Bring your own Stripe, explained

Bring your own Stripe means the platform you build on connects to a Stripe account you own, using your API keys. Customers pay you directly, the money lands in your Stripe balance, and the platform never holds it. The alternative is a merchant of record. They become the legal seller, and they pay you out.

Almost every write-up of this choice frames it as a fee comparison. Keep your own Stripe, keep the 5%. That framing is not wrong, exactly. It is the smallest part of what you are deciding, and the arithmetic underneath it is a lot closer than the headline makes it look.

What a merchant of record actually does

A merchant of record is the legal seller of your product. Not a payment processor and not a billing tool: in Stripe's own words, "the entity that takes legal and financial responsibility for a sale". They remit the applicable taxes. They handle the disputes and absorb the fraud liability. You sell to them, in effect, and they sell to your customer, and then they pay you out on a schedule.

That is a real service and it has a real price. Paddle and Lemon Squeezy, the two best known, both publish the same headline figure: 5% plus 50 cents per transaction (verified 22 October 2026 on paddle.com/pricing and lemonsqueezy.com/pricing). Paddle's own FAQ says products under $10 need bespoke pricing, which tells you where that fixed 50 cents starts to hurt.

The arrangement also shows up in small places. A Paddle charge reaches the customer's bank statement as PADDLE.NET* followed by your business name, trimmed to between two and ten characters (verified 22 October 2026 in Paddle's statement descriptor reference). Your brand is the suffix on someone else's prefix.

How bring your own Stripe works

You paste two keys into the platform and point a webhook at it. That is most of the mechanism.

The publishable key and the secret key come from your Stripe dashboard. The platform stores them, then uses the Stripe API to create customers, prices and subscriptions inside your account, as you. A webhook endpoint pointed back at the platform's server is how it hears about a payment succeeding, a card failing, a subscription being cancelled from Stripe's side. Open your Stripe dashboard afterwards and every subscription the platform created is sitting there, under your business, in your own list of customers.

So what is the platform actually doing, if Stripe holds the subscriptions? The part Stripe does not: plan versioning, entitlements, quota counting, the upgrade and downgrade rules, the trial clock, and all the product logic that decides what a customer is allowed to do between invoices.

BuildBase works this way. Stripe credentials are stored per organization, the secret key and webhook secret encrypted at rest, with a per-organization toggle for Stripe Tax. We are not a merchant of record and we take no percentage of what you bill. That is a design choice with trade-offs, and the rest of this post is the honest version of them.

The arithmetic, with every assumption shown

Here is the comparison people quote, on a product doing $10,000 a month across 100 charges of $100, all on domestic US cards.

A merchant of record at 5% plus 50 cents takes $500 on the percentage and $50 on the fixed fee. That is $550 a month.

Your own Stripe account, taking payments only, costs 2.9% plus 30 cents per successful domestic card charge (verified 22 October 2026 on stripe.com/pricing). That is $290 plus $30, so $320 a month.

A gap of $230 a month. This is the number the comparison posts stop at, and it is where they start being misleading, because the two columns are not buying the same thing yet.

The merchant of record included subscription management and tax handling. Buy those back and the picture moves:

• Stripe Billing, if you run your subscription logic through Stripe itself, is 0.7% of billing volume on the pay-as-you-go plan. On $10,000 that is $70.
• Stripe Tax is 0.5% per transaction on the no-code integration, or 50 cents per transaction on the API one. Call it $50.

Run both and your own-Stripe column is $440. The gap is now $110 a month, or $1,320 a year.

The 0.7% line is the one you can actually remove, and it is worth knowing why. If the platform you build on handles plans, versioning and metering itself, it is writing to Stripe's payments API rather than running subscriptions through Stripe Billing, so that fee never applies. Drop it and you are at $370 against $550, a gap of $180 a month. That is the strongest honest case for the model, and it is still a long way from what "keep the 5%" promises.

Two things move this further, both against your own Stripe account. Stripe adds 1.5% for international cards, so a customer base that is 40% outside the US adds about $60 to that column while the merchant of record's 5% stays flat. And at low ticket prices the fixed fees dominate everything: on a $5 product, 30 cents is 6% before the percentage fee even starts.

Run your own numbers before you believe mine. The crossover logic is the same one in the hidden revenue tax of billing platforms, which works through when a percentage beats a flat fee and when it stops.

When a merchant of record is the better answer

Tax. Almost always tax.

Selling software across borders means registering to collect in each place that says you must, filing returns there, and keeping up as thresholds change. Stripe Tax calculates the right amount at checkout, which is genuinely useful, but look at how the pricing is worded: per transaction, "where you're registered to collect taxes". The registering is yours. The filing is yours. A merchant of record does both, and if you are a two-person company selling into thirty countries, handing that over for a few points of revenue is not a bad trade. It might be the best money you spend.

Chargebacks are the other one. With your own Stripe account, a disputed charge is your dispute, your evidence to submit, your fee if you lose. A merchant of record takes that liability on.

There is a third case nobody writes about: access. Stripe does not operate everywhere, and the requirements for opening a business account vary enough by country that plenty of founders simply cannot get one. Then the choice has already been made for you.

So the model is a bad fit when you are small, selling globally, and would rather not become an amateur tax department. It is a good fit when your sales are concentrated in places you are already registered, when the volume makes a percentage expensive, or when you care about the thing in the next section.

What you are really buying

Ownership of the billing relationship, and the exit.

Under a merchant of record, the customer bought from them. Your subscriptions live in their system, your payout history is their record, and leaving means migrating customers who, legally speaking, were never yours. Under your own Stripe account, the subscriptions were in an account with your name on it the whole time. Move off the platform and the billing keeps running, because the platform was only ever an API client holding your keys.

That is worth more than $110 a month to some people and nothing at all to others, which is why this decision cannot be settled with a fee table. It depends on whether you expect to change platforms, and on whether an invoice with someone else's name on it bothers you.

Where this sits next to the other billing decisions

Bring your own Stripe decides who holds the money. It does not decide how you charge for anything, which is a separate question with its own answer: see what is usage-based billing for the pricing model, and what are entitlements for the layer that decides what a paying customer can actually do.

The one connection worth drawing is that these compound. A platform taking a percentage of revenue, on top of a processor taking a percentage of revenue, on top of a tax tool taking a percentage of revenue, is three lines on your P&L that all grow when you succeed. Knowing which of them you have actually signed up for is most of the work.