# Decoding take rates and interchange economics
A customer buys $100 of sneakers with a debit card at a store running on your fintech platform. By the time the money settles, your startup might keep somewhere between 30 and 80 basis points, roughly 30 to 80 cents. The other roughly $2 to $2.50 in fees flowed to parties you never see on the checkout screen. Understanding where that money goes, and why, is the difference between a fintech that scales profitably and one that quietly bleeds margin on every swipe.
A basis point (bp) is one hundredth of a percent. So 200 bps equals 2 percent. Payments people live in basis points because the numbers are small and the volumes are enormous.
Card payments run on a four-party model. For our $100 sneaker purchase:
Sitting in the middle is the card network (Visa, Mastercard). The network does not set the biggest fee, but it sets the rules.
Here is roughly how the $100 splits on a typical US credit transaction. Exact numbers vary widely by card type, merchant category, and country, so treat these as illustrative.
| Party | Fee taken | What it is |
|---|---|---|
| Issuer | ~$1.50 to $2.00 | Interchange |
| Network (Visa/Mastercard) | ~$0.10 to $0.15 | Network assessment |
| Acquirer / processor | ~$0.10 to $0.30 | Processing markup |
| Merchant keeps | ~$97.50 to $98.30 | Net of the "merchant discount rate" |
The total the merchant pays, often called the merchant discount rate (MDR), might land around 1.7 to 2.5 percent on that credit card swipe. Debit is usually much cheaper. More on why in a moment.
Interchange is the fee the merchant's side pays to the cardholder's bank (the issuer) for every transaction. It is the largest component of card fees, and the crucial point for fintech founders is this: you do not negotiate it. The card networks publish interchange rates.
You can see the real numbers yourself. Visa and Mastercard post their US interchange schedules publicly. Visa's is available here via its rules and rate documents.
Interchange exists because the issuer takes on risk (fraud, credit losses, funding the transaction) and the network wants issuers motivated to hand out cards. Higher interchange means banks earn more per swipe, so they push their cards harder. That is why your premium travel rewards card carries higher interchange: the bank funds those rewards partly from the merchant.
In the US, the Durbin Amendment (part of the 2010 Dodd-Frank Act) caps debit interchange for banks with more than $10 billion in assets. The cap is roughly 21 cents plus 0.05 percent of the transaction, plus a small fraud adjustment. On our $100 debit swipe from a large bank, that is around 22 to 25 cents total, dramatically less than credit.
This one rule shapes entire business models. Many US fintechs deliberately partner with small issuing banks (under $10 billion in assets) that are exempt from the Durbin cap. Those "exempt" banks can earn full, uncapped debit interchange, often over 1 percent, and share it with the fintech. This is the quiet engine behind a lot of "free" neobank and debit card products.
Your take rate is the percentage of payment volume your fintech keeps as revenue. Two very different structures produce it.
Think of a platform like a payments processor or a software company embeddingembeddingAn embedding is a numerical vector that represents data (text, images, or items) in a way that captures meaning, so similar items sit close together in space.Voir la définition complète → payments (this is often called payfac, short for payment facilitator, where you aggregate many small merchants under your own account).
You charge the merchant, say, 2.9 percent plus 30 cents (a common published rate). Out of that you pay interchange (~1.8 percent), network assessments (~0.13 percent), and your processor's cut. What is left is your gross take rate, often 30 to 80 bps once costs are stripped out.
The lesson: your headline price looks like 290 bps, but interchange and network fees are pass-through costs. Your real, keepable margin is a thin sliver on top.
Now flip sides. Your fintech issues a debit card to consumers or businesses through a sponsor bank. Here interchange is revenue *to your side*, not a cost.
On an exempt-bank debit program, the interchange might be ~1.2 percent. The sponsor bank and your processor (for example, a card issuing platform) take a share. You might net 60 to 100 bps. This is why business spend management and expense card fintechs love the issuing model: every dollar a customer spends on the card generates interchange revenue for the fintech.
Founders obsess over pricing pages. The real levers are structural.
1. Card mix. Credit generates more interchange than debit. Commercial cards generate more than consumer cards. A fintech skewed toward business credit spend has a structurally higher take rate than one pushing consumer debit, before any negotiation.
2. Bank partner and sponsor economics. On the issuing side, whether your sponsor bank is Durbin-exempt is often the single biggest determinant of revenue per swipe. On the acquiring side, your volume determines how good a wholesale rate you get from your processor.
3. Transaction type and channel. Card-present (in store, chip tapped) transactions carry lower interchange and lower fraud than card-not-present (online, keyed in). E-commerce fintechs live in the more expensive, higher-fraud world, which raises costs and can compress net take.
4. Volume and scale. Networks and processors offer better wholesale pricing at higher volumes. A fintech doing billions in annual volume negotiates rates a startup cannot touch.
5. Value-added services. The durable way to raise take rate is to stop competing on the swipe. Fraud tools, instant payouts, lending, foreign exchange, and analytics carry margins that dwarf 50 bps of interchange. Payments becomes the wedge; software and financial services become the profit.
6. Cross-border and FX. International transactions add cross-border and currency conversion fees. Fintechs serving global merchants or travelers earn meaningfully more per transaction here, though costs and compliance rise too.
Vérification des acquis
1. In the four-party card model, interchange represents a fee flowing from which party to which party?
2. Why do payments professionals typically express fees in basis points rather than percentages or dollars?
3. A fintech founder notices the platform's take rate per swipe is small but the business is still profitable at scale. What best explains this?
4. Select ALL correct answers about the roles in the four-party card model.
Sélectionnez toutes les réponses correctes.
5. Select ALL correct answers about interchange and merchant costs.
Sélectionnez toutes les réponses correctes.
When you evaluate a payments fintech, do not trust the headline "we process $X billion." Ask three questions.
What is net revenue, not gross? Many payments companies report gross payment volume (GPV) and gross revenue that includes pass-through interchange. The number that matters is net revenue after interchange and network fees. A company reporting 2.5 percent "revenue" on volume may keep only 0.5 percent.
Is the take rate rising or falling, and why? Rising because of value-added services is healthy. Falling because a large low-margin enterprise merchant now dominates volume is a mix shift, not necessarily a problem, but you must know which.
How concentrated is the volume? Large merchants negotiate rates down hard. A fintech dependent on a few big accounts has structurally thinner and more fragile take rates than one serving thousands of small businesses.
A fintech reports $10 billion in annual card volume and $250 million in "revenue." That looks like a 2.5 percent take rate. Impressive.
But if $200 million of that is interchange and network fees passed straight through to banks and Visa, the real net take rate is $50 million on $10 billion, or 0.5 percent (50 bps). Suddenly the economics, and the required scale to profit, look very different.
This gap between gross and net is the most common way payments metrics mislead.