its outside options | What would change the arrangement | | --- | --- | --- | --- | --- | | Other watch brands (Omega, Cartier, TAG, Breitling, Tudor) | Sell-through, brand presentation, reach | Multi-brand floor space and mono-brand boutiques | Their own boutiques and e-commerce | Weak sell-through; brands pulling margin (as after tariffs) | | Roberto Coin S.p.A. | US growth for its brand | Exclusive North American distribution, marketing (Dakota Johnson campaign) | Distribution through another party at contract end | Contract terms, which are not disclosed | | Landlords | Long leases with prime tenants | £83.9m a year of cash rent on £427.6m of lease liabilities | Other luxury tenants | Renewal pricing in prime locations | | Lenders | Covenants: net debt ≤ 3× EBITDA, fixed-charge cover; facilities to May 2028 | Net debt/EBITDA 0.28× (wosg-results-2026) | — | Only a severe downturn | | D&D minority sellers | Value for their 12.19% | Put rights booked at £8.9m | — | Profit-linked settlement could exceed book | Dependence runs both ways. WOSG cannot replace Rolex: allocation-led units carry over half of profit (01-business.md). Rolex can replace WOSG in a given city, but only by finding another operator willing to put up the capital, or by funding stores itself, which its history and stated aims argue against. The Bucherer purchase is evidence that Rolex acted to keep its Swiss network out of a competitor's hands (Rolex's own statement, rolex-bucherer); it is not evidence of Rolex's motive towards WOSG. The best evidence of motive is subsequent action: new agencies and the D&D transfer approved for WOSG through 2026. The counterparty move to watch is any Bucherer or Tourneau opening in a city where WOSG holds a Rolex agency. ## Current starting point (27 October 2026) The starting point is the portfolio at 3 May 2026. FY26 had 53 weeks; the model normalises the income statement by 52/53 and leaves balance-sheet claims as reported. FY26 revenue £1,827.9m, pre-IFRS16 adjusted EBIT £154.8m, operating profit after IFRS16 and exceptionals £170m. Gross margin 35.6% against 36.3% in FY25. The reported 18% ROCE is a group average, not a new showroom's return or a forward ROIIC. Deutsch & Deutsch's four locations were acquired on 22 January 2026; FY26 includes £16.4m of revenue against £56.7m for a full year. The missing £40.3m is already funded and enters the first forecast year once. The minority sellers' purchase rights were booked as an £8.9m obligation. Cash fell from £98.9m to £65.1m (−£33.6m movement, −£0.2m FX); gross bank borrowing fell to £122.1m and net bank debt to £57m. Restricted and in-transit cash is protected rather than treated as growth funding. The 3 September 2026 AGM update reiterated FY27 guidance (5–10% constant-currency revenue growth, 40–80bps adjusted EBIT margin expansion, £60–70m capex, about 70% FCF conversion) and listed Rolex Glasgow, Betteridge Greenwich and two Marlton, New Jersey showrooms for opening before Christmas; it published no balance sheet (wosg-agm-2026). The 23 October director RNS concerns an outside directorship. H1 FY27 trading is due 10 November and results 10 December (wosg-calendar). # Management and allocation incentives Reorganised 3 November 2026 under the stage-5 questions. Brian Duffy has been chief executive since 2014 and Anders Romberg finance director (2014 to January 2022, and again from 2023). Management controls showroom and technology spending, closures, working capital and acquisition selection; brands control allocation and dealer terms, so execution is judged within those limits. ## Ownership and incentives Insider table. Current holdings after the 2025 vestings and the FY26 annual report table were not re-extracted; treat the share counts as approximate. | Person | Shares (Aug 2024) | Cash pay | Open-market dealings (date, price) | Background | | --- | ---: | --- | --- | --- | | Brian Duffy, CEO | about 7.7m, plus about 0.98m options | Base £500k, unchanged since 2014; total about £3.1m | Sold 1.0m at £6.6 (Mar 2021); exercised a one-off option early and sold about 2.2m at £10.5 (Oct 2021); 16,870 deferred bonus shares awarded at £3.586 (Jul 2025, wosg-pdmr-2025) | Ralph Lauren Europe president and COO; Sara Lee; chartered accountant | | Anders Romberg, CFO | about 127k | Base £380k | Sold 1.0m at £7.2 (Apr 2021) and 341k at £14.2 (Dec 2021); bought 100k at £5.8 (Sep 2023) and 27k at £3.9 (Feb 2024); 10,684 deferred shares (Jul 2025) | Ralph Lauren CFO and COO roles | | Ian Carter, Chair | about 120k | £197k fee | Bought at £4.1, £11.5, £7.6, £5.8 and £3.7 (2020–2024), about £763k in total | Hilton president, Eataly chair | | Other NEDs | 0–33k each | £73–84k fees | Small purchases 2021–2023 | Retail, luxury and HR backgrounds | What the stake is worth: Duffy's roughly 7.7m shares at 762p (quote 3 November 2026) are about £59m. If the stock compounds at 10% a year for five years his gain is about £36m, against roughly £15m of total pay over the same period at the current level. His wealth is dominated by the share price, not by salary. Compensation metrics. Annual bonus up to 150% of salary for the CEO, 100% on adjusted EBIT with a 10% ESG underpin, one third deferred in shares for three years. The FY24–FY26 LTIP range was cumulative EPS of 178.2p–196.9p and ROCE of 22.7%–25.1%. Actual adjusted EPS was 38.0p + 41.6p + 45.2p = 124.8p (wosg-results-2024/2025/2026) and ROCE 19.5%, 19.0% and 18.0%. Both are below threshold, so on these numbers the FY24–26 award should lapse in full. That is unverified until the FY26 remuneration report is read, and it matters: key-management compensation rose from £2.1m to £4.3m in FY26 (wosg-annual-2026 note 24), which needs explaining if the LTIP lapsed. They implied far more than was delivered (FY26 adjusted EPS 45.2p is about £105m). Weighting is 80% on earnings and 20% on returns; earnings-per-share targets reward debt-funded acquisitions whose yield beats the after-tax interest rate even when it is below the cost of equity. ## Track record and references Capital-allocation record under this team, in the issuer's own numbers: - Built the US business from the 2017 Mayors entry to £927m of FY26 revenue and 62% of adjusted EBIT, through new showrooms and seven acquisitions (`04-capital-allocation.md`). - The FY24 long-range plan to double sales and profit by FY28 (wosg-results-2024) is off track: FY26 adjusted EBIT of £154.8m is below FY23's £165.1m. - Decisions that failed: the Europe venture, £44.5m of FY25 showroom impairments, Hodinkee (no material profit). There are no dividends; £25.1m was bought back. Sources also said the US team is less experienced than the UK team (p.14). Bench depth beyond Duffy and Romberg is the main gap: both came from Ralph Lauren and have worked together for nearly two decades; Duffy was born in 1955. No succession plan was found in the sources read. Dealings: Duffy and Romberg sold heavily in 2021 near the peak (£10.5 and £14.2) and Romberg and the chair bought at the 2023–24 lows. They know what the stock is worth. ## Capital allocation principles 1. **Zero-based allocation: yes.** Capital moved from Europe (nine stores divested) and the UK non-core estate (35 closures in FY25–FY26) to the US, where 65 showrooms now produce most of the profit. 2. **Funding strategies, not projects: partly.** Showroom spending follows a brand-partnership strategy, but approval is project by project on a sub-three-year pre-tax payback (wosg-results-2026). A payback rule ignores returns after year three and lease commitments that outlast it. 3. **Capital has a cost: partly.** Share awards are funded by buying stock (£28.5m for the employee trust in FY23–FY24) rather than issuing it, so compensation shows up as a cash cost, not hidden dilution. The hurdle is not stated as a cost of capital, and acquisitions are judged on "return on investment in line with our disciplined financial criteria, within an appropriate timeframe" with no number disclosed. 4. **Zero tolerance for bad growth: mostly.** Europe was exited and impairments were booked promptly. Hodinkee, which earns nothing material, is kept as a content and e-commerce platform. 5. **Price versus value: yes, on the evidence.** The only buyback (£25m, March to June 2025) bought about 6.3m shares at roughly 400p (239.57m issued shares before, 233.30m after, `sharesFy24`, `shares`), below today's 762p. No equity was issued for deals. The CEO reportedly engaged with take-private approaches in 2026 because he thought the market undervalued the company (reuters-takeover-2026, unverified). Governing objective: the framework speaks of "optimising capital deployment for the benefit of all our stakeholders, with a focus on long-term sustainable growth" (wosg-results-2026, Capital Allocation), and the long-range plan targets size (sales and profit). Value per share is not the stated objective, but the payout behaviour (buyback at low prices, no dilutive deals, insider selling at the peak) suggests management acts on it. ## Integrity and governance There is one share class with one vote per share (wosg-agm-votes), and no controlling shareholder. The issuer stays within its covenants with large headroom (net debt/EBITDA 0.28× against a 3× limit). Agency costs: - **Control of resources (empire building).** Moderate. The long-range plan, the "acquisitions" growth pillar and an 80% EPS-weighted LTIP all reward size. Acquisition multiples have been reasonable (8.8× after-tax profit measured, `07-assumption-register.md`), which limits the damage so far. - **Risk tolerance.** The CEO's wealth is concentrated in the stock, which favours conservative leverage. Leverage has indeed stayed low (peak net debt £96m, under 0.5× EBITDA). - **Time horizon.** LTIPs run three years plus a two-year hold, shorter than the brand relationships and leases that drive value. Succession is the larger horizon risk. - **Take-private conflict.** A management team negotiating with private-equity bidders may be offered equity in the new owner. If an offer comes, its independence and price need checking against value. Buybacks and compensation: of £53.6m of share purchases in FY22–FY26, £28.5m supplied employee awards and £25.1m was returned to owners, so about half the buying offset compensation. The valuation charges share compensation as a cost (£1.8m a year) and treats future awards as cash-replaced; it applies no separate agency discount. The take-private possibility is not valued. # Current investment-period assessment —12 November 2026 Current15November forward-deal bridge supersedes the blended purchase-yield prescription below: dealer-only matched consideration/profit, plus half-estate equivalent refits, yields8.13x all-in earnings and0.955x sales. Purchased targets grow only after paying additional capital. Budgets are conditional seller/integration ceilings, not historic extrapolation. [Exact inputs and effects](../updates/2026-11-15-reopened-underwriting.md). Prior measured-window comparisons below remain historical diagnostics. Read returns before and during the build. The public FY21–FY24 reconstruction gives15.4% incremental cash return; FY24–FY26 gives5.7%, or8.0% under its stated normalization. Different windows mix mature/immature assets, acquisitions and real dealer-margin changes. Recent weak group returns do not prove that every current project is poor; earlier strong returns do not guarantee the next investment. Continue funded high-quality showroom, CPO/ecommerce and supplier-approved acquisition opportunities. The revised forecast keeps total sustaining/new-capacity investment and continuing growth opex in actual cash; it shows earning power before that investment separately. Evaluate incremental cash contribution, lag, all-in price, opportunity and financing against capital cost. A short-term FCF/target miss is an observation to explain, not an automatic rejection. [Current assumptions, cash and before/after audit](../updates/2026-11-12-reassessment.md). The historical record below remains factual period evidence; current policy and valuation conclusions are in the12 November assessment. # Cash and capital allocation Reorganised and extended 3 November 2026. The five-year window is FY22–FY26, measured from the 53-week FY21 base so that both ends are 53-week years. Figures are the issuer's pre-IFRS16 APM cash-flow tables (rent is inside EBITDA as cash), GBP millions, from wosg-results-2022 to wosg-results-2026; named arithmetic is in `valuation/calculations.json` (steps `windowEbit` to `windowNetDebtChange`). ## Sources and uses of cash | | FY22 | FY23 | FY24 | FY25 | FY26 | Total | | --- | ---: | ---: | ---: | ---: | ---: | ---: | | Adjusted EBITDA (after cash rent) | 162.2 | 201.4 | 178.9 | 192.3 | 202.2 | 937.0 | | Share compensation add-back, JV | 1.7 | 3.5 | 2.1 | 2.0 | 1.8 | 11.1 | | Working capital | −29.8 | −22.5 | −20.3 | −52.2 | −4.3 | −129.1 | | Pension and tax | −16.3 | −27.3 | −34.2 | −30.4 | −24.0 | −132.2 | | **Operating cash** | 117.8 | 155.1 | 126.5 | 111.7 | 175.7 | 686.8 | | Maintenance capex (issuer label) | −3.0 | −4.6 | −2.7 | −2.8 | −3.0 | −16.1 | | Net interest | −2.7 | −4.7 | −6.2 | −11.1 | −11.0 | −35.7 | | **Issuer FCF** | 112.1 | 145.8 | 117.6 | 97.8 | 161.7 | 635.0 | | Expansionary capex | −41.0 | −67.5 | −78.0 | −72.6 | −65.9 | −325.0 | | Acquisitions (incl. contingent) and JV | −44.1 | −24.9 | −44.2 | −107.6 | −39.3 | −260.1 | | Own shares for employee awards | — | −21.3 | −7.2 | — | — | −28.5 | | Buyback for cancellation | — | — | — | −11.3 | −13.8 | −25.1 | | Dividends | — | — | — | — | — | 0 | | Exceptional cash, refinancing costs, disposals | −0.5 | −0.9 | −4.7 | −7.4 | −4.6 | −18.1 | | **Cash flow before borrowing** | 26.5 | 31.2 | −16.5 | −101.1 | 38.1 | −21.8 | | Net borrowing / (repayment) | — | — | −5.0 | 85.7 | −71.7 | 9.0 | | **Change in cash** | 26.5 | 31.2 | −21.5 | −15.4 | −33.6 | −12.8 | Reconciliation: the five annual changes sum to −12.8 (FY22 and FY23 show no borrowing line, so their cash flow equals the change in cash), and pre-IFRS16 net debt moved from £43.9m (FY21) to £57.0m (FY26), +13.1 (`windowNetDebtChange`); the difference is FX and non-cash debt movements. FY26 closing cash of £65.1m reconciles to its own opening balance (`reconciledClosingCash`). Over five years, of £635m of issuer FCF: 51% went to expansionary capex, 41% to acquisitions, 4.5% to shares for employee awards (a compensation cost) and 4.0% to the only distribution to owners, a buyback (`distributionShareOfFcf`). The company spent slightly more than it generated and never held more than £16.4m of net cash (FY23). The allocator's real priorities are new and refurbished doors, then deals, with returns to owners a residual, which matches the stated framework (wosg-results-2026, Capital Allocation: showrooms, strategic acquisitions, then buybacks or dividends from surplus). The FY26 detail: the £161.7m headline FCF is before £65.9m expansion capex and £5m exceptional cash, leaving £90.8m before acquisitions and buybacks. Charging £1.8m economic compensation and excluding £1.7m cash income gives an £87.3m trailing owner-cash diagnostic. The model then replaces £75.5m accounting rent with £83.9m of actual lease cash, keeps physical depreciation (£43.4m) and software amortisation (£2.2m) as wear, adds back only £1.3m purchase-accounting amortisation and a £0.5m disposal loss, replaces the £0.2m pension expense with £0.7m of contributions, and allows £2m a year of ongoing exceptional cash. ## Investment rates per sales change Revenue rose £922.8m from FY21 to FY26 (`windowDeltaSales`; both 53-week years; includes acquired revenue and translation). | Rate (Rappaport) | Numerator | Rate | Calculation | | --- | ---: | ---: | --- | | Incremental M&A | 259.4 acquisition cash | 28.1% | `mnaRate` | | Fixed capital | 341.1 capex − 202.4 D&A = 138.7 | 15.0% | `fixedCapitalRate` | | Working capital | net operating WC 85.1 → 259.9, +174.8 (incl. stock acquired with Roberto Coin and D&D) | 18.9% | `workingCapitalRate` | | Internally generated intangibles | e-commerce, marketing and Hodinkee app spending are expensed and not separately disclosed | not measurable | — | Every £1 of extra sales took about 62p of capital, or about 29p excluding acquisitions and the stock they brought (net capex plus cash working capital, 267.8 / 922.8). The inflection points: expansionary capex jumped from £21m (FY21) to £68–78m from FY23 as the US flagship and refurbishment programme started; acquisitions peaked with Roberto Coin in FY25; working capital turned from an outflow of £52m in FY25 (including a supplier payment-terms change) to only £4m in FY26 as underlying stock was cut. The same team (Duffy and Romberg, with Romberg away from January 2022 to 2023) ran the whole window. No peer rates were computed from sourced data. ## Reinvestment and returns Matched measurement (`windowReinvestmentRate`, `roiicFromTrough`, `roiicLaggedFromFy22`), NOPAT = pre-IFRS16 adjusted EBIT × 0.73: - Owner earning power: FY20 £40.8m (lockdown-hit, 6.9% margin), FY21 £56.6m (8.6% margin, the same as FY26's 8.5%), FY22 £95.1m, FY26 £113.0m (53 weeks; £104.4m normalised to 52 weeks with cash rent). - Growth investment FY22–FY26: net capex 138.7 + working capital 129.1 + acquisitions 259.4 = £527.2m, against matching NOPAT of £536.3m: a **reinvestment rate of 98%**. - Lagged ROIIC from FY22: (113.0 − 95.1) / £462.1m invested in FY22–FY25 = **3.9%**. FY22 starts at a boom margin (10.5%), so this is a distorted lower bound. Identity check: growth ≈ reinvestment × ROIIC. From FY22, cumulative investment of 462.1 is 4.9× FY22 NOPAT; at 3.9% that predicts 19% cumulative NOPAT growth, which is what happened (95.1 to 113.0). The identity holds by construction; the information is in the low return. Separating operating leverage: FY23 to FY26 revenue rose £285m while EBIT fell £10m, so the decline is in margin (brand margin-rate cuts, US e-commerce investment, gold-price mix, a debtor write-off) rather than in capital turnover. If margins had held at FY23's 10.7%, FY26 EBIT would have been £196m and the lagged ROIIC about 10%. The central measured group figure is about 11–12% after tax over five years (15% before FY25, 6–8% since), below the 15.3% five-year forecast ROIIC in the base model. That gap is the most important finding of this stage. The mix: organic fixed capital and stock £267.8m (51%); acquisitions £259.4m (49%); payouts £25.1m. By use, Roberto Coin (about 6× after-tax profit) has been the best measured deal; showroom refurbishments carry the under-three-year payback hurdle but their aggregate result is not disclosed, and FY25–FY26 impairments (£44.5m, £8.8m) show that some did not meet it. Capacity projects such as Old Bond Street have no disclosed IRR (01-business.md). ### Forward capital-allocation policy Offer positive cash to approved, attractively priced acquisitions after organic investment, interest, dated claims, reserves and debt obligations; distribute unplaced surplus. No acquisition borrowing is assumed. Current seller/integration ceilings and mixed-dealer/refit economics are in [the15November audit](../updates/2026-11-15-reopened-underwriting.md). The old60%deployment hold-back double-counted reserves and was retired; no second hold-back is implied by the historical funding table. ## Acquisition record Historical transactions, not the prospective dealer yield. | Closed | Target | All-in consideration | Disclosed revenue / profit | Current-rate profit used | | --- | --- | ---: | --- | ---: | | Sep 2020 (FY21) | Analog:Shift | 0.3 | £0.7m revenue, £0.4m loss part-year | 0 (undisclosed) | | Sep–Dec 2021 (FY22) | Ben Bridge, Timeless, Betteridge (5 showrooms) | 48.3 | £32.5m revenue, £5.7m net profit part-year | 10.3 (FY22 full-year 13.2 restated to FY26 US margin) | | Jun 2022 (FY23) | Bernie Robbins (1 showroom) | 21.2 | £10.5m revenue part-year; profit "not material" | 0 | | Nov 2023 (FY24) | 15 Ernest Jones/Signet showrooms | 44.2 | none | 0 | | May 2024 (FY25) | Roberto Coin Inc. (US wholesale) | 107.8 | FY25 £111.9m revenue, £23.5m PBT | 24.5 (FY26 segment EBIT, 52 weeks) | | Oct 2024 (FY25) | Hodinkee | 10.7 | £6.1m pro forma revenue; PBT not material | 0 | | Jan 2026 (FY26) | Deutsch & Deutsch (4 showrooms, 87.81%) | 40.9 incl. £8.9m minority put | £56.7m pro forma revenue, £7.6m PBT | 7.6 | Aggregate: £273.4m all-in for disclosed profit of £42.5m pre-tax, £31.0m after tax, which is **8.82×** (`measuredDealMultiple`), with undisclosed deals at zero; **6.73×** if undisclosed deals earn the disclosed showroom yield (`imputedDealMultiple`); 12.65× excluding Roberto Coin. Deployment pace £45.6m a year all-in (`dealDeploymentPace`), £43.2m in cash (`cashDealPace`). Post-integration margins: the FY22 cohort was bought at peak and its profit is restated to FY26's lower US margin; Roberto Coin's segment margin fell from 21.8% to 19.7% after a debtor write-down but its profit grew; D&D is too new. Full locators, matched definitions and caveats are in `07-assumption-register.md` (3 November 2026 section). Traps checked: the D&D figure is the seller's pro forma, not a WOSG-owned year; the FY22 "net profit" label is ambiguous between pre- and post-tax; refurbishment spent on acquired showrooms is not separately disclosed, so the true all-in price of showroom deals is higher than shown. Sources: wosg-results-2022, wosg-results-2023, wosg-results-2024, wosg-results-2025, wosg-results-2026 (APM cash-flow tables, balance-sheet summaries, capital-allocation framework); wosg-annual-2021 p.210, wosg-annual-2023 pp.188, 202, 226–227, wosg-annual-2025 pp.263–265, wosg-annual-2026 pp.225–226, 254–255. # Watches of Switzerland — current view Current15November thesis: allocation resilience, funded channel expansion and supplier-approved dealer consolidation can create durable owner cash. Current5y downside/base/upside−4.34/7.75/24.99%; weighted10.06%. This revises all joint paths, not just a bullish multiple. [Scenario definitions and premium-market sensitivities](../updates/2026-11-15-reopened-underwriting.md) govern the current model; the original reports remain historical calibration. | Horizon | Downside | Base | Upside | Weighted | | --- | ---: | ---: | ---: | ---: | | 5years | -4.34% | 7.75% | 24.99% | 10.06% | | 10years | 2.90% | 9.62% | 13.56% | 9.81% | WOSG has supplier-approved access, paid showroom capacity and a fragmented US consolidation opportunity. Brands benefit from retailer investment and service while retaining allocation and margin power. Growing pre-owned, jewellery and distribution channels can add cash profit if the stock, fit-out and customer returns justify investment. Negative FCF during productive investment is not impairment by itself. ## Pre-mortem - **Understandability:** store/stock cash can be rebuilt, but undisclosed brand allocations and cost splits make this a several-part hurdle. - **Cycle:** a durable franchise still sells discretionary goods; replacement, scarce allocations and US opportunity do not remove the cycle. - **Value trap:** revenue rises while brands capture the profit or slow stock consumes the cash. - **Productive asset:** the case rests on repeatable showroom/service and acquisition returns, rather than liquidation or a takeover. - **Agency cost:** supplier concentration, management incentives and minority puts need explicit treatment. - **Execution:** base requires competent use of funded capacity and capped profitable acquisitions; it does not assume flawless execution. - **Turnaround:** current operations are profitable. Future acquired operations receive a profit ramp, not free mature synergies at closing. - **Promises:** the H1 outcome supports the near term; long-term targets are not achieved results. - **Competition:** there is no proprietary technology preventing disintermediation; distribution rights and counterparty incentives matter. - **Concentration/fad:** scarce watches are not the whole inventory. A structural brand shift can defeat every ordinary scenario. ## Exit multiple The forecast values actual future unit profit under separately stated growth, incremental-return and funding assumptions. Excess opportunities fade by fixed2041, followed by funded2%mature nominal growth. Effective2031enterprise/cash-earnings multiples are8.70/11.24/11.97x across downside/base/upside, for the intrinsic continuations. Five-year upside instead includes the separately judged20xmaintained-equity-cash market realization; its effective enterprise/cash multiple is larger. Net financial items, claims and unseasoned acquisitions enter the engine bridge once. This is a fundamental-value continuation, not a prediction of market pricing. The [premium-market sensitivity](../updates/2026-11-15-reopened-underwriting.md) keeps operating cash/distributions fixed. A stronger operating outcome with25xequityFCF after expansion yields about20.2%five-year IRR. Missing exact source definitions do not disprove that conditional case. Premium duration, supplier returns and market pricing remain uncertainties. ## Bear case Current strong H1 is followed by two years with no additional real-volume opportunity, non-core margin/stock pressure, partial recovery and adaptive project costs. Core cash resilience remains; no below-hurdle acquisitions are forced. Full agency withdrawal, prolonged failed investment or unavailable refinancing may be worse. A model guardrail is not a legal covenant or proof of zero recovery. ## Why the opportunity exists The prior margin reset and supplier-owned retail competition explain a rational discount. No current float-turnover, short-interest or coverage study establishes neglect. No takeover premium is credited. At764.5p the modeled base and weighted returns remain below12%; improved business understanding does not by itself establish a favorable entry price. ## Variant view This research distinguishes productive investment and use of already-paid capacity from permanent deterioration. It also charges sustaining wear, stock and ongoing projects rather than equating reported FCF with earning power. At the current price, the base's12% hurdle entry is about644p. In a single-driver inverse check, holding every other base path and the fixed exit constant, a2.07-percentage-point gross-margin improvement is needed to earn12%, versus the base's unchanged terms. This is an inspectable price-implied sensitivity, not an observed consensus or an adopted forecast. Several other driver combinations could support the same price. There is no demonstrated positive mispricing against this hurdle. ## What would prove us wrong The10 December H1 accounts must distinguish actual organic/acquired sales, gross terms, setup versus routine costs, inventory, capex and minority cash. Weak mature-unit returns, allocation losses, undisclosed refit costs or poor acquisition cash would undermine the productivity case. Stronger funded unit conversion and sustained dealer economics could show the current forecast understates value. Swiss exports19 November are shipments; they cannot alone verify WOSG allocation or sell-through. The13 November Richemont original report is supportive industry context with a different brand/channel perimeter; it does not justify copying its growth or margin into WOSG. Existing frozen predictions remain unchanged for scoring. Five-year upside combines stronger operations with a conditional 20× market valuation of maintained after-interest cash; all forecast growth costs remain paid. Its IRR is 24.99%, versus 13.97% for intrinsic continuation alone. Weights are 25%/55%/20%; the market premium fades toward the independently valued ten-year continuation. [Cash definitions, evidence and full sensitivity](../updates/2026-11-15-reopened-underwriting.md). # Risks and evidence against the case Current15November: ordinary downside no longer stacks a second core margin cut with falling scarce-brand volumes. Other-brand/wholesale stock, credit, supplier rents and financing remain risks. Permanent supplier withdrawal is a severe unquantified tail;−4.34%is not a loss bound. Unspecified source detail is not contradictory evidence. [Joint paths and sensitivity audit](../updates/2026-11-15-reopened-underwriting.md). Current assessment13 November2026; financial scenarios unchanged from the [12 November reassessment](../updates/2026-11-12-reassessment.md). ## Paths to permanent loss 1. **Lost allocation or economics.** Brands may redirect stock/agencies, internalize distribution or reduce dealer margins. Mutual investment incentives support the franchise, but do not guarantee its continuation. 2. **Unproductive capital.** More showrooms, CPO stock or acquisitions can absorb cash at inadequate mature returns. The model funds continued growth and renewal; classifying spending as investment creates no profit or cash. 3. **Inventory and leases.** Slower non-scarce stock, ageing CPO and committed rents can prevent rapid adaptation. Ordinary downside assumptions are not loss bounds. 4. **Claims and financing.** D&D minorities keep their income until staged exercise, and the put is paid once. Reported1 November net debt24m is not a new full balance sheet: restricted19.9m and inventory remain carried proxies. Operating buffers, later claims and refinancing matter. 5. **Management dependence.** Brand relationships and acquisition judgment rely partly on the current team. A routine successor scenario needs evidence about the system carrying those relationships; no unsupported numerical key-person discount is added. ## Material estimates and current limitations - The growth/routine split of ordinary6m opening/closure cost is estimated at4m/2m; stress0–6m. The identified3.5m wholesale debtor charge is normalized by3.0m with a0.5m continuing credit allowance. - Brand/category profit shares, stock turns and capacity cost are estimates, not disclosed unit accounts. Group historical returns mix changes in supplier terms, investment timing and operating maturity. - Base dealer terms are unchanged; downside gross-margin pressure is1.5 percentage points with stated unit sensitivities. Future project costs and already-paid capacity can differ from these paths. - Estimated acquisitions use a finite seller budget, a0.8x sales proxy, normalized all-in cash yield and an85% first-full-year profit ramp. Unknown seller supply, refurbishment or integration costs can reduce their returns. - The base/upside12.33x NOPAT exit includes value creation beyond exit. Its duration and a current comparable peer table remain unverified. The ordinary downside uses8.33x, not a forced-sale floor. - December accounts must replace carried inventory/restricted-cash proxies and confirm actual minority/put timing. A model2x profit guardrail is not the legal covenant. ## Discriminating evidence Read the10 December original H1 gross-margin, inventory, project/cash and ownership bridge. Test actual allocations and mature/immature store profit, paid capital and acquisition returns rather than revenue momentum. Richemont's13 November results are industry evidence, not WOSG's allocation or dealer margin. Swiss export data due19 November measure shipments rather than this retailer's sell-through. Preserve `valuation/prospective-checks.json` predictions and append outcomes only when matching periods are reported. # Current assumptions and definitions 15 November 2026 reopened underwriting. Quote764.5p (£7.645),13November close; valuation date15November; fixed exits27October2031/2036. This supersedes the earlier same-day no-change article assessment. Source observations and original historical calibrations are preserved; current forecasts are independent judgments. Named arithmetic: `valuation/calculations.json`; company-owned yearly `unitVolume`, acquisition capacity/growth and terminal policies: `valuation/model.json`. [Full input/effect audit](../updates/2026-11-15-reopened-underwriting.md). | Driver | Current treatment | Evidence / uncertainty | | --- | --- | --- | | Funding |1November net debt24 plus protected restricted cash19.9; settled cash reserve and actual claim payments |H1release; gross cash/debt proxy pending10Decemberaccounts | | Starting earning power |Existing setup4of6 as investment,3of3.5credit cost normalized, full sustaining wear/cash rent and recurring closure costs retained |Original fact/estimate bridge unchanged; setup split0–6 sensitivity | | Initial growth |H1-constrained first-year country paths unchanged |FY27 revenue£1962.9m; about7.4%against53-week reported base | | Emerging channels |Pre-owned8.3%of watch sales; retail jewellery121.2; wholesale external116.7; yearly contributions within existing categories |Country CPO split50/50estimated; paid rollout and client conversion required | | Acquisition cost |Dealer-only matched profit plus explicit extra refits; all-in8.13x earnings/0.955x sales |Exclude wholesale entry; target margin, central allocation and refit share uncertain | | Owned target growth |5→4%base,8→6%up; incremental fixed/stock capital funded |Mixed dealer economics, never automatic parent scarce-brand growth | | Seller capacity |Base38.3→49.1→60;up54.7→72.3→90; cash-constrained, no borrowing |Ceilings require actual targets, brand approval and integration; zero-deal stress retained | | Ordinary adverse case |Temporary no-growth digestion, non-core margin/stock pressure; core resilience, adaptive costs |Severe supplier withdrawal remains outside representative paths, not disproven | | Exit |Actual future mix/funded continuation; upper5y20xmaintained-equity-cash realization fades toward intrinsic byyear10 |Duration/productivity and market realization uncertain; no fixed group P/E | | Weights |5y25/55/20;10y25/45/30 |Dated subjective joint paths, local edits suspend weights | Price/date/horizon held fixed in the change audit. Cash, owner-earnings, ROIC/ROIIC and chart data all come from the same yearly engine. Reported group returns, acquired cash-yield proxies and model ROIIC are different measurements. Full maintenance, expensed growth, stock, refits and claims are paid once. Current opportunities, not prior growth or a required return target, determine paths. Five-year upside combines stronger operations with a conditional 20× market valuation of maintained after-interest cash; all forecast growth costs remain paid. Its IRR is 24.99%, versus 13.97% for intrinsic continuation alone. Weights are 25%/55%/20%; the market premium fades toward the independently valued ten-year continuation. [Cash definitions, evidence and full sensitivity](../updates/2026-11-15-reopened-underwriting.md). # Forward underwriting plan 15 November 2026 reopened underwriting. Quote764.5p (£7.645),13November close; valuation date15November; fixed exits27October2031/2036. This supersedes the earlier same-day no-change article assessment. Source observations and original historical calibrations are preserved; current forecasts are independent judgments. [Full opportunity, capital and fixed-price audit](../updates/2026-11-15-reopened-underwriting.md). ## Opportunity ledger | Opportunity | Causal path | Cost / constraint | Disconfirming evidence | | --- | --- | --- | --- | | Scarce-brand throughput |Existing paid estate, named expansions, repeat clients and bounded future supply |Supplier approvals, wear, cash rent, incremental capacity/stock; price separate |Allocation loss, lower cash profit per sale, conversion deterioration | | CPO and retail jewellery |Disclosed small starting categories, rollout and cross-selling;15%base/25%up initially then fade |Do not grow whole non-core category25%; pay stock and fit-out |Stock ageing, cannibalization, weak category gross profit | | Wholesale distribution |Named boutiques/shop-in-shops and retailer reach;12%base/20%upinitial then fade |Wholesale margins/credit/stock independent of Rolex economics |Receivables losses, inventory excess, boutique under-use | | Dealer consolidation |Mixed profitable agencies; supplier-required capital can motivate sales |8.13x all-in cash earnings; cash and conditional seller ceilings bind |Higher prices/refits, no approved targets, poor acquired cohort cash | | Owned acquired growth |Customer retention/integration after first contribution;5→4%base,8→6%up |Incremental fixed capital31.0%and net working capital18.9%of new sales |Target growth only comes from buying more agencies; weak cash return | ## Scenario mechanisms Downside is ordinary investment/trading digestion, not an unsupported stack of permanent core losses. No unattractive acquisitions are forced. Terminal continuation values future unit profit, pays future reinvestment, fades excess growth by2041 and retains funded2%mature nominal growth. Premium market realizations are separate sensitivities, not proof that a higher IRR is wrong. | Horizon | Downside | Base | Upside | Weighted IRR | | --- | ---: | ---: | ---: | ---: | | 5 years | -4.34% | 7.75% | 24.99% | 10.06% | | 10 years | 2.90% | 9.62% | 13.56% | 9.81% | | Five-year operating path | 15x equity FCF | 20x equity FCF | 25x equity FCF | EV/cash earnings needed for 20% IRR | | --- | ---: | ---: | ---: | ---: | | downside | 1.77% | 7.10% | 11.54% | 31.82x | | base | 6.39% | 12.27% | 17.13% | 20.23x | | upside | 8.90% | 15.08% | 20.16% | 15.77x | ## Conditional market realization Five-year upside combines stronger operations with a conditional 20× market valuation of maintained after-interest cash; all forecast growth costs remain paid. Its IRR is 24.99%, versus 13.97% for intrinsic continuation alone. Weights are 25%/55%/20%; the market premium fades toward the independently valued ten-year continuation. [Cash definitions, evidence and full sensitivity](../updates/2026-11-15-reopened-underwriting.md). # Current capital and reinvestment definitions —15 November 2026 Current15November forecast diagnostics replace the prior forecast estimates. Store examples, acquired cash yield and group model ROIIC are distinct. Lease/capital attribution limits remain. Matched dealer/refit yield replaces the blended target proxy; each acquired growth increment pays fixed capital and working capital once. | Five-year path | Revenue | Maintained after-tax operating earnings | Actual after-interest FCF | Forecast ROIIC proxy | Period-weighted reinvestment | | --- | ---: | ---: | ---: | ---: | ---: | | downside | £2151.0m | £124.5m | £108.2m | 7.77% | 31.31% | | base | £2762.1m | £198.4m | £144.8m | 18.46% | 60.18% | | upside | £3227.3m | £259.3m | £169.5m | 21.14% | 73.73% | [Perimeters and funding reconciliation](../updates/2026-11-15-reopened-underwriting.md). Growth expense after usable tax relief enters both investment and matched before-growth earnings. Historical intangible stock is undisclosed, so the capital proxy remains partial. Examine pre-programme and later windows, mature and immature cohorts, and supplier economics separately. ## Measured history (FY22–FY26) | Measure | Value | Calculation | | --- | ---: | --- | | Growth investment (net capex 138.7, working capital 129.1, acquisitions 259.4) | 527.2 | `windowGrowthInvestment` | | Reinvestment rate (investment / matching NOPAT 536.3) | 98% | `windowReinvestmentRate` | | ROIIC FY21 → FY26, margin-neutral ends (8.6% / 8.5%) | 10.7% | `roiicFromTrough` | | Same, timing-adjusted (debtor write-off, D&D not yet earning) | 11.8% | `roiicFy21to26Adj` | | FY24 → FY26 as reported / timing-adjusted | 5.7% / 8.0% | `roiicFy24to26`, `roiicFy24to26Adj` | | From boom-margin FY22, capital lagged (distorted lower bound) | 3.9% | `roiicLaggedFromFy22` | | Issuer ROCE, pre-tax, FY23 → FY26 | 27.9% → 18.0% | `roceFy23`, `roceFy26` | ## Forecast diagnostics Current forecasts are the table above and the dashboard year-by-year cash/capital charts. Other/wholesale stock and credit differ from scarce-brand economics. Dealer-only purchase cash yield, group book-capital ROIC, partial ROIIC and store examples have different denominators and cannot substitute for one another. # Scenario probabilities 15November2026. Subjective joint-path estimates, not measured frequencies. [Definitions, old/new inputs and impact](../updates/2026-11-15-reopened-underwriting.md). | Horizon | Downside | Base | Upside | | --- | ---: | ---: | ---: | |5years |25% |55% |20% | |10years |25% |45% |30% | The prior stacked downside was not entitled to its probability merely because it was already in the model. Ordinary downside preserves current core resilience while testing digestion and non-core pressure. Base and upside now include disclosed channel opportunities, funded acquired growth and actual future profit mix. Longer-term growth/valuation uncertainty broadens credible outcomes, without automatically increasing the adverse weight. Weights approximate representative fundamental-value paths, not a complete probability distribution of all market exit multiples. Premium market realization and severe agency loss can fall outside them. No price move or desired IRR sets a weight. Local scenario changes suspend the researched weights. ## Conditional market realization Five-year upside combines stronger operations with a conditional 20× market valuation of maintained after-interest cash; all forecast growth costs remain paid. Its IRR is 24.99%, versus 13.97% for intrinsic continuation alone. Weights are 25%/55%/20%; the market premium fades toward the independently valued ten-year continuation. [Cash definitions, evidence and full sensitivity](../updates/2026-11-15-reopened-underwriting.md). # Independent bull–bear review —15November2026 ## Bull case Fresh advocate: superior allocation-led cash productivity, resilient demand, supplier-required capital and scale support repeatable value creation. Current issuer evidence supports showroom, pre-owned, jewellery and distribution opportunities. Freezing purchased earnings after integration and valuing future profits at today's business mix understated these mechanisms. Paid investment can reduce current FCF while increasing earning power. Neither a store-level return nor a waitlist proves a group return, but unspecified accounting is not a refutation. ## Bear case Fresh advocate: future dealer acquisitions must exclude the cheap wholesale entry and pay refits, stock, integration and central costs. An agency transfer does not create new production. Supplier-owned competition and rent extraction can impair value. Other-brand/wholesale inventory and credit differ from scarce watches. Future sellers and allocation are not guaranteed merely because the company has cash. A fifteen-year runway and exact probabilities remain low-precision judgments. ## Rebuttals Bull concedes dealer-only purchase perimeters, additional refit funding and target-specific growth. Bear concedes acquired businesses should grow if funded, current core cash margins were resilient, and the stacked ordinary downside lacked support. Both reject using missing fieldwork details as evidence of false claims. CPO/jewellery/wholesale contributions must be rebuilt rather than diluted into old low category rates. The upper operating schedule now reaches a low/mid-teen total-sales outcome without assuming every agency or category grows25%. ## Judgment Separate neutral judge, not the original base-case author: change the entire scenario set and test it before accepting the earlier view. Adopt funded acquired growth, dealer/refit economics, real channel contributions and actual future profit-mix valuation. Separate ordinary digestion from severe supplier withdrawal. Finite excess growth fades into paid mature nominal growth; its calendar end cannot reset at each exit. The specific10–12%discount rates and2041runway are judgments, not facts proving that the market cannot award a premium. Preserve a market-realization sensitivity and solve the exact valuation needed for a higher return. Original calibration is unchanged. | Horizon | Downside | Base | Upside | Weighted IRR | | --- | ---: | ---: | ---: | ---: | | 5 years | -4.34% | 7.75% | 24.99% | 10.06% | | 10 years | 2.90% | 9.62% | 13.56% | 9.81% | | Five-year operating path | 15x equity FCF | 20x equity FCF | 25x equity FCF | EV/cash earnings needed for 20% IRR | | --- | ---: | ---: | ---: | ---: | | downside | 1.77% | 7.10% | 11.54% | 31.82x | | base | 6.39% | 12.27% | 17.13% | 20.23x | | upside | 8.90% | 15.08% | 20.16% | 15.77x | ## Conviction Moderate in the economic mechanisms; lower in category cost allocation, future seller/refit economics, long-run allocation and exact market valuation. The revised operating upside fairly represents the stronger growth thesis. A higher return remains credible under a premium cash-flow valuation; today's fundamental continuation result is not its refutation. This open-book reassessment is neither held-out validation nor proof of investment skill. Source/evidence, decision plans and fixed-price factor effects are in [the dated review](../updates/2026-11-15-reopened-underwriting.md). Next confirmed report10December2026. ## Conditional market realization Five-year upside combines stronger operations with a conditional 20× market valuation of maintained after-interest cash; all forecast growth costs remain paid. Its IRR is 24.99%, versus 13.97% for intrinsic continuation alone. Weights are 25%/55%/20%; the market premium fades toward the independently valued ten-year continuation. [Cash definitions, evidence and full sensitivity](../updates/2026-11-15-reopened-underwriting.md). ## Public Primary & Secondary Sources - `wosg-annual-2026` (issuer annual report): Watches of Switzerland Annual Report and Accounts 2026 — https://www.thewosgroupplc.com/media/ix2hz00d/wosg-annual-report-2026.pdf (published 2026-07-31) - `wosg-results-2026` (issuer results): WOSG FY26 results, 53 weeks to 3 May 2026 — https://www.thewosgroupplc.com/media/spwhxg4u/wosg-fy26-results-statement-final.pdf (published 2026-07-14) - `wosg-strategy-2026` (issuer strategy): WOSG Strategy Deep Dive July 2026 — https://www.thewosgroupplc.com/media/4v0nzw14/wosg-deep-dive-140726-final.pdf (published 2026-07-14) - `wosg-agm-2026` (issuer trading update): WOSG AGM trading update, 3 September 2026 — https://www.thewosgroupplc.com/media/1xah51o3/wosg-agm-trading-update.pdf (published 2026-09-03) - `wosg-calendar` (issuer calendar): WOSG financial calendar — https://www.thewosgroupplc.com/investors/financial-calendar/ - `wosg-ir` (issuer announcements): WOSG financial reports and current announcements — https://www.thewosgroupplc.com/investors/ - `wosg-director-2026` (issuer RNS reproduced): WOSG 23 October 2026 update on director details — https://6ix.com/news/directorate-change-1285 (published 2026-10-23) - `fh-september-2026` (trade association): Federation of the Swiss Watch Industry September 2026 statistics — https://www.fhs.swiss/eng/statistiques - `rolex-bucherer` (supplier disclosure): Rolex acquisition of Bucherer — https://newsroom.rolex.com/about-rolex/rolex-acquires-bucherer (published 2023-08-24) - `richemont-q1-fy27` (issuer counterparty results): Richemont Q1 FY27 original issuer sales release — https://www.globenewswire.com/news-release/2026/07/15/3327455/0/en/Richemont-posts-strong-start-to-the-year-with-sales-up-by-20-at-constant-rates-for-its-first-quarter-ended-30-June-2026.html (published 2026-07-15) - `capital-method-roiic` (methodology): Mauboussin and Callahan: Return on Invested Capital — https://www.morganstanley.com/im/publication/insights/articles/article_returnoninvestedcapital.pdf (published 2022-10-06) - `capital-method-cash` (methodology): Dan Shuart, Eagle Point Capital: A Framework For Spotting Value Traps — https://eaglepointcapital.substack.com/p/spotting-value-traps-the-anti-compounders (published 2023-08-26) - `scenario-method` (methodology): Aswath Damodaran: Probabilistic Approaches in Valuation, chapter 33 — https://pages.stern.nyu.edu/~adamodar/pdfiles/val3ed/c33.pdf - `wilsdorf-foundation` (counterparty owner primary disclosure): Fondation Hans Wilsdorf: ownership and objectives — https://hanswilsdorf.ch/en/the-foundation/ - `rolex-bulle-2026` (counterparty primary disclosure): Rolex 2026 recruitment announcement: new Bulle site — https://www.carrieres-rolex.com/content/Zoom-comptoir-gruerien-2026/ - `wosg-annual-2025` (issuer annual report): Watches of Switzerland Annual Report and Accounts 2025 — https://www.thewosgroupplc.com/media/3xlo5mcw/wosg-annual-report-and-accounts-fy25.pdf (published 2025-07-31) - `wosg-annual-2023` (issuer annual report): Watches of Switzerland Annual Report and Accounts 2023 — https://www.thewosgroupplc.com/media/oukb2xlx/wos-ar23-web.pdf (published 2023-07-12) - `wosg-annual-2021` (issuer annual report): Watches of Switzerland Annual Report and Accounts 2021 — https://www.thewosgroupplc.com/media/0iwl15d0/6426-wos-ar21-21-07-30-web.pdf (published 2021-07-30) - `wosg-dd-2026` (issuer annual report section): WOSG Annual Report 2026: Deutsch & Deutsch FY26 progress — https://www.thewosgroupplc.com/media/ix2hz00d/wosg-annual-report-2026.pdf (published 2026-07-31) - `wosg-results-2022` (issuer results statement): WOSG FY22 results, 52 weeks to 1 May 2022 — https://www.thewosgroupplc.com/media/nfkptm3v/final-full-rns.pdf (published 2022-07-07) - `wosg-results-2023` (issuer results statement): WOSG FY23 results, 52 weeks to 30 April 2023 — https://www.thewosgroupplc.com/media/du5hatic/wosg-fy23-final.pdf (published 2023-07-13) - `wosg-results-2024` (issuer results statement): WOSG FY24 results, 52 weeks to 28 April 2024 — https://www.thewosgroupplc.com/media/qucdqw5k/fy24-wos-results-final.pdf (published 2024-06-27) - `wosg-results-2025` (issuer results statement): WOSG FY25 results, 52 weeks to 27 April 2025 — https://www.thewosgroupplc.com/media/qbkp2c0z/fy25-wosg-results.pdf (published 2025-07-03) - `wosg-agm-votes` (issuer AGM result announcements): WOSG results of AGM: 14 Oct 2020, 2 Sep 2021, 1 Sep 2023, 3 Sep 2024 — https://www.thewosgroupplc.com/media/xmqai2om/agm-2024-results-030924.pdf (published 2024-09-03) - `wosg-pdmr-2025` (issuer PDMR notification): WOSG PDMR notice, deferred bonus share awards 25 July 2025 — https://data.fca.org.uk/artefacts/NSM/RNS/5755743.html (published 2025-07-28) - `reuters-takeover-2026` (news report (Reuters, via republisher)): Reuters: Watches of Switzerland held talks over potential takeover offers, sources say (13 July 2026) — https://www.globalbankingandfinance.com/exclusive-watches-switzerland-held-talks-over-potential/ (published 2026-07-13) - `fca-shorts-wosg` (FCA net short position disclosures (republished)): WOSG disclosed net short positions (FCA data republished by Investegate) — https://www.investegate.co.uk/company/WOSG/short-positions (published 2026-04-10) - `bucherer-tourneau-2018` (trade press report): aBlogtoWatch: Swiss retailer Bucherer acquires Tourneau LLC (2018) — https://ablogtowatch.com/swiss-retailer-bucherer-acquires-tourneau-llc (published 2018-06-01) - `wosg-h1-fy27-tu` (issuer trading update (RNS)): WOSG H1 FY27 Trading Update, 26 weeks to 1 November 2026 — https://www.investegate.co.uk/announcement/rns/watches-of-switzerland-group--wosg/h1-fy27-trading-update/9866387 (published 2026-11-10) - `wosg-h1-fy27-original-pdf` (primary): WOSG original H1 FY27 trading update PDF — https://www.thewosgroupplc.com/media/fvknqdcx/h1-fy27-trading-update.pdf (published 2026-11-10) - `richemont-h1-fy27` (issuer counterparty results): Richemont H1 FY27 original issuer interim results release — https://www.richemont.com/news-media/press-releases-news/richemont-posts-strong-results-for-the-six-month-period-ended-30-september-2026-driven-by-sustained-double-digit-sales-growth/ (published 2026-11-13) - `richemont-fy27-h1-transcript-2026-11-13` (primary): Richemont FY27 interim results presentation transcript — https://www.richemont.com/media/dldixyzg/richemont-fy27-interim-results-presentation-transcript.pdf (published 2026-11-14) ---