Payments. ADYEN (Euronext Amsterdam) / AXP (NYSE). Run 17 September 2026. Built from Adyen annual reports FY2023-25, shareholder letters H2 2024 to H1 2026, Investor Day 2025 and M&A call transcripts; American Express 10-Ks FY2021-25, Q2 2026 10-Q and 2026 earnings-call transcripts; plus cited web sources for the event sweep. Events swept through 17 September 2026; most recent events checked: Amex CFO comments at an investor conference (16 Sep 2026) and Adyen CFO Ethan Tandowsky's exit effective 31 Aug 2026. Adyen figures in EUR, Amex in USD; the duel compares ratios and growth rates, not absolutes (size gap at ~1.13 USD/EUR, approx., unverified: Adyen net revenue ~$2.7bn vs Amex $72.2bn, roughly 27x). Not a valuation and not a recommendation.
Pairing note: these two sit at opposite ends of the same card transaction. Adyen is a merchant-side processor and acquirer earning a take rate; Amex is a closed-loop issuer, network and acquirer earning merchant discount, card fees and interest. They compete for the same pool (what merchants pay to accept a payment), so the duel asks which economic position compounds better, not who has more share of one product.
Three Adyen cells carry the next five years. North American enterprise acceptance (+30% cc in H1 2026, 27% of net revenue) grows by moving large US merchants and platforms such as Toast and OpenAI onto a stack they have not used before, so growth is share gain rather than market growth. Unified Commerce (+27% cc, 32% of net revenue) grows because the same identity and settlement stack runs online and in-store: 838K transacting terminals (+27%) and 486 merchants at scale on both channels, now with Talon.One loyalty bolted on. Platforms (+40% cc) adds issuing, capital and accounts to vertical SaaS, a second revenue layer on volume Adyen already processes (H1 2026 letter). Amex's forward engine is the premium membership refresh cycle, which works (Platinum is its fastest-growing US consumer portfolio, Q2 2026 call), but each cycle costs more in benefits and lifts a mature base by high single digits; the 2026 guide is about 10% revenue growth.
Adyen converts through operating leverage on one platform: interchange and scheme fees are passed through, so 88% of mapped revenue is gross profit, and the heavier line (people, 44% of revenue) is falling as a share because the same code serves every region; SG&A-equivalent went from 46.1% to 41.9% of revenue in two years. Amex converts through pricing power on card fees ($92 to $117 average fee per card, 2023-25, $131 in Q2 2026) but recycles it: rewards, card-member services and partner payments grew faster than revenue in 2025, and the company chose to reinvest its 2026 revenue beat rather than raise EPS guidance (Q2 2026 call). The tension worth noting is that Amex wins the power lens yet loses the margin lens: its moat is real, but it is paid for every year.
Amex's exposure is the balance sheet behind the spend: $151.8bn of card-member loans funded largely by $152.5bn of deposits, provisions already 7-8% of revenue, and discount revenue (52% of revenue) tied to consumer spending; in 2020 revenue fell 17% (10-K FY2025, FY2021). Adyen's exposure is price: tiered pricing hands take rate back as merchants scale (Platforms take rate about 12bps vs 17bps in Unified Commerce, H1 2026, inferred from letter volumes and revenue), and a few hundred enterprise merchants drive about 60% of growth, so a Stripe-led price war on the largest accounts would show up as growth and take rate falling together. In a cyclical downturn Amex breaks first because credit losses and spend fall at once; in a structural disruption (merchants steering shoppers to cheaper methods, surcharging) Amex is also more exposed, and Adyen is building the steering tool (Personalize, H2 2025 letter).
These five cells carry roughly 80% of the combined economics of both firms; every score is argued in the tabs below.
| Cell (product x region) | Adyen | Amex | Why (one clause, sourced) |
|---|---|---|---|
| US consumer premium spend | 0 | 5 | Adyen has no consumer product; Amex USCS is 48% of revenue, fees +20% and Platinum its fastest grower (10-K FY2025; Q2 2026 call). |
| Enterprise merchant acceptance, EMEA | 5 | 2 | Adyen EMEA 58% of net revenue, +22% a year 2023-25 on own EU license; Amex exited its EU network business under fee-cap regulation (AR FY2025; 10-Q Q2 2026). |
| Merchant acceptance, North America | 4 | 3 | Adyen NA +30% cc in H1 2026 from share gains; Amex network volumes +7% but discount rate slipping 2.29% to 2.24% and surcharging rising (letter H1 2026; 10-K FY2025). |
| Commercial and platform payments, US-led | 4 | 3 | Adyen Platforms +40% cc, issuing volume 8x in 2025 from a small base; Amex CS billed business +3%, revenue +7% (letters H2 2025, H1 2026; 10-K FY2025). |
| International cards and acceptance, APAC, LatAm, Canada, Japan | 3 | 4 | Adyen APAC +14% and LatAm +12% a year 2023-25, below group; Amex ICS billed business +14% and pretax income +55% in 2025 (AR FY2025; 10-K FY2025). |
Scores are anchored to the exhibits in the three tabs. Scores are per cell and are not summed; the three lenses get their verdicts in the Three Answers above.
The filings do not line up, so the map is normalized. Adyen reports pillars (Digital, Unified Commerce, Platforms) and regions (EMEA, North America, APAC, LatAm) separately, with no pillar x region matrix, so pillar-region cells are ND and the two dimensions are shown side by side. Adyen's regions follow merchant billing location, and North America includes APAC-headquartered merchants billed there (H2 2025 letter). Amex segments are already product x geography: US Consumer (USCS), Commercial (CS, US), International Card Services (ICS, all cards outside the US including Canada) and Global Merchant and Network Services (GMNS).
The contrast is stark. Adyen's two newer cells are 44% of net revenue and grew 30% and 50% in 2025; its largest cell, Digital, grew 9% reported, held back by one large customer moving volume away (volume +8% in 2025, +21% excluding that customer, H2 2025 letter). Amex's largest cell, US Consumer, is also its best: 11% a year 2023-25 with rising card fees. Its merchant and network cell is the laggard at 4%, which matters because merchant economics are exactly where Adyen operates.
| Cell | Adyen FY25 (% of NR) | Growth | Amex FY25 (% of rev) | Growth, 2y CAGR | Margin signal / leader |
|---|---|---|---|---|---|
| Engine 1: Digital / US Consumer | EUR 1,335m (56%) | +9% (+15% cc H1'26) | $34,814m (48%) | +11% | Adyen Digital take rate 16.8bps (inferred); Amex USCS pretax margin 19.6%. Amex leads on pricing, Adyen on growth. |
| Engine 2: Unified Commerce / Commercial | EUR 765m (32%) | +30% (+27% cc) | $16,926m (23%) | +7% | Adyen UC 17.3bps, highest pillar take rate (inferred); Amex CS pretax 21.7%. Adyen leads. |
| Engine 3: Platforms / International Cards | EUR 264m (11%) | +50% (+40% cc) | $13,000m (18%) | +12% | Adyen Platforms 12.3bps, tiering visible (inferred); Amex ICS pretax 12.3%, lowest segment margin. Split. |
| Engine 4: none / Merchant and Network | ND | ND | $7,759m (11%) | +2% | Amex GMNS pretax 51% on allocated segment revenue; discount rate 2.29% to 2.24%, 2023-25. |
| Geography: home / outside | EMEA 58% / 42% | +19% / +18% | US ~78% / ~22% | ND / ICS +13% | Adyen is the more global business; Amex is a US franchise with a growing international card book. |
Adyen, share of wallet. About two thirds of H1 2026 growth came from customers onboarded in 2024 and earlier; the rest from the 2025 cohort, new wins and financial products, net of volume tiering (H1 2026 letter). Management's Investor Day framing is that newer cohorts contribute 2-3x more growth in year two than in year one (Investor Day 2025 transcript). Acquired growth is now in the mix: Talon.One (EUR 750m, about EUR 60m ARR growing 30-40%) and Orb ($335m) closed on 1 July 2026 and add about 1 point to 2026 growth and 1-2 points to 2027 (M&A call 23 Apr 2026; H1 2026 letter).
Amex, membership refresh. Net card fees grew 18% in 2025 and 17% in H1 2026 on new premium acquisition, retention and product refreshes; proprietary new cards acquired fell from 13.0m (2024) to 12.5m (2025), so the lift is price and mix more than volume (10-K FY2025; 10-Q Q2 2026). ICS growth in 2026 is partly inorganic: Swisscard became wholly owned on 12 January 2026 (10-Q Q2 2026).
Adyen sells direct to enterprises and through platforms, prices interchange-plus with volume tiers, and passes scheme and interchange costs through, so it controls its margin but not its take rate: take rate moved 16.2bps (H2 2024), 17.1bps (H2 2025), 16.2bps (H1 2026) on mix and tiering (letters). Amex sets both prices in a closed loop, card fees to consumers and merchant discount to merchants, but the merchant side is being contested: discount revenue as a share of billed business slipped from 2.29% to 2.24% over 2023-25 and 2.23% in H1 2026, merchant surcharging is rising where allowed, and Australia's central bank ends the requirement to permit surcharging on 1 October 2026 and opened a review of three-party networks in June 2026 (10-K FY2025; 10-Q Q2 2026).
| Input | Adyen | American Express |
|---|---|---|
| Card network access | Depends on Visa and Mastercard scheme licenses and rules (AR FY2025 risk section) | Owns its network; exited EU and Australia network licensing under regulation |
| Funding / licenses | Own banking licenses in EU, US, UK; net cash EUR 4.9bn excluding merchant balances (H1 2026) | Customer deposits $152.5bn fund $151.8bn of loans (10-K FY2025) |
| Compute | Own data centers; capex pulled forward to 7% of net revenue in 2026 to lock in supply | Not a disclosed constraint |
| Partners | No single customer disclosed as material; concentration "within risk appetite" | Delta cobrand ~13% of billed business, ~21% of loans, contract to end-2029 |
Adyen's cells are contested by Stripe (payment volume +34% in 2025, approx., unverified), Checkout.com, JPMorgan and the legacy acquirers Fiserv and Worldpay/Global Payments, which are flat to shrinking (approx., unverified); Adyen is a share-taker. Amex's premium cells are contested by bank issuers on Visa and Mastercard rails, notably JPMorgan and Capital One (which now owns the Discover network); Amex holds share in premium and takes share internationally, with Millennials and Gen Z about 65% of new accounts (investor conference, 16 Sep 2026, via Yahoo Finance).
Adyen Digital: large merchants renegotiating tiers while Stripe moves up-market would compress take rate faster than wallet share rises; this is where Adyen is more exposed. Amex US Consumer: each refresh adds fixed benefit costs (card-member services +27% in 2025, +49% in H1 2026) that must be covered by fee growth; if premium consumers trade down in a slowdown, costs stay and fees soften. Amex Merchant and Network: regulation of three-party pricing (Australia review, EU cobrand fee-cap case in the Netherlands after the April 2026 CJEU ruling) cuts discount rate directly; Adyen gains from the same shift.
1) One global stack plus own banking licenses in the EU, US and UK; replicating licensed, single-code acquiring across regions takes years (Medium-High). 2) Cross-channel identity data: about 95% of roughly 400m Black Friday shoppers recognized in real time across online and in-store (H2 2025 letter) (Medium). 3) Peak reliability, 99.9999% uptime on 837m BFCM transactions (Medium). Stripe is building the same with money and time, so none is High.
1) Premium membership brand with proven pricing: 86.6m proprietary cards, average fee per card up 12-14% a year, $10.0bn of card fees in 2025 (High). 2) Closed-loop network seeing both sides of $1,897bn of network volume (High). 3) Rewards and cobrand ecosystem (Delta, Marriott, Hilton, Membership Rewards) (High, partner-dependent). Sources: 10-K FY2025.
Adyen: enterprises multi-source by design, which is why share of wallet rises slowly from under 20% to over 40% over a decade; the same fact means merchants keep live alternatives and win lower tiers as they grow, so Adyen's pricing power is exercised on value-added products, not on core processing (H1 2026 letter; H1 2026 call). Amex: consumers pay rising annual fees without volume collapse (fees +18%, cards-in-force +4% in 2025), which is pricing power exercised; merchants, the other side, push back through surcharging and lower discount rates, and one partner (Delta) carries 21% of loans (10-K FY2025).
Adyen's critical suppliers are the card schemes, but their fees are passed through (costs from financial institutions 7.2% of IFRS revenue in 2025), and it funds itself from equity and float, holding EUR 4.9bn of own cash (AR FY2025; H1 2026 letter). Amex's critical inputs are funding and partners: interest expense and deposit pricing move with rates, and business development payments to partners reached $6.5bn in 2025, rising with volumes (10-K FY2025). In a rate or credit shock Amex absorbs it in provisions and funding cost; Adyen's finance income falls (EUR 268m in 2025 vs 350m in 2024) but operating margin is untouched.
| Cell | Route control | Pocket price | Continuity | Outcome (share / margin) | Confirming KPI |
|---|---|---|---|---|---|
| Enterprise acceptance, North America | Adyen direct to merchant; Amex direct via GMNS and OptBlue acquirers | Adyen below legacy acquirers on tiers; Amex above bankcard rates | Both high | Adyen share up, margin flat; Amex share flat, discount rate down | Adyen NA cc growth at or above 25%; Amex discount rate at or above 2.20% |
| US consumer premium spend | Amex direct to consumer; Adyen absent | Amex fee per card rising | Amex high, Delta to 2029 | Amex share up, margin flat (VCE absorbs fees) | Amex VCE at or below 43% of revenue |
| Enterprise acceptance, EMEA | Adyen direct, own license; Amex capped by EU rules | Adyen parity to below; Amex capped | Adyen high; Amex regulatory risk | Adyen share up, margin up; Amex flat | Adyen EMEA growth at or above 15% |
1) Balance-sheet model (Major): Amex carries credit and funding risk as the price of issuing; Adyen carries neither. The gap is the business model, so neither side closes it. 2) Who pays for growth (Major): Amex spends about 43% of revenue on rewards, benefits and partner payments to grow near 10%; Adyen gives price back through tiers but keeps 88% after pass-through. Not closable without Amex weakening its value proposition. 3) Consumer pricing power (Moderate): Amex's is real and Adyen has none; Adyen's route to comparable lock-in is moving into loyalty and billing (Talon.One, Orb, Intelligent Money Movement), a 3-5 year project that money alone does not finish.
The statements are not like-for-like and the mapping is stated once. Adyen presents costs by nature under IFRS: revenue = non-interest revenue plus interest income; COGS proxy = costs incurred from financial institutions + cost of goods sold (terminals) + interest expense, so gross profit equals Adyen's net revenue; SG&A = wages, social security and pension, and other operating expenses; D&A (4.6-5.1%) sits outside; EBIT = income before net finance income and taxes, which excludes EUR 268m of 2025 finance income. Amex presents by function as a bank: revenue = total revenues net of interest expense; COGS proxy = provisions for credit losses + card-member rewards + card-member services + business development; SG&A = marketing + salaries and benefits + other, net; EBIT proxy = pretax income. Neither company discloses R&D, so it is ND for both.
| % of mapped revenue, 3y avg FY2023-25 | Adyen | American Express | Gap | What drives it |
|---|---|---|---|---|
| COGS proxy | 11.9% | 49.6% | -37.7pts | Amex pays rewards, benefits, partners and credit losses; Adyen passes interchange through |
| R&D | ND | ND | ND | Not disclosed; technology sits in staff and other costs at both |
| SG&A | 44.3% | 31.7% | +12.5pts | Adyen is people-centric (5,020 FTE); Amex spreads fixed cost over a far larger revenue base |
| Gross margin | 88.1% | 50.4% | +37.7pts | Position in the value chain |
| EBIT margin | 38.7% | 18.7% | +20.0pts | Gross gap only partly offset by Amex's lighter overhead |
| Year | Adyen revenue (EURm) | COGS | SG&A | GM | EBIT | Amex revenue ($m) | COGS | SG&A | GM | EBIT |
|---|---|---|---|---|---|---|---|---|---|---|
| 2023 | 1,863 | 12.7% | 46.1% | 87.3% | 35.3% | 60,515 | 49.4% | 33.2% | 50.6% | 17.4% |
| 2024 | 2,253 | 11.4% | 44.9% | 88.6% | 39.4% | 65,949 | 49.2% | 31.2% | 50.8% | 19.6% |
| 2025 | 2,672 | 11.5% | 41.9% | 88.5% | 41.5% | 72,229 | 50.1% | 30.8% | 49.9% | 19.1% |
The persistent gap is gross margin: about 38 points in every year, explained by position rather than one-offs. Amex earns the full merchant discount and card fees but returns roughly half to cardmembers, partners and credit losses, and that share rose to 50.1% in 2025 as the Platinum refresh added benefits. Adyen's gross line is stable near 88% because scheme and interchange costs pass through, and its operating leverage shows in SG&A falling 4.2 points in two years. This confirms the power map rather than contradicting it: Amex's pricing power is visible in card fees, but it is spent on the value proposition that creates it. The confidence caveat is Adyen's 2026 step: EBITDA margin guided about 1 point lower with acquisitions and capex at 7%, so the next two years test whether leverage survives the move beyond payments.
| Metric | Threshold | By when | If it hits, it favors | Where published |
|---|---|---|---|---|
| Adyen net revenue growth, constant currency | FY2026 at or above 21% (guide 21-23%); H1 2027 at or above 18% | Feb 2027 / Aug 2027 | Adyen | Shareholder letters |
| Adyen take rate | At or above 15.5bps in H2 2026 and H1 2027 (H1 2026: 16.2bps) | Aug 2027 | Adyen (below 15bps favors Amex) | Shareholder letters |
| Adyen EBITDA margin | FY2027 at or above 52%; FY2028 above 55% | Feb 2028 / Feb 2029 | Adyen | Annual report |
| Amex net card fee growth and average fee per card | FY2026 at or above 15%; fee per card at or above $140 by Q4 2027 (Q2 2026: $131) | Jan 2027 / Jan 2028 | American Express | 10-Q, 10-K |
| Amex discount revenue as % of billed business | Holds at or above 2.20% (FY2025: 2.24%) | FY2027 10-K | Amex if held; below 2.18% favors Adyen | 10-K Table 5 |
| Amex net write-off rate, principal, consumer and small business | At or below 2.3% (FY2025: 2.0%) | FY2027 10-K | American Express | 10-K Table 7 |