Stripe Keeps 40 Cents on a Hundred Dollars. That's Not the Business — It's the Doorway.
Everyone reads Stripe's '2.9% + $0.30' and assumes it pockets ~3% of every sale. It doesn't. After interchange and network fees, Stripe keeps roughly 0.40% — about 40 basis points. The fat margin lives in everything it sells on top of the rail.
Comes with a free Profit-Engine Map template.
A founder reads Stripe's pricing page, sees "2.9% + $0.30," and does the quick math everyone does: nearly three cents of every dollar I sell goes to Stripe. It is the most repeated number in fintech, and it is almost entirely wrong about what it implies. Stripe collects that 2.9%, yes — but on a hundred-dollar sale it gets to keep about forty cents.4 The rest leaves almost as fast as it arrives, paid out to banks and card networks Stripe doesn't own and can't negotiate with. The headline is a toll Stripe announces. The toll mostly belongs to someone else.
The official story is that Stripe is a payment processor that skims roughly 3% off the internet's commerce. The real story is that the 3% is gross revenue Stripe collects and largely hands away — and that the company most people picture as a fat-margin tollbooth actually runs a near-pass-through payment rail with a software business bolted on top, and the software is where the money is.
Where the 2.9% actually goes
Watch a single card payment break apart. Stripe's own explainer is blunt about the plumbing: a card transaction's fee is split three ways — the issuing bank takes interchange, the card network takes an assessment, and the processor takes a markup.6 Interchange is the biggest slice, it's set by the networks, and no merchant — and no Stripe — can negotiate it down.6 So when Stripe quotes 2.9% + $0.30, it is quoting the full stack: the bank's cut, the network's cut, and its own thin margin, all bundled into one clean number for the merchant's convenience.5 Strip the pass-through out and what's left for Stripe is about 0.40% of the volume that flows through it.4 The simplicity of "2.9% + $0.30" is the product. The economics underneath are anything but simple.
| What it's called | Who keeps it | Negotiable? | |
|---|---|---|---|
| Largest slice | Interchange | The cardholder's issuing bank | No — set by networks |
| A thinner slice | Assessment | The card network | No |
| What's left | Markup / net take | Stripe (~0.40%) | Custom for big merchants |
| Merchant sees | 2.9% + $0.30, bundled | — | — |
Stripe processed $1.4 trillion in volume in 2024.1 At the headline 2.9% that would imply tens of billions of pure revenue — which is exactly the error that floods the coverage. At the real ~0.40% net take-rate, the math lands near Stripe's reported net revenue of roughly $5.1 billion for 2024.34 The difference between those two numbers is everything Stripe never gets to keep.
The single most common mistake in payments coverage is multiplying volume by the sticker fee. A company can 'touch' $1.9 trillion and keep a few billion of it — the rail is mostly a conduit. Whenever you see a processor's volume times its quoted rate, assume you're looking at gross, then ask what survives interchange and network fees. With Stripe, conflating the two overstates the business by roughly three to four times.
The rail is the doorway. The software is the house.
If forty basis points were the whole story, Stripe would be a low-margin utility, not a company valued at $159 billion.8 The thin rail is the point of entry — once a business runs its payments through Stripe, Stripe sells it everything else a company needs, and those products carry software margins, not pass-through margins. The standard checkout fee may have no setup or monthly cost, but real-world merchants stack on layers: international-card surcharges, currency conversion, Radar for fraud screening, Billing for recurring revenue, Connect for paying out to platforms.5 Each is a separate line, each is higher-margin than the swipe itself, and each gets stickier the deeper a company embeds. The Revenue suite alone — Billing, Invoicing, Tax — is on track to reach a $1 billion annual run rate in 2026.2 That is the business hiding behind the 0.40%.
The depth of that embedding shows up in who can't leave. Stripe says 90% of the Dow Jones Industrial Average and 80% of the Nasdaq 100 run on it, and a full quarter of new Delaware corporations are now incorporated through Stripe Atlas.2 A company born inside Stripe Atlas, billed through Stripe Billing, paid through Stripe Connect, and protected by Stripe Radar is not a merchant shopping for the cheapest forty basis points. It is a tenant. And the 2024 acquisition of stablecoin platform Bridge for $1.1 billion — Stripe's largest deal ever, at roughly 5.5 times Bridge's $200 million Series A valuation from months earlier — is the same move in a new room: own the rails for the next form money takes, then sell software on top of those too.7
Isn't a 0.40% take just a bad business?
The fair objection is that 40 basis points is a terrible margin to build an empire on — a near-commodity rail anyone with a banking partner could underprice. And on the rail alone, that's correct; payment processing is a race to the bottom, which is precisely why Stripe doesn't try to win it on price. The honest counter is that the thin rail is doing the work of customer acquisition, not profit. It puts Stripe inside the financial guts of millions of businesses at a price that's hard to argue with, and then the higher-margin software — fraud, billing, tax, payouts, soon stablecoins — does the earning. The proof is in the cash, not the take-rate: Stripe was profitable in 2024 and expects to stay that way, with free cash flow doubling to $2.2 billion.13 A genuinely bad business does not throw off that kind of cash. A thin rail that sells a thick stack does.
Stripe's model is a lesson in where to put your margin. Make the thing everyone compares on — the swipe — nearly a pass-through, so price stops being a reason to leave. Then build the high-margin products that only make sense once a customer is already inside: billing, fraud, tax, payouts. The trap is mistaking the doorway for the business and trying to fatten the toll; that just invites a cheaper rail. The discipline is the opposite — keep the entry thin on purpose, and let the embedded software compound. The deeper a customer builds on you, the less the headline rate ever matters.
So the next time someone says Stripe makes 3% on the internet, remember the forty cents. The 2.9% is a number Stripe collects and mostly gives away — to banks and networks it can't out-negotiate, on a rail it deliberately keeps thin. The genius was never the take-rate. It was recognizing that the cheapest way into a company's books is to barely charge for the door — and that once you're inside, you can sell them the whole house. Stripe doesn't win the swipe. It wins everything that happens after it.
Profit-Engine Map
A one-page map that pulls a business apart into the hook that gets the customer in the door and the engine that quietly earns the margin. Use it to see where the real profit lives, how the two halves are wired together, and what breaks if the link is cut. Blank to dissect your own P&L; filled as the worked example of a business whose advertised product is not where it makes its money.
The worked example unlocks with a subscription. See plans →
Sources
Where this comes from — the filings, records, and reporting behind it.
- 1Stripe processed $1.4 trillion in total payment volume in 2024, up 38% year-over-year, equivalent to ~1.3% of global GDP; Stripe was profitable in 2024 and expects to remain so.Stripe, Stripe Annual Letter 2024 ↗ · 2025-02-27
- 2Stripe processed $1.9 trillion in total payment volume in 2025, up 34% from 2024; the February 2026 tender offer valued Stripe at $159B; Stripe's Revenue suite (Billing, Invoicing, Tax) is on track to hit $1B annual run rate in 2026; 90% of the Dow Jones Industrial Average and 80% of the Nasdaq 100 use Stripe; 25% of all Delaware corporations are created via Stripe Atlas.
- 3Stripe's net revenue was approximately $5.1B in 2024 (up ~28% YoY from ~$4.0B in 2023); free cash flow doubled to $2.2B in 2024.
- 4Stripe's core checkout stack converts ~3% gross fees into a ~0.40% net take-rate after interchange, network, and bank-partner costs; at its February 2025 valuation of $91.5B, Stripe was valued at ~17.9x its 2024 net revenue of $5.12B; Sacra estimates $6.9B net revenue and $1.2B EBITDA in 2025; 2024 pre-tax profit was $101.9M vs. a $1.2B pre-tax loss in 2023.Sacra, Stripe revenue, valuation & funding ↗ · 2026-04-08
- 5Stripe's standard US pricing for online card payments is 2.9% + $0.30 per transaction; no setup or monthly fees; custom/enterprise pricing available for high-volume merchants; additional fees apply for international cards, currency conversion, Radar, Billing software, and Connect payouts.Stripe, Pricing & Fees ↗ · 2026-04-20
- 6Transaction fees are split among the issuing bank (interchange), the card network (assessment fee), and the payment processor (markup); interchange fees are set by card networks and cannot be negotiated by merchants; interchange fees make up the largest share of a card transaction's cost.
- 7Stripe announced on October 21, 2024 it would acquire stablecoin platform Bridge for $1.1B (Stripe's largest acquisition ever); the deal closed February 4, 2025; Bridge's Series A in August 2024 valued it at ~$200M post-money, implying a ~5.5x acquisition premium; Bridge's customers at announcement included Coinbase and SpaceX.
- 8Stripe's valuation peaked at $95B in 2021, fell to $50B in March 2023 (Series I, $6.5B raise), recovered to $91.5B in February 2025 via tender offer, and reached $159B in February 2026 via a new tender offer led by Thrive Capital, Coatue, and a16z — a 74% increase in one year and a new all-time high.
- 9Stripe's Dublin-based entity recorded pre-tax losses of $1.2 billion in 2023, per statutory accounts filed under Irish law; the same accounts confirm $1.4 trillion in total payment volume and profitability in 2024.