Adyen N.V. (Euronext Amsterdam: ADYEN.AS) — The Single-Stack Compounder the Market Re-Rated From Network to Acquirer, Cheapest Since It Listed

An independent equity-research note. All figures IFRS, in euros (€), unless stated. Price reference: €824.9 close, 2026-06-26. Fiscal year ends 31 December. Adyen is a foreign issuer with no SEC filings; primary sources are Adyen’s half-year shareholder letters, its annual report, and Euronext/AFM disclosures.

⚡ Claude’s Take

This block is the author’s own subjective opinion and general information only — not investment advice and not a recommendation. The detailed analysis that follows takes no position, names no price target, and carries no buy/sell; the single directional view in this piece is fenced inside this clearly-labeled block.

Verdict: HOLD / accumulate-on-weakness — a BUY-quality business at a now-reasonable price, but not yet a fist-pounding entry. Medium conviction. Directional fair-value zone ~€950–1,200 (≈24–28x reported / ~28–32x ex-float FY26 EPS for a sustained high-teens compounder); the €750–900 band where it trades is where you are paid to hold the bear case; rich above ~€1,400; value-trap floor ~€600–700.

Adyen is the best-architected business in payments — one global codebase, built entirely in-house, no acquisitions, no goodwill, its own banking licenses, ~88% gross margins on net revenue, 50%+ incremental operating margins, ~100% cash conversion, and a fortress balance sheet — and the market has just repriced it from a ~150x-earnings ZIRP darling to ~24x, the cheapest multiple since its 2018 IPO, even as net revenue still compounds ~20% in constant currency. That is the inverse-of-richest setup: a genuine quality compounder de-rated rather than broken. A close public-market analog is Dynatrace (a compounder left for dead), crossed with the competitive-share caveat of PayPal — and explicitly not a value trap like Fiserv, which is organically shrinking while Adyen still grows ~20%. At ~24x earnings the price embeds only ~14% terminal EPS growth (PEG ~0.6–0.7 on near-term guidance); the market is underwriting convergence to a sub-scale acquirer, and on the balance of evidence that is too pessimistic.

What keeps this a HOLD rather than a table-pounding BUY is that the bear case is partly real, not imaginary. The take rate has structurally compressed from ~22bps to ~16–17bps as Adyen rode price-sensitive mega-merchants; Stripe — now privately valued at $159bn, roughly six times Adyen’s public multiple on comparable scale and faster growth — contests it on capability, not just price; the 2023 loss of US large-merchant volume to Braintree proved the franchise competes on price at the top of the market; ~21% of EPS is rate-sensitive float income now rolling over as the ECB eases; and the tape is an intact downtrend below all major moving averages — a knife that has slowed but not turned. Layer on a surprise CFO departure (effective 31-Aug-2026, no successor named) and a management that openly finds the market’s short-term anxiety “difficult to understand,” and the prudent posture is to accumulate into weakness rather than chase. Framing: contrarian/quality-at-a-discount, not deep value and not momentum. Conviction flips bullish on a first-ever buyback at this multiple plus sustained 20%+ cc prints with a stable-to-rising take rate; it flips bearish if the take rate breaks below ~16bps, a marquee merchant is publicly lost to Stripe, or 2026 growth guides below ~18%. Tag: “Rolls-Royce engineering, marked down to a Ford multiple — buy the dips, not the rip.”

📈 Stock Price Action — Five-Year Event Map

Text-only by design. The price move is a Fact; the attributed cause is Interpretation. No price target, no support/resistance, no chart-pattern reading.

The arc. Adyen listed on 2018-06-13 (€240 offer, opened €400, closed €475), rode the pandemic-e-commerce-plus-zero-rates boom to an all-time high of €2,766 on 2021-08-24, round-tripped to a €631 trough on 2023-09-21 after a single-day −39% crash, recovered to €1,846 on 2025-02-18, then de-rated through 2025–26 to a fresh low of €785 on 2026-06-11, and trades at ~€825 today — roughly −70% off the all-time high, below its 21-, 50- and 200-day moving averages, in a 52-week close range of €785–€1,583.

Cycle narrative. (1) The IPO-to-ATH +482% was a real digital-payments boom amplified by zero rates inflating the multiple to nosebleed levels — growth was genuine, but 90–220x earnings was a regime artifact. (2) When rates rose, the multiple compressed −58% with the whole unprofitable-growth complex, before any operating miss. (3) The defining event was 2023-08-17’s −39%: H1’23 paired a growth deceleration with a self-inflicted margin collapse (headcount surged from ~3,300 toward ~4,000+) and the revelation that Adyen had lost North American large-merchant volume to cheaper rivals — the first hard evidence the franchise competes on price at the top of the market. (4) Capitulation carried it to €631. (5) The November-2023 Investor Day (+37.8%) restored credibility by committing to margin recovery and hiring discipline, which (6) the company delivered through 2024 into a €1,846 high. (7) The fresh re-derate came 2026-02-12 (−21.9%) when the FY2025 print and a 2026 guide trimmed to 20–22% cc crystallized the market’s three fears — take-rate compression, Stripe-led share competition, and float-income rollover. (8) The drift to €785 by June 2026 added a key-person overhang (the surprise CFO departure), leaving the stock at the cheapest multiple in its public life.

1. Executive Summary

Adyen N.V. is a Dutch, Amsterdam-listed payments company that operates a single global financial platform — gateway, acquirer, processor, risk engine, card issuer, and licensed bank — built entirely in-house on one codebase with zero acquisitions in its history. It earns “net revenue” (the slice it keeps after passing interchange and scheme fees to issuers and to Visa/Mastercard) on the volume it processes for a concentrated book of very large, mostly digital-native merchants — Uber, Spotify, Microsoft, McDonald’s, Meta, eBay, LVMH, and newer wins Starbucks and Temu.

The numbers are those of a high-quality compounder downshifting from hyper-growth. FY2025 net revenue was €2,364m, +18% reported / +21% constant-currency, on €1,394bn of processed volume (implied take rate ~17bps). Reported EBITDA was €1,246m (53% margin); operating income (EBIT) €1,096m (46.6%); net income €1,062m, diluted EPS €33.61. The business is genuinely capital-light — capex ~5% of net revenue, SBC ~1.7%, ~100% normalized cash conversion, no debt, no goodwill, ~€167/share of tangible book — and earns a ~26% ROE that is declining only because it hoards cash. The single most important caveat to the earnings: roughly 21% of EPS is float income (interest earned on merchant settlement balances under the banking license), which is rolling over as the ECB cuts; ex-float diluted EPS is ~€26.4.

The investment tension is sharp and singular. Adyen is one of the best businesses in payments, yet it has been re-rated from ~150x earnings in 2021 to ~24.5x today — the cheapest multiple since its 2018 IPO — while still compounding ~20%. At that price the market embeds only ~14% terminal EPS growth (PEG ~0.6–0.7 on guidance), effectively underwriting convergence toward a sub-scale acquirer multiple. The bull case is that Feb-2026’s −22% was a multiple overshoot handing a quality compounder to buyers at a trough; Unified Commerce (+33% cc) and Platforms (+49% cc) are accelerating, North America has re-accelerated >30%, and there is unpriced optionality in embedded finance and a possible first buyback. The bear case is that the take rate is structurally compressing (~22bps → ~16–17bps), Stripe (a $159bn private mark) contests Adyen on capability, the 2023 share loss proved the large-merchant layer is price-competitive, float income is fading, and management is dismissive of shareholders into a CFO transition. This note argues the franchise is intact and the valuation reasonable-to-attractive, but that the deceleration and competitive risks are real enough to keep the conclusion measured rather than emphatic.

2. Business Overview

What Adyen is. Adyen is best understood not as a “payments app” but as one global financial operating system that a merchant integrates once and then uses to accept money in any channel, currency and geography. It sits between large merchants and the card networks as a combined gateway, acquirer, processor, risk engine, issuer and — uniquely — a licensed bank. The defining architectural fact, and the foundation of the entire bull case, is that Adyen built this system 100% in-house, on a single codebase, with effectively zero acquisitions and zero goodwill on the balance sheet. This is the deliberate inverse of the incumbent model (Fiserv, Global Payments, FIS/Worldpay), which assembled scale by acquisition and now runs fragmented, stitched-together stacks. Adyen’s argument is that one stack, one data model and one ledger across 30+ countries is both a cost advantage (no integration tax, no duplicate systems) and a capability advantage (a global merchant gets unified reporting, cross-border routing and one settlement relationship). Whether that is a durable moat or merely a nicer architecture competitors can also build is the subject of the Competitive Position section.

How it makes money — and the metric that governs all analysis. Adyen charges merchants a processing fee plus a small margin over the interchange and scheme fees it passes through to issuers and to Visa/Mastercard. The single distinction that governs every ratio in this report: gross “settlement” volume — the full ticket, including interchange/scheme costs passed straight through — is NOT revenue. Adyen’s true top line is net revenue, the slice it actually keeps. (Reported IFRS “revenue” before 2023, e.g. €8.9bn in FY2022, is gross settlement and produces meaningless single-digit “margins”; we use net revenue throughout, which corresponds to the company’s headline figure.) In FY2025 net revenue was €2,364m on processed volume of €1,394bn — an implied blended take rate of ~17 basis points. That take rate is the most important operating ratio in the business, and its trajectory is double-edged.

The take-rate trend — secular compression, recent stabilization. Over the platform’s life the blended take rate has fallen roughly a quarter to a third — from ~22bps in 2018 (€349m net revenue on ~€159bn volume) to ~16–17bps today — because Adyen’s growth has been powered by very large, price-sensitive digital merchants who command steep volume discounts. This is the mathematical signature of “land-and-expand into mega-merchants”: volume compounds faster than net revenue, so the rate bleeds. Crucially, that compression recently reversed at the margin — H2-2025 take rate was ~17.1bps versus ~16.2bps in H2-2024 — but the uptick is a mix effect: the lowest-take digital volume (notably eBay, rolling off as a customer) is shrinking as a share, while higher-take in-person (Unified Commerce) and Platforms volume grows. The honest read: the secular pressure of large-merchant discounting is real and structural, currently offset by deliberate mix shift; whether that offset is durable or a temporary eBay base-effect is a live question (see Variant Perception).

The four growth pillars. Adyen reports along product pillars; the H2-2025 split is highly informative about where the franchise actually grows:

The mix tells the story: the mature Digital pillar (~55% of revenue) is now a ~10% grower (with volume actually −1% on the eBay roll-off), while Unified Commerce (+33%) and Platforms (+49%) do the heavy lifting. Unified Commerce is the genuine differentiator — Adyen lets a global retailer run online + physical store + app on one platform with one view of the shopper; point-of-sale volume reached €311bn in FY2025 (+34%). Platforms (Adyen for Platforms) embeds Adyen’s acquiring, issuing and lending inside software companies and marketplaces — the embedded-finance land grab Stripe also chases. Issuing/Banking-as-a-Service/Capital is still tiny but is the optionality leg (issuing volumes up ~8× off a small base).

Customers and “land-and-expand.” Adyen targets the enterprise and digital-native top of the market — a concentrated book of very large names rather than a long SMB tail. Marquee customers include Uber (live in 70+ countries on Adyen), Spotify, Microsoft, McDonald’s, Meta, eBay and LVMH, plus newer wins Starbucks (rolled to 900+ European stores in seven weeks) and Temu. The model is to land a single use-case, then expand across channels, geographies and products; net revenue retention has historically run above 100% (roughly 110–120%), the quantitative proxy for the expand motion working. The risk embedded in this strategy is concentration and price exposure: a handful of mega-merchants drive a disproportionate share of volume, they have negotiating leverage, and — as 2023 proved — they will move volume to a cheaper processor at the margin.

Verdict. A high-quality, capital-light, organically-built global payments franchise whose revenue is recurring-in-character (it recurs as long as merchants keep processing) and whose growth engine has rotated from a maturing digital core to faster in-person and platform pillars — but whose unit economics (the take rate) are under genuine, quantified secular pressure that mix shift is currently, but not necessarily durably, masking.

3. Industry Dynamics

The value chain and where Adyen sits. Card payments split into three economic layers, and the profit pool is wildly unequal across them. (1) The networks (Visa, Mastercard) — a rational, two-decade-stable duopoly earning ~50% net margins and ~50%+ ROIC on true two-sided network effects; they set the rails, take a rising toll, and are not Adyen’s competitors — they are its suppliers. (2) Issuer processing — the back end running banks’ card portfolios (Fiserv, FIS, the divested Global Payments Issuer unit) — a sticky, slow-growth annuity. (3) Merchant acquiring / PSP — Adyen’s home — historically the most competitive, lowest-margin, most capital-flooded layer. The most important structural fact in payments, evident across the listed networks and acquirers (Visa, Mastercard, Global Payments), is that durable pricing power sits at the network layer, not the acquirer layer — which is why networks trade at 25x+ earnings and acquirers (GPN ~5x, Fiserv ~7x) trade in single digits. Adyen’s bull case is, in effect, that it has built the one acquirer good enough to escape the gravity of its own layer.

Secular tailwinds (the demand side). The demand backdrop is genuinely good and durable: cash→digital conversion still has a long runway globally; e-commerce continues to take retail share; unified commerce (merchants wanting one platform across online + store) is a real and accelerating buyer preference that favors a single-stack provider; and embedded payments/finance (software platforms monetizing payments inside their products) is a structural land grab. These are exactly the segments where Adyen’s Unified Commerce (+33%) and Platforms (+49%) pillars grow fastest. But — channeling Marathon’s capital-cycle lens — strong, well-advertised demand growth is precisely what attracts the capital that erodes returns. Demand tailwinds are not a moat; everyone in payments reads the same TAM slides.

Competitive intensity — capital is flooding in. The acquiring/PSP layer is crowded and getting more so:

• Stripe (private, $159bn February-2026 mark) is the most dangerous competitor — born developer-first/SMB, now pushing aggressively up-market into the enterprise/global accounts Adyen owns, with ~$1.9T processed volume (+34%) and a similar single-stack, in-house architecture. Stripe is the one rival that contests Adyen on capability, not just price.
• PayPal / Braintree is the proven enterprise price-competitor and the protagonist of Adyen’s 2023 trauma.
• Checkout.com, Block/Square (SMB + Cash App), Toast (vertical/restaurants), Nuvei, Worldpay (now standalone under GTCR — see the GPN report), and Fiserv/Clover round out a deep field, plus regional acquirers everywhere.

The competitive evidence that matters most: in 2023, Braintree/PayPal won North American volume from Adyen by discounting aggressively, and that share loss — alongside a margin-crushing hiring spike — drove the −39% single-day crash. That episode is the empirical proof that large-merchant acquiring is price-competitive: even Adyen’s vaunted platform did not stop a determined competitor buying volume on price. Adyen’s counter — that it has since re-accelerated NA (>30% ex-FX in H2-2025) by selling unified commerce rather than competing on raw online price — is credible and important, but it concedes the underlying point: the moat is in the integrated offer, not in the commodity online-acquiring rate.

Marathon capital-cycle read. Payments processing is mid-to-late in a capital boom. The signs are textbook: a decade of capital — VC into Stripe/Adyen/Checkout, PE into Worldpay/Nuvei, strategic M&A (the 2026 GPN–Worldpay deal) — has flooded the acquiring layer; sell-side coverage expanded; take rates are compressing industry-wide; and the 2025–26 reshuffle of acquiring assets (GPN buying Worldpay, FIS selling it, GPN selling Issuer to FIS) is the classic late-cycle signature of incumbents buying scale because organic returns compress. Against that, the supply side of Adyen’s specific niche — enterprise unified commerce + platforms/embedded finance — is far less commoditized than vanilla online acquiring, because it requires capabilities (one global stack, banking licenses in 30+ jurisdictions, in-person hardware + online + issuing under one roof) that take years and hundreds of millions to build and cannot be bought off the shelf. The capital cycle is eroding returns at the commodity end of the layer (where Adyen is exposed via large digital merchants and the secular take-rate decline) while leaving its integrated niche relatively protected — for now.

Regulation. The regulatory vector is one-directional against the spread acquirers can hold: interchange settlements, debit-routing rules, proposed credit-card-competition legislation, and open banking (which threatens to disintermediate cards with account-to-account rails). Lower merchant card costs pressure the bps acquirers capture. Adyen is partly hedged — as a licensed bank and direct network member it captures more of the chain than a thin reseller, and it builds A2A/local payment methods itself — but the secular direction is margin-unfriendly for the layer.

Verdict — a structurally mixed industry: bad layer, good niche. Merchant acquiring as a layer is a structurally below-average place to sit: capital-flooded, weak-moated at the commodity end, regulation-pressured, take rates compressing. But Adyen does not operate at the commodity end — it operates in enterprise unified commerce and embedded-finance platforms, a genuinely high-barrier niche with strong, durable demand tailwinds and a far thinner competitive set (really only Stripe at capability parity). Structurally attractive for a top-two single-stack global platform; structurally unattractive for everyone selling acquiring as a near-commodity. The risk is that the boundary between those two worlds is porous — 2023 showed mega-merchants will treat even Adyen’s offer as substitutable on price when a rival discounts.

4. Competitive Position

Greenwald taxonomy — naming the moat. Adyen’s competitive advantage, if real, is a combination of two of Greenwald’s three genuine types, reinforced by a data-network effect: (1) an economies-of-scale / supply-cost advantage from the single global in-house platform, and (2) customer captivity via switching costs from deep enterprise integration. The pure demand sources (habit, search costs) barely apply — enterprise payments is a considered, RFP-driven purchase. We pressure-test each against the only test that matters: if this advantage disappeared, what financial outcome would deteriorate?

(1) The single-platform scale/cost advantage — real but contestable. Adyen runs one codebase, one data model and one ledger across all geographies and channels, with no acquired-system integration tax. The financial fingerprint is visible: ~46–47% operating EBITDA margins, ~5% capex intensity, and the ability to onboard Starbucks across 900+ stores in seven weeks — operating leverage and deployment speed the acquisition-built incumbents (Fiserv, FIS/Worldpay, GPN) structurally cannot match because they carry fragmented stacks and goodwill. This is a genuine supply advantage rooted in proprietary engineering and a decade-plus learning curve, plus an economies-of-scale element (the fixed cost of the global platform spread over €1.4tn of volume). But Greenwald is explicit that technology/supply advantages are the weakest and most transient, and that scale is only a durable barrier when combined with customer captivity. The fatal qualifier: Stripe has built the same kind of single in-house stack at comparable volume. When a second well-capitalized entrant replicates the architecture, it is not a barrier to entry — it is table stakes for the top tier. The platform advantage is real against the legacy incumbents; it is parity, not advantage, against Stripe.

(2) Switching costs — the strongest leg, but it protects the integration, not the price. A global enterprise that has integrated Adyen across online, in-store, app, issuing and settlement in 30 countries faces a real, multi-quarter, risk-laden project to rip it out. That captivity is genuine and is the best explanation for >100% net revenue retention (~110–120%) and the stickiness of the enterprise book. But the 2023 Braintree episode is the crucial disconfirming evidence: mega-merchants did move material North American volume to a cheaper processor. The resolution of that paradox is the key insight for the whole thesis — switching costs protect Adyen’s seat at the merchant (the relationship is sticky) but they do NOT protect the take rate (the price is re-negotiable, and large merchants multi-source). So the moat caps churn risk but not margin risk; it is why retention stays >100% even as the blended take rate compresses ~25–30% over time. A real but bounded moat.

(3) The data-network / authorization-rate advantage — plausible but unquantified. Adyen’s pitch is that processing €1.4tn across thousands of merchants gives it cross-merchant data to optimize routing and lift authorization rates (the “Adyen Uplift,” “Intelligent Payment Routing,” “RevenueAccelerate” products) — i.e., it makes more of the merchant’s attempted transactions succeed, worth far more than a basis point of price. If real and durable, this is the most attractive part of the moat because it is a genuine scale-driven data advantage that improves with volume. But it is the least quantified: Adyen publishes no credible, audited auth-rate uplift versus Stripe, and Stripe (and the networks themselves, via tokenization) make the identical claim with comparable scale. Treat it as a real but unproven and probably-not-unique advantage — an open question, not an established moat.

(4) Banking licenses + direct network membership — the most under-appreciated, most durable leg. Adyen holds its own banking license (EU/DNB) plus acquiring licenses across the US, UK and 30+ jurisdictions and is a direct principal acquiring member of Visa/Mastercard, letting it own the full chain — gateway → acquiring → issuing → settlement → banking — with no third-party intermediaries, and to layer on issuing/BaaS/lending. This is the hardest leg to replicate: assembling banking and acquiring licenses across so many jurisdictions is a multi-year regulatory slog most competitors will not undertake. It is a genuine, government-protection-adjacent barrier and, more than the codebase, the part of the moat we underwrite as durable.

Greenwald tests applied. ROIC test: sustained ROE ~26–31% and ROIC ~16% over a decade clears Greenwald’s “advantages present” bar (>15%) decisively — this is not a commodity 6–8% business. Market-share-stability test: here the moat is weaker than bulls claim. Logo retention is high and Adyen keeps winning marquee accounts, but it visibly lost North American share to Braintree in 2023 — not the <2% drift Greenwald associates with formidable barriers, but the >5% movement that signals contestability. Adyen passes the profitability test emphatically and the share-stability test only partially.

Marathon overlay — returns are mean-reverting, and the numbers show it. The capital cycle is already visibly compressing Adyen’s economics: EBITDA margin fell from a ~59% peak (2021) to a ~42% trough (2023) before recovering to ~53%; ROE is sliding from 35% toward 26%; the blended take rate has compressed ~25–30% over the platform’s life; and 2026 revenue guidance was cut from “mid-twenties” to 20–22% cc. This is exactly what Marathon predicts when capital floods a high-return industry: the direction of nearly every key return metric (except the recovering margin) is down. The bull rebuttal — that this is deliberate investment and mix shift, not competitive erosion — is partly right (the margin recovery and the UC/Platforms acceleration are real), but the burden of proof sits with the bulls to show the take-rate stabilization and 20%+ growth are durable rather than the early innings of mean reversion.

Verdict — a durable advantage that is narrower and more price-exposed than the bull case claims. Adyen is not a commodity processor with a nice UI — the ROIC, license breadth, in-house single stack and >100% retention place it in the top tier. But it is not unassailable either. The moat is (a) genuinely wide against the legacy incumbents, (b) merely at parity against Stripe, © protective of the merchant relationship but not of the take rate, and (d) reliant on a “data network” auth-rate advantage asserted more than proven. The correct characterization: best-in-class, narrowly-moated single-stack platform competing in a structurally pressured layer, with one true peer (Stripe) and a take rate that secular forces and aggressive competitors can compress. Durable advantage: yes — but bounded, and currently mean-reverting.

5. Growth History and Forward Opportunities

The historical record: a ~31% compounder downshifting to the low-20s. Net revenue has compounded at roughly 31% over seven years: €349m (2018) → €534m → €684m → €1,002m → €1,330m → €1,626m → €1,988m → €2,364m (FY2025). But the trajectory is unmistakably maturing: after +46% (2021) and +33% (2022), growth settled at +22% (2023), +22% (2024) and +18% reported / +21% cc (2025). Q1-2026 printed €620.8m, +16% reported / +20% cc — the bottom of the 20–22% cc guide, with the reported figure dragged ~4 points lower by a weak US dollar. The deceleration is real, but the gap between high-teens reported and low-20s constant-currency is the framing bears under-weight: the underlying engine is still a 20% grower; FX is doing much of the optical damage.

Decomposing the growth — the share-of-wallet engine. Management is explicit that “the biggest part of our growth in any given year comes from growth with our existing customers” — the land-and-expand / net-revenue-retention motor running above 100% as merchants add markets, channels and products on one stack. The 2025 new-logo cohort was, in management’s words, “the strongest we’ve seen.” The third leg — same-merchant market volume — is the cyclical wildcard, and in 2025 it landed at the low end of the framework as a handful of APAC cross-border online retailers (the Temu/Shein-type discounters caught by tariff and de minimis changes) softened. So the 2025 algorithm was: solid existing-customer expansion + a strong new cohort − weak market volume − FX. Higher quality than the headline suggests, but it exposes the two soft spots: consumer-cyclical volume and FX translation.

Growth by geography — North America is the re-acceleration story. After ceding large-merchant US/APAC volume to cheaper rivals in 2023, Adyen rebuilt North America into its single biggest investment market, growing >30% ex-FX in H2-2025; management ties the strength to a decade of investment plus a US market “only getting more complicated” (payment-method fragmentation, the unified-commerce imperative). EMEA, the mature base, decelerated to ~17% on lumpy ramp timing. The emerging frontier is genuinely promising: LatAm was the fastest-growing region in H2-2025 on cc, and Japan and India have entered the “domestic phase” — Adyen’s playbook is to enter a country with its global merchants, then convert domestic logos, and management says both have reached that inflection. These are multi-year runways, not 2026 needle-movers.

Forward opportunities — five vectors, in rough order of proven traction. (1) Unified Commerce / in-person POS — Adyen entered physical retail late and has large runway; Starbucks, McDonald’s and Uber airport kiosks evidence the land grab. (2) Platforms & embedded finance — the +49% pillar, where Adyen’s combined online + in-person + financial-products offering “future-proofs” software platforms; the most direct Stripe contest. (3) Issuing + Capital + bank accounts — embedded lending/BaaS optionality; management flags “a nice inflection in issuing” and “stronger traction in capital and bank accounts,” all tiny today. (4) AI / “Dynamic Identification” — a foundational data layer over “trillions of interactions,” feeding the Uplift suite (Personalize pilots cited at +6% conversion at −3% merchant cost), RevenueAccelerate and fraud detection; ~two-thirds of new merchants now activate parts of Uplift from the start — a real wallet-share lever. (5) Agentic commerce — protocol work with OpenAI, Google, Visa and Mastercard; explicitly “immaterial” today and “not a big part of 2026,” a 2027-28+ option that is more defensive than a near-term driver.

Medium-term targets (current). 2026 net revenue growth 20–22% cc; 2026 EBITDA margin “broadly in line with 2025” (~53% on Adyen’s reported, float-inclusive definition; ~47% on the operating measure); EBITDA margin >55% by 2028; capex ≤5% of net revenue. Management reaffirmed the Nov-2023 framework of “~20% growth for the upcoming years.” Note the margin target is a step-up but a gentle one — Adyen is deliberately reinvesting (team growing a few points faster than 2025, weighted to US tech hubs and financial-products specialists), so 2026 is not a margin-expansion year.

Verdict — high-quality growth, but maturing and increasingly cyclical/FX-exposed. This is genuinely high-quality growth: overwhelmingly organic (zero acquisitions ever), driven by >100% net-revenue retention and a strong new-logo engine, capital-light, with credible multi-year runways in UC, Platforms, North America, Japan/India/LatAm and AI-driven authorization uplift. But the quality cannot disguise the downshift from ~31% to 20–22% cc / high-teens reported, with a 55%-of-revenue Digital core growing only ~10% (and shrinking in volume), consumer-cyclical exposure, meaningful FX drag, and growing reliance on still-unproven new products to keep the consolidated rate near 20%. High quality, lower altitude.

6. Financial Quality

Read the top line correctly or every ratio is wrong. Built on net revenue, the model is one of the cleanest growth profiles in payments:

*Pre-2023 EBIT-margin figures are distorted by the gross-revenue denominator and are not comparable; the EBIT margin on net revenue is the relevant figure from 2023 once presentation normalized. Net-revenue CAGR 2018–2025 ≈ 31%. Gross margin on net revenue runs ~88–89%.

The EBITDA-margin story — and a reconciliation that matters. Note that third-party financial databases mislabels Adyen’s operating income (EBIT) as “EBITDA”: its €1,096m FY25 figure is after €135.8m depreciation/amortization. Adyen’s own reported EBITDA = EBIT + D&A = €1,246m, a 53% margin (the number the market and guidance use); EBIT margin is 46.6%. Both tell the same arc. Margin peaked at 59% in 2021, then collapsed to a 43% trough in H1-2023 — the single most important thing to understand about the business. The collapse was not demand loss; it was a deliberate hiring surge (headcount roughly +25% in ~18 months, ~3,300 toward ~4,000+) as management front-loaded investment into the US, Unified Commerce and new geographies. The market misread it as structural — the stock fell 39% in a session — compounded by the genuine scare that Adyen was losing US large-merchant volume to lower-priced rivals. Management then slowed hiring, and operating leverage did exactly what a capital-light platform’s leverage should: margin recovered to 50% (FY24) and 53% (FY25), with H2-2025 at 55%. The proof is in the incremental operating margin: 53.1% (FY25) and 52.6% (FY24) versus −0.2% in the 2023 investment year. New net revenue now drops through at >50%; management guides to >55% EBITDA margin by 2028.

The float is real money, but it is rate-sensitive money — quantify it. A large, under-appreciated slice of pre-tax profit is net financial income — interest earned on merchant settlement balances and Adyen’s own cash, held under its banking license:

FY25 pre-tax income of €1,394m = operating income €1,096m + €298m float. Strip the float and tax operating income at the effective 23.8%: ex-float net income ≈ €835m (vs €1,062.5m reported) and ex-float diluted EPS ≈ €26.4 (vs €33.61) — so roughly 21% of EPS is float-derived. High-quality cash, but it shrinks as the ECB cuts, and the effect is already visible: reported net income grew only +14.8% in FY25 while EBIT grew +25.6%, precisely because float fell from €354m to €298m and the tax rate ticked up. Headline earnings understate operating momentum on the way down the rate cycle (and would re-inflate it on the way up). For valuation, the honest lens is ex-float: at €825 the P/E is 24.5x reported but ~31x ex-float. This is a forward-EPS headwind, not a quality flaw.

FCF quality — ignore reported OCF year-to-year, normalize it. Operating cash flow is violently distorted by the timing of merchant settlement balances at each 31-December (working-capital swings of +€1,424m (FY22), +€1,011m (FY23), +€654m (FY24), then −€244m (FY25)). These are merchant funds in transit, not Adyen’s cash, and net to roughly zero over time. Reported FCF was €907m in FY25 (OCF €1,030m − capex €124m; 86% conversion). The clean number is normalized FCF = net income + D&A − capex = €1,062.5 + €135.8 − €123.7 ≈ €1,075m, ~101% of net income. Capex is €124m (5.2% of net revenue, ≈ D&A), and SBC is just €40.6m (1.7%). Cash conversion is effectively 100% on a normalized basis, and earnings are genuinely clean: no goodwill amortization, no acquired-intangible drag, no one-time gains — the only two subtleties are the float income and the settlement-balance noise.

Returns — the “low” ROIC is a cash artifact, not weak economics. ROE is 26.3% (FY25), down from 30.3% (FY24), 31.0% (FY23) and 34.8% (FY22) — declining purely because equity is bloating with retained cash, not because returns deteriorate. Standard databases report ~16.3% ROIC, but that buries €10bn of cash in invested capital. The operating business runs on almost no capital (~5%-of-revenue capex, near-zero goodwill, minimal working capital), so the return on operating capital is effectively triple-digit and the return on incremental capital is extraordinary (>50% incremental margin on near-zero incremental capital). In Greenwald’s terms this is a genuine high-return, capital-light compounder; the headline 16% is the price of management hoarding cash (see Capital Allocation).

Balance sheet — decode the banking model. Total assets are €12.26bn against €5.29bn equity, and cash is €10.42bn — but that cash is mostly not Adyen’s. Of €6.79bn current liabilities, ~€6.72bn is merchant payables: funds collected on cards but not yet settled, matched by an equal slug of cash. The economically meaningful figure is corporate net cash ≈ cash €10,425m − merchant payables €6,718m − leases €252m − taxes payable €165m ≈ ~€3.5bn, growing ~€0.9bn/year from FCF. There is no financial debt (only €252m IFRS-16 leases). Equity is €5,285m, of which intangibles are a trivial €7.5m and goodwill is ~zero (never doing M&A), so tangible book ≈ €5,278m ≈ €167/share — among the very few payments names with large positive tangible book (contrast Fiserv/FIS/GPN, all deeply negative). Adyen’s banking license requires CET1 capital, so a portion of equity is genuinely “working,” but the requirement is a fraction of €5.3bn. Net: liquidity and solvency are a fortress (current ratio 1.72x, zero leverage), and the balance sheet is over-capitalized — precisely the capital-allocation problem.

Verdict — yes, economics improve markedly with scale. 88% gross margin, >50% incremental operating margin, ~100% normalized cash conversion, capital-light, minimal dilution. The two honest caveats: (i) ~21% of EPS is rate-sensitive float that is now shrinking, and (ii) headline net income flatters/deflates with the rate cycle and must be read ex-float. Neither dents the conclusion that this is a high-quality earnings stream.

7. Capital Allocation

The record is a study in two opposite halves: outstanding discipline, and a worsening idle-cash problem.

The good half — discipline that has created value by what it avoided. Since its 2018 IPO Adyen has done zero M&A — not one acquisition in the company’s history — and built its entire single global platform in-house on <5%-of-net-revenue capex (€124m FY25, ≈ depreciation). This is the Marathon “capital-returns” virtue in its purest form: the payments industry’s dominant value-destruction mechanism is serial, goodwill-funded M&A — Fiserv (First Data), FIS (Worldpay), and Global Payments (TSYS/Worldpay) all rolled up competitors at high multiples, manufactured negative tangible equity, and watched consolidated ROIC sink toward WACC. Adyen sidestepped the entire trap. Its organic platform produces ~zero goodwill, positive tangible book, and a >50% incremental operating margin — i.e., growth that creates rather than buys revenue. Dilution is negligible (shares 29.5m in 2018 → 31.5m in 2025, ~+7% over seven years, SBC just 1.7% of net revenue), so there is no need to buy back stock merely to offset grants. On capex and M&A, management has allocated capital intelligently and deserves full credit.

The bad half — the cash is now piling up faster than any plausible use. Adyen has never paid a dividend and never repurchased a share. It retains 100% of earnings, and the surplus corporate cash (~€3.5bn, excluding merchant float and regulatory capital) compounds at roughly €0.9bn/year. That cash earns the ECB deposit rate — ~2–3% pre-tax and falling — i.e., ~2% after tax, well below the cost of equity. For a business that demonstrably needs almost no incremental capital to grow at 20%+, hoarding a low-returning and growing cash pile is quietly value-leaky, and the leak widens as the pile grows and rates fall. The argument sharpens because the stock is at its cheapest-ever multiple — a buyback here would be genuinely accretive, the textbook moment to act.

Has the policy shifted? Signaled openness, no commitment. Pressed directly on the H2-2025 call on whether a buyback would “signal focus on shareholder value” given the stock’s ~34% fall since the November Investor Day, the CFO moved off the old absolutist “no returns” line: “we should always be open to what options are available… we’re certainly not dogmatic here… we constantly assess what’s the right decision.” But he immediately re-anchored on growth and flagged that the embedded-finance/banking build-out (issuing, capital lending, bank accounts) is balance-sheet-consumptive and a reason to keep flexibility. The stance has softened from “never” to “open, but not now” — a hint, not a plan. There is a credible strategic case for some retention (regulatory capital + funding embedded-finance assets), but it does not justify €3.5bn-and-growing.

Verdict — mixed, leaning positive on philosophy but increasingly hard to defend on execution. Best-in-class on the things that destroy most payments companies (M&A and capex), genuinely shareholder-aligned in spirit, but the failure to return any capital while sitting on a growing low-return cash pile at a trough valuation is a real and now pointed criticism. A first buyback/return announcement would be the single clearest signal the policy has actually changed rather than merely been “evaluated.”

8. Changes and Headwinds — Last Two Years

Timeline (built from shareholder letters, Adyen press releases and trade press):

1. Aug-2023 — the first crash (−39%). H1-2023 revealed an EBITDA-margin collapse to ~43% from a deliberate headcount surge colliding with US/APAC large-merchant volume lost to cheaper rivals (Braintree/PayPal) — the proof that enterprise acquiring is price-competitive and Adyen is a low-cost/high-capability processor, not an unassailable toll road.
2. 09-Nov-2023 — Investor Day reset (+37.8%). Net-revenue growth “low-20s to high-20s %” through 2026, EBITDA margin >50% by 2026 (later >55% by 2028), capex <5%. Restored credibility.
3. 2023 — co-CEO transition. Ingo Uytdehaage (CFO 2011–2023) elevated to co-CEO alongside founder Pieter van der Does; Ethan Tandowsky promoted to CFO.
4. 2024 — recovery and North America re-acceleration. Stock to €1,570 (Mar-2024); FY24 net revenue +22%, margin back to 50%.
5. Feb-2025 — peak (€1,846) on FY24 results.
6. 12-Feb-2026 — the de-rate (−21.9%) on FY25/H2 results (diagnosed below).
7. 06-May-2026 — Q1-2026 update. Net revenue €620.8m, +20% cc / +16% reported (FX drag); volume +21%; UC +28% cc, Platforms +40% cc, Digital +13% cc — in line, but the reported softness kept sentiment weak.
8. 27/28-May-2026 — CFO departure. Ethan Tandowsky resigned effective 31-Aug-2026 to “pursue an opportunity outside fintech”; the Supervisory Board said it was “surprised by the timing”; no successor named. The second finance-leadership transition in three years, compounding co-CEO succession opacity — a genuine governance/continuity caution.

The Feb-2026 de-rate, diagnosed. This was not an earnings miss but a growth-rerating. Primary cause — guidance below consensus: Adyen guided 2026 net revenue growth to 20–22% cc against ~22.8% consensus — a miss at the midpoint, with Q1 promptly printing exactly +20% cc; three months after an Investor-Day framework implying “low-to-mid-20s,” crystallizing it at 20–22% read as a downgrade. Secondary cause — soft volume and a shrinking core: H2-2025 processed volume €745bn, +12% (missing ~€771bn consensus), with Digital pillar volume −1% — a confidence shock for a stock priced on volume network effects. Aggravating cause — the guidance-philosophy change: Adyen shifted to a once-a-year bottoms-up “expectation” and explicitly disowned sandbagging (“the market rewards beats and positive revisions, which is not the game we play”), removing the beat-and-raise cushion and compressing the multiple independently of fundamentals. What did NOT cause it: EBITDA was a beat (H2 margin 55%); the take rate rose (16.2→17.1bps); the float/ECB drag was already known and structural; the CFO news came later. The de-rate is a textbook high-multiple-compounder repricing — the entire question reduces to whether 20% cc is a durable multi-year floor or the first rung of continued deceleration.

Competitive, macro and regulatory headwinds. Stripe pushing up-market; PayPal/Braintree price competition; the longer-tail threat of account-to-account/open-banking and stablecoin rails disintermediating card economics; European consumer softness and the APAC online-retailer (tariff/de-minimis) volume hit; ECB rate cuts shrinking float; and USD weakness depressing reported vs cc growth by ~3–4 points. New products (the AI “Dynamic Identification” layer and Uplift suite, embedded-finance build-out, agentic-commerce protocol work with OpenAI/Google/Visa/Mastercard, new terminals, SAP partnership) deepen the moat but are mostly slow-burn.

Verdict — net mixed, modestly thesis-weakening at the margin, but the franchise is intact. The negatives cluster on the income statement and guidance (deceleration, FX and float drag, a consensus-below guide, lost beat-and-raise cushion), on people/continuity (surprise CFO exit, co-CEO opacity), and on the competitive horizon (Stripe, A2A/stablecoin optionality). The positives are real but slower-burning (NA >30% recovery, UC/Platforms 33–49%, LatAm, Japan/India domestic phase, AI uplift, embedded-finance inflection). On balance the last two years lowered Adyen’s growth ceiling and stripped the valuation cushion — which is why the multiple re-rated to its cheapest-ever level — without impairing the moat, the fortress balance sheet, or the ~100%-conversion cash generation.

9. Risk Analysis

Catastrophic-loss risk is low: no leverage, no goodwill, fortress liquidity, a regulated bank with capital well above requirement. The risk here is multiple/growth compression and value leakage, not solvency or a wipeout.

10. Valuation Discussion (Embedded Expectations)

No price target, no recommendation. The figures below characterize what the market is pricing.

The spot setup. At €824.9 and 31.54m diluted shares, market capitalization is ~€26.0bn. Stripping ~€3.5bn corporate net cash, enterprise value is ~€22.5bn. On that base:

The core finding is own-history, not cross-sectional. Adyen has never been this cheap as a public company. The stock carried a P/E of 90–220x through the 2018–2021 ZIRP bubble, 50–75x in 2022–2023, and the FY2025 range spanned ~33–55x. At ~24.5x trailing it sits at the cheapest multiple since its 2018 IPO — and on ~31x ex-float still near its own floor. This is the inverse-of-richest setup that recurs across quality names de-rated rather than broken: it rhymes with Dynatrace (a compounder left for dead at the 15th percentile of its own history) far more than with a structurally-impaired value trap.

Peer comp table — priced between the cheap acquirers and the premium networks, while growing faster than either.

The read: the market awards network-like multiples (V/MA ~21x EV/EBITDA, ~25x earnings) to two-sided franchises growing ~11–13%, and single-digit multiples to acquirers/processors growing low-single-digit-to-negative (PYPL, GPN, FISV). Adyen grows two-to-three times faster than any name in the table yet trades closer to the network multiple on EV/EBITDA while at a discount to its own history. The table poses the whole question: is Adyen a faster-growing peer of the networks (under-priced), or a faster-growing acquirer whose multiple still has room to compress toward the GPN/PYPL cohort (correctly priced, or even rich)?

A blunt relative datapoint. Stripe’s February-2026 employee tender struck a $159bn valuation (+74% YoY), on $1.9T processed volume growing +34%. Adyen processed €1.39T and is valued at ~€26bn (~$28bn) — roughly one-sixth of Stripe’s private mark on comparable scale and only modestly slower growth. Private capital pays a vastly richer multiple for the leader of the same category the public market is fleeing. Context, not an anchor (private marks are illiquid and investor-set), but the dispersion is striking.

Reverse-DCF / embedded expectations. Adyen is a near-zero-reinvestment business (capex ≈ D&A, negligible working capital, no M&A), so FCF ≈ NOPAT and the multiple maps cleanly to growth. At a ~9% cost of equity, the 24.5x trailing multiple solves, via P/E ≈ 1/(r−g), to an implied perpetual earnings-growth rate of ~4.9%. A two-stage version is more telling: to fade to an ~18x terminal multiple (g 3.5%, r 9%) over ~7 years, the price embeds only ~14% EPS CAGR over that window. Against management’s 2026 guide of 20–22% cc with 50%+ incremental margins (>25% near-term EPS growth ex-float), the embedded ~14% blended path implies a PEG of ~0.6–0.7 on near-term growth. Plainly: at €825 the market does not believe the 20% compounding is durable; it underwrites convergence toward a GDP-plus acquirer within a few years.

Scenario analysis (illustrative to FY2030, ~31.5m shares — assumptions explicit, not a target).

What the market is underwriting correctly vs incorrectly. Correctly: deceleration is real; the take-rate slide from ~22bps to ~16–17bps is genuine; float income (~21% of EPS) rolls over as the ECB eases; Stripe is a real, faster-growing, better-funded rival, with the 2023 NA loss as proof the large-merchant layer is price-competitive. Plausibly incorrectly: the price extrapolates those into a ~14% terminal grower while Unified Commerce (+33%) and Platforms (+49%) accelerate, NA re-accelerates >30%, the balance sheet is a fortress, and embedded-finance optionality and a potential first buyback at the cheapest-ever multiple sit unpriced. The valuation debate reduces to one question — structurally-advantaged 20% compounder treated like a sub-scale acquirer, or an acquirer whose multiple still has room to fall.

11. Variant Perception

Consensus belief. Wall Street remains overwhelmingly constructive even after the de-rating: covering analysts carry an average 12-month target of ~€1,380–1,460 and a “Buy”/“Strong Buy” consensus — implying roughly +70% upside from €825. The sell-side reads Feb-2026 as a multiple overshoot on a guidance trim, not an estimate collapse, and still models a high-teens-to-low-20s compounder. The market price, by contrast, embeds only ~14% terminal EPS growth — so consensus targets and the tape are sharply at odds, and the share price is the more bearish of the two.

Strongest bull case — cheapest-ever quality compounder. Adyen is the best-architected platform in payments — one global codebase built entirely in-house, end-to-end gateway/acquirer/processor/issuer with its own banking licenses — de-rated to ~24x, the lowest since IPO, while still growing net revenue ~18–22%. The growth engines are accelerating, not fading: Unified Commerce +33% cc, Platforms +49% cc, in-person volume +34%, North America reaccelerating >30%. Margins are recovering toward >55% EBITDA by 2028 with 50%+ incremental flow-through; the balance sheet is a fortress (~€3.5bn corporate net cash, no debt, no goodwill, ~100% cash conversion); there is no acquisition risk and the cleanest earnings in the group. Real unpriced optionality sits in issuing/embedded finance and — for the first time — a credible path to a maiden buyback at a trough multiple. If the 20% compounding is even half-durable, the stock is being given away.

Strongest bear case — a structurally compressing acquirer dressed as a compounder. The take rate has fallen ~25–30% over the platform’s life as large-merchant discounting dominates — structural, not cyclical. Stripe is winning — $1.9T volume +34%, a $159bn mark — and the 2023 NA loss proves Adyen is a low-cost processor competing on price, not an unassailable franchise; switching costs protect the integration, not the economics. ~21% of EPS is float income rolling over as the ECB cuts, masking slower operating growth (reported NI +14.8% in FY25 vs EBIT +25.6%). Growth has decelerated for years and the company itself trimmed the guide from “mid-twenties” to 20–22%. Add disintermediation risk (stablecoins, A2A), a surprise CFO departure with no successor named, co-CEO succession opacity, and a “we only think 3–5 years out” management posture that reads as dismissive. In this reading ~24x is not cheap — it is an acquirer multiple with room to compress toward GPN/PYPL, and “cheapest-ever” is a value-trap echo of a multiple that was always a bubble.

The 3–5 assumptions that matter most. (1) Take-rate durability — does mix shift to UC/Platforms hold the blended rate ~16–17bps? (2) Stripe / competitive share — can Adyen hold and re-gain enterprise share, or is 2023 the template? (3) Growth-rate path — is 20–22% cc durable for 3+ years, or does it fade to low-teens? (4) Float normalization — how far does ~21%-of-EPS float fall, and can operating growth absorb it? (5) Capital return — does management deploy the idle ~€3.5bn at a trough multiple, or let it leak?

What would falsify each side. Bull falsified by: take rate breaking below ~16bps; a marquee merchant publicly lost to Stripe; 2026 actuals undershooting 20–22%; float rollover dragging NI growth to single digits with no operating offset. Bear falsified by: sustained 20%+ cc prints with stable/rising take rate; NA share re-gains; UC/Platforms holding 30%+; and a first buyback monetizing the idle cash at the trough.

Price-action / positioning read. the quantitative factor read for this Amsterdam line is limited, so the positioning read is built from the Euronext Amsterdam price history. Beta has compressed to ~0.84 (from ~1.0+ historically) as Adyen fell out of the high-beta growth cohort and now trades on idiosyncratic de-rating — itself a tell that the market has re-classified it from momentum-growth toward a beaten-down name. The stock is −70% off its ATH, sits below its 21-, 50- and 200-day moving averages (€867 / €901 / €1,105), all declining, printed a fresh multi-year low of €785 on 2026-06-11, bounced to €904 (06-18), and slipped back to €825. Honest characterization: an abandoned-growth / de-rated former darling in an intact downtrend that has stopped accelerating but has not confirmed a turn — a falling knife that is decelerating. This supports the contrarian framing (washed-out positioning) but is not a clean momentum-bottom signal; the tape argues for patience over aggression.

12. Fact vs. Interpretation

13. Open Questions

1. Geographic net-revenue split (EMEA/NA/APAC/LatAm) — not in the press release; pull from the 2025 annual report to size the NA re-acceleration precisely.
2. Durability of the 16.2→17.1bps take-rate uptick — genuine mix-shift floor, or a temporary eBay roll-off base effect that fades once eBay is fully out?
3. Hard auth-rate evidence vs Stripe — is the “Uplift” data-network advantage real and quantifiable, or an unprovable marketing claim?
4. Float-income glide path — how far does net financial income fall as the ECB eases toward neutral, and what is the steady-state contribution to EPS?
5. Combined founder ownership and any 2025–26 disposals — reconcile against the 2025 annual report and the AFM substantial-holdings register (no Form 4 tape exists for a foreign issuer).
6. Capital-return policy — does the softened tone convert into an actual buyback/dividend, and at what trigger?
7. CFO succession — internal or external, and how it affects the credibility of guidance and the embedded-finance build-out.
8. Incentive design — does management compensation embed any return-on-capital metric (proxy not yet read), given the idle-cash issue?

14. What Must Be True

For the bull case (Adyen compounds and re-rates):

• Net revenue sustains ~20% cc for at least the next 2–3 years, with Unified Commerce and Platforms holding 30%+ and offsetting the maturing Digital core. Falsification test: any full-year cc net-revenue print below ~18%, or a UC/Platforms deceleration below ~25%, breaks the durable-compounder thesis.
• The blended take rate holds ~16–17bps (mix shift offsets large-merchant discounting). Falsification test: a reported blended take rate breaking below ~16bps.
• Adyen holds enterprise share against Stripe and re-gains North America. Falsification test: a publicly-disclosed loss of a marquee enterprise merchant to Stripe, or NA growth decelerating below ~15% cc.

For the bear case (Adyen converges to an acquirer):

• Growth fades from 20% toward mid-teens-then-GDP-plus as scale, Stripe and disintermediation compound, and float income keeps masking the operating slowdown. Falsification test: two consecutive years of 20%+ cc net-revenue growth with a stable-to-rising take rate.
• The take rate continues its secular slide below 16bps and switching costs fail to defend price. Falsification test: the take rate stabilizing or rising for four-plus consecutive halves.
• The idle cash stays idle and value leaks at the cost of equity. Falsification test: a first buyback/dividend authorization that monetizes surplus cash at the trough.

The shared falsification clock is the FY2026 actuals (and the once-a-year February guides): a 20%+ cc print with a stable take rate is the bull’s vindication; a guide-down below ~18% with continued take-rate slippage is the bear’s.

15. Source Appendix

(See the full source list in Appendix B below.) Primary sources: Adyen H2-2025 and H1-2025 shareholder letters and press releases (adyen.com/investor-relations); Adyen Q1-2026 business update (2026-05-06); Adyen CFO-departure release (2026-05-28); Adyen FY2024 annual report; the Adyen H2-2025 earnings-call transcript. Quantitative data: Adyen’s own reporting cross-checked against third-party financial databases (financial statements, ratios, valuation multiples, enterprise value — with the net-vs-gross-revenue and EBIT-vs-EBITDA labels corrected); Euronext Amsterdam five-year price history. Peer context: the public filings and reporting of Visa, Mastercard, PayPal, Block, Fiserv, Toast, Global Payments. Third-party: CNBC/TechCrunch (Stripe $159bn tender, 2026-02-24); MarketScreener/Investing.com (consensus); FXC Intelligence, Crowdfund Insider, Payments Dive (2026 results/CFO coverage); Greenwald & Kahn, Competition Demystified, and Marathon/Chancellor, Capital Returns, for the analytical frameworks.

APPENDIX A — Standard Diligence Questionnaire

Adyen N.V. (Euronext Amsterdam: ADYEN.AS) · Prepared 2026-06-27 · All figures IFRS, EUR. Supplemental to the main note; Fact / Interpretation / Assumption labels applied where material.

General

What thoughtful questions have other investors asked about this company? The dominant question is whether the take-rate compression (~22bps in 2018 → ~16–17bps today) is structural mean-reversion or a benign mix artifact; whether Stripe’s up-market push erodes Adyen’s enterprise franchise the way Braintree did in North America in 2023; how far float income (~21% of EPS) falls as the ECB eases; whether 20% constant-currency growth is a durable floor or the first rung of a longer fade; and — increasingly loudly since the stock halved — when management will return any of the growing ~€3.5bn idle cash pile. The Feb-2026 −22% drop and the May-2026 surprise CFO departure sharpened all of these.

Cyclicality & Earnings Nature

Are earnings at a cyclical high or low? Interpretation: mid-cycle on the operating line but at a float-income high that is now rolling over. Operating margins have recovered from the 2023 trough (~43%) to ~47% EBIT / ~53% reported EBITDA, below the 2021 peak (~59%) — neither extreme. The ~21%-of-EPS float income, however, peaked with ECB rates in FY2024 (€354m) and is falling (€298m FY25), so reported EPS carries a fading tailwind.

Driven by the external environment or internal actions? Both. Volume is consumer-cyclical (external); the margin trajectory is largely internal (the deliberate 2022–23 hiring surge and subsequent discipline); float income is purely external (rates).

How stable are revenues? Fact/Interpretation: highly recurring in character — net revenue recurs as long as merchants keep processing, with >100% net revenue retention — but the growth rate is volatile and decelerating, and a slice is consumer-volume-cyclical. Not contractually recurring like SaaS, but stickier than transactional.

Outlook for products/services. Strong secular demand (cash→digital, e-commerce, unified commerce, embedded finance); the question is share and price, not demand.

How big will this market be? Global digital-payments TAM is enormous and growing double-digits; Adyen’s served niche (enterprise unified commerce + platforms/embedded finance) is a high-barrier subset with multi-year runway. International by design (28 offices, 30+ jurisdictions).

Business Quality & Competitive Moat

Is the industry getting more or less competitive? More — capital has flooded the acquiring layer (Stripe, Checkout.com, Nuvei, PE-backed Worldpay) and take rates compress industry-wide (Marathon late-capital-cycle signature). Adyen’s specific niche is less commoditized than vanilla online acquiring.

How profitable is the business (ROIC, ROE)? Fact: ROE ~26% (FY25, declining only as cash builds); third-party financial databases ~16% (depressed by €10bn cash in the denominator); return on operating capital is effectively triple-digit (capex ~5% of net revenue, near-zero goodwill, >50% incremental margins).

How profitable is the industry — competitors, barriers? Bifurcated: networks (V/MA) earn ~50% margins on true barriers; acquirers/processors mostly earn single-digit-multiple economics. Barriers in Adyen’s niche: single global stack, banking/acquiring licenses across 30+ jurisdictions, deep enterprise integrations.

Can the business be easily understood? Yes, with one caveat — the bank-like balance sheet (merchant float) and the net-vs-gross revenue distinction must be decoded, or every ratio misleads.

Can it be undermined by foreign low-cost labor? No — this is a technology/licensing/scale business, not a labor-arbitrage one.

Do brands matter? Modestly — Adyen has strong enterprise brand equity (“the payments company for ambitious enterprises”), but purchases are RFP-driven on capability and price, not brand.

Nature of competition? Capability + price. Adyen wins on integrated single-stack capability (unified commerce, global reach, auth-rate optimization); it can lose on price at the large-merchant layer (the 2023 Braintree episode).

Customers’ switching costs? Real and high for the integration (multi-quarter rip-and-replace across channels/geographies) — explains >100% retention — but they do not protect the take rate, which large merchants re-negotiate and multi-source.

Financial Condition & Balance Sheet

Assets not fully recognized on the balance sheet? The in-house-built platform and banking/acquiring licenses are expensed/carried at little book value — genuine economic assets understated by IFRS. Interpretation.

Off-balance-sheet liabilities? None material beyond operating leases (already on-sheet under IFRS-16, €252m). Merchant settlement obligations are on-sheet and matched by cash.

How conservative is the accounting? Conservative and clean — no goodwill, no acquired-intangible amortization, no one-time gains, minimal SBC (1.7%). The two subtleties are float income (real but rate-sensitive) and merchant-settlement working-capital noise in OCF.

How CapEx-hungry? Very light — capex ~5% of net revenue (≈ depreciation); the entire global platform was built on this intensity.

Capital Allocation & Management

How much FCF, and how is it used? Normalized FCF ~€1.07bn (FY25, ~100% of net income). Use: retained entirely — no dividend, no buyback ever. Surplus corporate cash ~€3.5bn, growing ~€0.9bn/yr, earning sub-cost-of-equity returns. This is the central capital-allocation criticism.

Significant acquisitions? None — zero M&A in the company’s history. A genuine differentiator and the source of its clean balance sheet (Marathon virtue).

Buying back shares? No — never. The softened management tone (“not dogmatic… we constantly assess”) hints at a possible first return; a buyback at the cheapest-ever multiple would be accretive.

Issuing large amounts of stock to insiders? No — SBC is just 1.7% of net revenue; shares grew only ~7% in seven years.

Compensation policy. The “Adyen Formula” — flat hierarchy, modest fixed salaries, deliberately limited bonuses, restrained equity. Genuinely low-extraction. Open question: whether incentives embed any return-on-capital metric (proxy not yet read), which sits awkwardly with the idle cash.

Motivations of management. Founder-led, long-term-oriented (“we think 3–5 years out and ignore the short term”) — principled but at times dismissive of legitimate shareholder concerns about the collapsed price and zero capital return.

Valuation & Market Data

ADR, MLP, or K-1 issuer? None — Adyen is an ordinary share listed on Euronext Amsterdam (ISIN NL0012969182). Not a US filer; no K-1; no UBTI issue. US investors typically hold the ordinary shares or an OTC ADR; FX (EUR) exposure applies.

Dividend policy? No dividend. 100% retention.

How profitable? Among the most profitable in payments on operating metrics (~47% EBIT margin, ~88% gross margin on net revenue, ~26% ROE).

Is net income diverging from cash from operations? Reported OCF swings violently with merchant-settlement timing and should be ignored year-to-year; on a normalized basis (NI + D&A − capex) cash conversion is ~100%. No adverse divergence.

Risks & Downside

What would cause the stock to decline? A guide-down below ~18% cc; a take rate breaking below ~16bps; a marquee merchant lost to Stripe; faster-than-expected float-income rollover; further multiple compression toward the acquirer cohort; execution wobble around the CFO transition.

Risk of catastrophic loss? Low — no leverage, no goodwill, fortress liquidity, a regulated bank with capital well above requirement. The risk is multiple/growth compression, not solvency.

Chance of a total loss? Negligible on any reasonable horizon.

Recent News & Events

Has the business environment changed recently? Yes — growth decelerated to 20–22% cc, the company adopted once-a-year bottoms-up guidance (removing the beat-and-raise cushion), float income began rolling over with ECB cuts, and Stripe’s private valuation re-rated to $159bn. The franchise, moat and balance sheet are unchanged.

Significant acquisitions? None.

Change in accounting policies? None material; presentation normalized to net revenue from 2023.

Recent changes — markets, facilities, management? Co-CEO transition (Uytdehaage alongside van der Does, 2023); North America rebuilt into the largest investment market; Japan/India entering the domestic phase; LatAm the fastest-growing region; and the surprise CFO departure (Tandowsky, effective 31-Aug-2026, no successor named) — the key recent negative.

APPENDIX B — Source Appendix

Adyen N.V. (Euronext Amsterdam: ADYEN.AS) · Compiled 2026-06-27. Primary sources first. Adyen is a foreign issuer (no SEC filings); the source of record is the company’s half-year shareholder letters, annual report, and Euronext/AFM disclosures. Quantitative figures cross-checked against third-party financial databases and reconciled to the company’s own reporting; the Euronext Amsterdam price history underlies the price-action map.

Primary — Company Filings & Disclosures

• Adyen N.V., H2/FY-2025 shareholder letter & results (reported 2026-02-12/13). https://www.adyen.com/press-and-media/adyen-publishes-h2-2025-financial-results-3pgu2
• Adyen N.V., H1-2025 shareholder letter & results (2025-08-14).
• Adyen N.V., Q1-2026 business update (2026-05-06). https://www.adyen.com/press-and-media/adyen-publishes-q1-2026-business-update-4gyhh5
• Adyen N.V., CFO Ethan Tandowsky to step down (2026-05-28). https://www.adyen.com/press-and-media/adyen-cfo-ethan-tandowsky-to-step-down-to-pursue-opportunity-outside-fintech
• Adyen N.V., FY-2024 Annual Report and FY-2024 shareholder letter (IFRS, EUR). https://www.adyen.com/investor-relations
• Adyen N.V., Investor Day 2023 materials (medium-term targets reset, 2023-11-09).
• Adyen N.V. H2-2025 earnings-call transcript (management commentary, treated as hypothesis and validated against the financials).
• Adyen investor-relations hub (financial calendar, share data, governance). https://www.adyen.com/investor-relations

Quantitative Data Sources

• Third-party financial databases — income statement, balance sheet, cash flow, profitability ratios, valuation multiples and enterprise value (multi-year). Used as a starting point and reconciled to Adyen’s own reporting; two labels corrected in this report: (i) the pre-2023 “revenue” line is gross settlement, not net revenue — net revenue corresponds to the company’s headline figure; (ii) the database “EBITDA” line is operating income (EBIT) — reported EBITDA adds back D&A.
• Euronext Amsterdam price history (daily prices, split/dividend-adjusted, with moving averages and beta) — basis for the five-year event map and the price-action/positioning read.
• Note: standard US factor models do not cover this Amsterdam-listed line; the positioning read was built from the Euronext Amsterdam price history instead.

Peer & Industry Context

• The public filings and reporting of Visa (V), Mastercard (MA), PayPal (PYPL), Block (XYZ), Fiserv (FISV), Toast (TOST), Global Payments (GPN) — for the payments value chain, take-rate mechanics, and competitor economics/multiples.
• Stripe valuation — February-2026 employee tender at $159bn (+74% YoY); ~$1.9T processed volume (+34%). CNBC (2026-02-24): https://www.cnbc.com/2026/02/24/stripe-value-stock-sale-tender-offer.html ; TechCrunch (2026-02-24): https://techcrunch.com/2026/02/24/stripes-valuation-soars-74-to-159-billion/
• Consensus / targets — MarketScreener (https://www.marketscreener.com/quote/stock/ADYEN-N-V-44211922/consensus/) and Investing.com (https://www.investing.com/equities/adyen-consensus-estimates).
• Results / event coverage — CNBC (2026-02-12, share reaction); FXC Intelligence (FY25 analysis, https://www.fxcintel.com/research/analysis/ct-adyen-fy-25-earnings); Crowdfund Insider (2026-02-12, https://www.crowdfundinsider.com/2026/02/263162-adyen-shares-drop-up-to-20-following-cautious-outlook-for-future-growth/); Payments Dive (CFO exit, https://www.paymentsdive.com/news/adyen-cfo-heads-for-exit/821365/); Investing.com (Investor Day 2023 targets, https://www.investing.com/news/stock-market-news/adyen-gains-after-unveiling-new-20-longterm-growth-and-higher-margin-targets-4348357).

Analytical Frameworks

• Bruce Greenwald & Judd Kahn, Competition Demystified — barriers to entry, the three genuine advantage types, market-share-stability and ROIC tests.
• Edward Chancellor (ed.) / Marathon Asset Management, Capital Returns — supply-side capital-cycle analysis, the asset-growth anomaly, high-returns-attract-capital mean reversion.

All multiples and per-share figures recomputed at the 2026-06-26 close of €824.9 and 31.54m diluted shares; a stale database price (€1,375) was not used. Management commentary is treated as hypothesis and validated against the financial statements throughout.