Merchant acquiring and payment infrastructure. ADYEN (Euronext Amsterdam) / Stripe, Inc. (private, no audited financials published). Run 17 September 2026. Adyen evidence: annual reports FY2023 to FY2025, shareholder letters H2 2025 and H1 2026, Talon.One deal document and call transcripts (research folder sources). Stripe evidence: Stripe annual updates for 2022 to 2025, Stripe newsroom, and press reporting (The Information, Axios, TechCrunch, American Banker), labeled press or inferred. Events swept through 17 Sep 2026; most recent events checked: Stripe Link integration with Meta's Muse agent (8 Sep 2026) and Adyen India expansion report (9 Sep 2026). Figures in EUR; Stripe USD converted at ECB average EUR/USD 1.0824 for 2024 and approx. 1.13 for 2025 (approx., unverified). Not a valuation and not a recommendation.
Three cells should carry the next five years, and Stripe holds the structural driver in two of them.
Adyen's guide of 21 to 23% constant-currency growth for 2026 includes about 1 point from acquisitions. (H1 2026 letter)
Adyen's conversion is audited and mechanical, and it comes from a single platform and a thin head count. Personnel costs fell from 36.5% to 32.0% of net revenue between 2023 and 2025 as FTEs rose 14% and net revenue 45%, lifting EBITDA margin from 45.7% to 52.7%. Adyen processes about EUR 296M of volume per employee vs about EUR 151M at Stripe (2024, inferred) (AR 2023 to 2025; TechCrunch, 21 Jan 2025).
Stripe converts through price instead: roughly double the take rate on a long-tail, list-priced mix, and a press-reported free cash flow of about 47% of net revenue in 2025. Those Stripe figures are unaudited and on an undisclosed definition.
This is the run's central tension. The growth winner monetizes each unit better, while the margin winner runs each unit at lower cost. If Stripe's take rate compresses as its AI and enterprise customers scale into custom interchange-plus pricing, Adyen's cost engine becomes the decisive advantage.
Stripe. Its growth is increasingly carried by a small, fast-scaling AI cohort whose largest members can multi-source or insource. OpenAI added Adyen in H1 2026, and Anthropic is reported to be building billing and fraud tools in-house. Its long tail of startups is the first volume to shrink in a downturn. (Adyen H1 2026 letter; The Information, 4 Sep 2026, press)
Adyen. Its exposure is volume tiering and concentration. 300 merchants account for about 60% of growth, and management calls tiering "the only explanation" for take-rate moves. One large customer cut FY2025 volume growth from 21% to 8%, and EMEA growth has slowed from 26% to 15%. (H1 2026 letter and call; H2 2025 letter)
Adyen's blue-chip base bends, where Stripe's cohort-driven growth snaps.
These six cells carry the bulk of both companies' economics. Every score is argued in the tabs below. Cell revenue is not disclosed as a cross-tab by either company, so the cells are normalized from Adyen's pillars and regions and from Stripe's disclosed customer metrics.
| Cell (product x region) | Adyen | Stripe | Why (one clause, sourced) |
|---|---|---|---|
| Enterprise online payments x EMEA | 4 | 3 | Adyen EMEA EUR 1,360.7M, 57% of FY25 net revenue, still +15% in H1 2026 (AR 2025, H1 2026 letter); Stripe counts half the DAX 40 and CAC 40 as users, but share of wallet is ND (Stripe Tour Berlin, Paris 2026) |
| Enterprise online payments x North America | 4 | 4 | Adyen NA +30% cc, 27% of net revenue (H1 2026 letter); Stripe home market, used by 90% of the Dow and 80% of the Nasdaq 100, US share ND (Stripe 2025 update) |
| Unified commerce (in-person + online) x global retail, F&B | 5 | 2 | Adyen in-person EUR 175.7B +28%, 838K terminals, 486 cross-channel merchants at scale (H1 2026 letter); Stripe Terminal volume undisclosed since 2022 |
| Platforms / embedded payments x NA + EMEA | 4 | 5 | Adyen Platforms +40% cc, 37 platforms above EUR 1B (H1 2026 letter); Stripe serves 5M+ businesses directly or via platforms and powers Shopify (Stripe 2025 update; share lead inferred) |
| AI and software-native monetization (payments + billing) x global | 2 | 5 | Stripe is the default for major AI launches, with a Revenue suite near $1B run-rate plus Metronome (Stripe 2025 update, Jan 2026 close); Adyen's OpenAI win and Orb purchase are from a small base (H1 2026 letter) |
| Long-tail SMB and startups, self-serve x global | 0 | 5 | Stripe: 57% of 2025 new businesses were ex-US (Stripe 2025 update); Adyen is merchant-side enterprise by choice and reaches SMBs only through platforms (H1 2026 call) |
Scores are anchored to the exhibits in the three tabs and are not summed. The three lenses get their verdicts in the Three Answers above.
Normalization first. Adyen reports net revenue by pillar (Digital, Unified Commerce, Platforms) and by billing region (EMEA, North America, APAC, LatAm), but never as a cross-tab. Stripe discloses only total volume, customer counts and product run-rates, and its revenue comes from press reports. So the cell view below mixes disclosed Adyen figures with Stripe signals. Where a Stripe number does not exist it is ND, and the butterfly plots growth rates rather than cell revenue, because Stripe cell revenue would have to be invented.
The headline contrast is not the growth gap itself but where it comes from. Adyen's fastest cells are Platforms (+40% cc) and North America (+30% cc), while its largest cell, Digital in EMEA, is where growth has slowed most. Stripe's growth is broad at the total level and disproportionately fed by AI and software-native customers.
| Cell | Adyen net rev (share) | Adyen growth | Stripe rev (share) | Stripe growth | Margin signal / leader |
|---|---|---|---|---|---|
| Digital (enterprise online), all regions | EUR 1,334.9M (56%) FY25 | +15% cc H1'26 | ND | ND | Adyen take rate on large digital merchants tiers down with volume; Stripe leads new logos, Adyen leads scaled share of wallet |
| Unified Commerce, all regions | EUR 765.4M (32%) FY25 | +27% cc H1'26 | ND | ND | Adyen leads: terminal plus acquiring plus data on one stack |
| Platforms, all regions | EUR 263.8M (11%) FY25 | +40% cc H1'26 | ND | ND | Stripe leads on breadth (Connect, Shopify); Adyen grows faster from a smaller base |
| Region: EMEA | EUR 1,360.7M (57%) | +15% cc | ND | ND | Adyen incumbent; Stripe penetrating (800K French and 300K+ German businesses) |
| Region: North America | EUR 633.3M (27%) | +30% cc | ND | ND | Stripe home market; Adyen gaining share of wallet |
| Total | EUR 2,364.2M | +21% cc FY25 | ~EUR 6.0B ($6.8B, press) | +33% FY25, +41% H1'26 (press) | Adyen 53% EBITDA margin (audited); Stripe "robustly profitable" (company) |
Adyen pillar shares from AR 2025 note on net revenue; H1 2026 growth from H1 2026 letter. APAC EUR 236.5M (10%) and LatAm EUR 133.8M (6%) collapsed into the total.
AI and software monetization. Stripe's driver is default adoption plus attach: Billing, Metronome for usage-based billing (closed 14 Jan 2026), Link with 250M+ users, and the Agentic Commerce Protocol with OpenAI (29 Sep 2025). Adyen's response is acquired rather than organic: Orb (usage billing) and Talon.One (EUR 750M, loyalty), both closed 1 Jul 2026 and expected to add about 1 point to 2026 growth and 1 to 2 points in 2027 at about 1 point of margin dilution. It also launched Adyen Agentic and won OpenAI. Durability favors Stripe: an integrated default is harder to dislodge than a bolted-on alternative. (Stripe newsroom; Adyen deal call 23 Apr 2026; H1 2026 letter)
Unified commerce. Adyen's driver is cross-channel retailers consolidating online and in-store acquiring (Tod's moved 100+ stores and 35 markets in under three months), with 838K terminals up 27%. Stripe has Terminal (S710, T600) but no disclosed volume. The cost to grow here is hardware certification and local acquiring licenses, a multi-year build that favors Adyen. (H1 2026 letter; Stripe Sessions 2026)
Platforms. Both grow through vertical SaaS. Adyen has 293K active business customers (+51%) and expanded Toast into the US. Stripe runs Connect at far larger breadth, including Shopify. Adyen's growth here is organic. (H1 2026 letter; Stripe 2025 update)
Stripe sells self-serve at list price (US 2.9% + 30c online, EEA 1.5% + EUR 0.25 for standard cards) to its long tail, and custom interchange-plus to enterprise. So every long-tail customer pays a blended price that includes the scheme cost. Adyen sells direct to enterprises on interchange-plus with volume tiers, which is why its take rate falls as merchants scale ("our existing merchants give us more volume and therefore come to lower tiers", co-CEO, H1 2026 call). Stripe controls net price at the bottom of the market; Adyen trades price for share of wallet at the top. (Stripe pricing pages, Sep 2026; Adyen H1 2026 call transcript, third-party)
| Input | Adyen | Stripe |
|---|---|---|
| Scheme and rail access | Own banking licenses EU/US/UK; direct CB and STET participant in France; UAE Category II license (2026) | In-house processing per press (Payments Dive, 1 May 2024); license detail ND |
| Compute | Own data centers; 2026 capex raised to 7% of net revenue to lock in supply | ND |
| Balance sheet rails | EUR 4.9B own cash, EUR 12.4B total incl. merchant funds (H1 2026) | Bridge OCC trust charter conditionally approved 18 Feb 2026 (American Banker) |
| Hardware | Terminals sourced; inventory costs in COGS EUR 102.3M FY25 | BBPOS acquired 2022 (own reader maker) |
Both compete with PayPal/Braintree (Stripe's $53B+ bid with Advent was dropped on 27 Aug 2026, press), Worldpay, Checkout.com, Fiserv and JPMorgan in enterprise acquiring. Market shares are not disclosed by either company in a comparable form. Adyen claims to grow faster than the market (H1 2026 letter); Stripe's growth against global e-commerce implies it is a share-taker (inferred).
AI cell. Stripe is more exposed. Large AI labs can split volume (OpenAI to Adyen) or insource (Anthropic, press 4 Sep 2026), and the cell is Stripe's fastest. EMEA enterprise. Adyen is more exposed. It is 57% of net revenue, growth has halved from 26%, and management attributes some of the slowdown to global merchants shifting where they sell (H1 2026 call). Unified commerce. Adyen is exposed only if Stripe proves in-person scale, and no evidence of that has been disclosed.
Single global platform plus own licenses (High). Its banking licenses in the EU, US and UK and direct rail memberships take years of regulatory work, not money (AR 2025; H1 2026 letter).
Unified commerce footprint (High). 838K transacting terminals on the same stack as online, with 486 merchants at scale across both channels.
Enterprise share-of-wallet curve (Med). Share of wallet runs from under 20% to over 40% after year 12, but its large customers can and do reallocate (one customer cut FY25 volume growth by 13 points).
Developer and new-business default (High). All major 2025 AI launches used it, and 57% of new businesses are ex-US. Habit is set at company formation (Stripe 2025 update).
Two-sided network (Med to High). Link has 250M+ consumers, and there is a 92% chance Stripe has seen a card before, which feeds Radar and conversion (Sessions 2026; 2024 update).
Suite breadth (Med). Billing, Tax, Capital, Issuing and stablecoins via Bridge; much of it is buyable, as Adyen's Orb deal shows.
Adyen: direct enterprise sales, merchant-side only by choice ("large merchants consider their shoppers their domain", co-CEO, H1 2026 call), with 300 merchants at about 60% of growth. Net price falls with volume through tiers, so its customers hold price power as they scale. Stripe: self-serve list pricing to millions of businesses gives it net-price control in the long tail. Its largest customers negotiate custom pricing and, in AI, multi-source. Implication: margin pressure sits with Adyen's top merchants in a soft market and with Stripe's top AI accounts in a competitive one.
Adyen: it owns its licenses, scheme connections and data centers. Its costs from financial institutions ran between 6.7% and 7.9% of gross revenue across 2023 to 2025, and it pulled 2027 data-center capex forward amid supply constraints (AR 2025; H1 2026 letter). Stripe: it is reported to process in-house, but its license footprint and cost of scheme fees are ND. Its list pricing passes scheme fee changes straight through to the long tail (inferred). Implication: Adyen is more insulated from partner failure; Stripe is more able to pass cost increases through.
| Cell | Route control | Pocket price | Continuity | Outcome (share / margin) | Confirming KPI |
|---|---|---|---|---|---|
| Enterprise online x NA | Both direct | Stripe above Adyen (blended vs tiered IC++) | Both high | Adyen share up, margin flat; Stripe share flat to up, margin down slightly as mix rises | Adyen NA cc growth at or above 25% (H2 2026 letter) |
| Enterprise online x EMEA | Adyen direct incumbent; Stripe direct challenger | Parity for large merchants (inferred) | Adyen higher (local rails) | Adyen share flat, margin flat; Stripe share up | Adyen EMEA cc growth at or above 17% (FY2026 letter) |
| Platforms x NA + EMEA | Via SaaS partners for both | Stripe above (Connect pricing) | Both high | Both share up; Adyen margin up from scale | Adyen platforms above EUR 1B at 45+ (H1 2027 letter) |
Tension: Adyen grows fastest in NA and Platforms, exactly where Stripe holds the stronger access position. That growth is won by lower pocket price through tiering, not by price power.
1. Merchant mix and pricing model (Major). Stripe serves the long tail at list price, while Adyen serves enterprise on tiers. Closing this would take Adyen a self-serve motion it rejects culturally, so it is effectively not closable. 2. Software layer beyond payments (Moderate). Stripe built Billing organically and added Metronome; Adyen bought Orb and Talon.One in 2026, with integration likely to take 2 to 3 years (inferred). 3. In-person scale (Major, favors Adyen). Stripe would need certified hardware, local acquiring and retail sales capacity across dozens of markets, a multi-year build that money alone does not shortcut.
Mapping note. Adyen presents costs by nature under IFRS: costs from financial institutions and terminal COGS, wages, social costs, D&A, and other operating expenses. It has no R&D or SG&A lines. COGS proxy = costs incurred from financial institutions + costs of goods sold, as % of non-interest (gross) revenue. Because Adyen books most interchange net, this gross margin is not comparable to a card acquirer that books gross. SG&A proxy = other operating expenses (which include IT and advisory), as % of net revenue. R&D is ND; the best proxy is that tech roles are 55% of FTE in 2025 and 56% in 2024. Stripe publishes no income statement, so every Stripe P&L ratio is ND. The comparable metrics are cash margin, take rate and per-employee productivity, built from press revenue and company volume and headcount, and labeled inferred.
| Metric, 3y avg (2023 to 2025) | Adyen | Stripe | Gap | What drives it |
|---|---|---|---|---|
| COGS proxy (% gross revenue) | 11.5% | ND | ND | Scheme/bank costs plus terminal COGS; net interchange presentation |
| R&D (% net revenue) | ND (tech FTE 55%, 2y) | ND | ND | By-nature statement; development mostly expensed in wages |
| SG&A proxy (other opex % NR) | 15.7% | ND | ND | Sales & marketing only 2.8% of NR in 2025: direct enterprise sales, little advertising |
| Gross margin proxy | 88.5% | ND | ND | Not comparable across presentations |
| EBIT margin (% NR) | 44.0% | ND | ND | Personnel fell from 36.5% to 32.0% of NR |
| FCF (% NR) | 42.7% | ~45.1% (2y avg, press) | ~2 pts to Stripe | Definitions differ; Adyen pre-tax, Stripe ND |
| Take rate (bps) | 16.4 | ~36.1 (2y avg, inferred) | ~2.2x | Long-tail list pricing vs enterprise tiers |
| Adyen, by year | COGS / gross rev | Personnel % NR | Other opex % NR | EBIT % | EBITDA % | Take rate bps |
|---|---|---|---|---|---|---|
| 2023 | 12.7 | 36.5 | 16.3 | 40.4 | 45.7 | 16.8 |
| 2024 | 10.7 | 35.1 | 15.5 | 44.5 | 49.7 | 15.5 |
| 2025 | 11.2 | 32.0 | 15.4 | 46.9 | 52.7 | 17.0 |
EUR M: net revenue 1,626.1 / 1,996.1 / 2,364.2; personnel 594.0 / 701.2 / 756.2; other opex 264.7 / 309.6 / 363.3; EBIT 657.6 / 887.8 / 1,109.9 (AR 2024, AR 2025). 2023 EBIT includes EUR 24.3M other expense. Stripe per-year ratios: all ND.
The most persistent structural difference is the take rate: roughly 36 bps at Stripe vs 15.5 to 17.0 bps at Adyen in both years where Stripe data exist. It is explained by the route-to-market mechanism in the Growth Map, long-tail list pricing vs enterprise interchange-plus tiers. Adyen's offset is productivity. It handled about EUR 296M of volume per employee in 2024 vs about EUR 151M at Stripe, which is why a half-size take rate still yields a similar cash margin.
This confirms the power map in both directions. Stripe's access power shows up as price, and Adyen's infrastructure ownership shows up as cost. It also exposes a weakness in Stripe's evidence: its margin claim rests on press figures with no disclosed definition, stock-based compensation, or tax treatment. That caps confidence in any Stripe margin advantage.
| Metric | Threshold | By when | If it hits, it favors | Where published |
|---|---|---|---|---|
| Stripe total volume growth, 2026 | 30% or above on $1.9T base | Feb 2027 | Stripe | Stripe 2026 annual update |
| Stripe take rate (press net revenue / volume) | 0.34% or above | Mid 2027 | Stripe (below 0.30% favors Adyen) | Press plus Stripe update |
| Adyen Digital pillar cc growth | 18% or above | Feb 2027 (FY2026 letter) | Adyen (below 12% favors Stripe) | Adyen H2 2026 letter |
| Adyen EMEA net revenue cc growth | 17% or above | Aug 2027 (H1 2027 letter) | Adyen | Adyen shareholder letter |
| Adyen EBITDA margin path | Underlying at or above 53% in 2026, above 55% by 2028 | Feb 2027 / Feb 2029 | Adyen | Adyen shareholder letters |