Disclosure up front: I'm Oded, co-founder of UniPaaS, the FCA-authorised Payment Institution (No. 929994) behind paas.build. So I have a horse in this race. The framework below holds whoever you pick - I've tried to be fair about when Stripe Connect is the right answer, because often it is.
The setup
You're building a marketplace, a community with paid creators, a booking app that pays venues - anything where money flows from your users to your other users. You reach for Stripe Connect, because that's the default answer. And Connect is genuinely excellent infrastructure.
But "we need Connect" is one of those decisions people make on autopilot, and for a lot of platforms it's more than they need - with real cost.
What Connect actually asks of you
Full Connect - the model where you own the platform experience - means you're on the hook for:
- Onboarding every seller through KYC yourself. Each connected account needs identity and banking verification before it can be paid. You build and maintain that flow.
- Platform liability. With the common separate-charges-and-transfers setups, chargebacks and negative balances can land on your platform account.
- The build. Connect onboarding, payout scheduling, the dashboard your sellers expect - that's weeks of engineering, and it isn't your product.
- Time to first payout. Your sellers can't get paid until they clear Stripe's verification, and newer platforms get held for review.
If your team is happy owning all of that, Connect is a great choice - use it. The rest of this is for platforms that reached for Connect and then realised they'd signed up to build a payments company on the side.
What most platforms actually need
Strip it back and the requirement is usually just two things:
- Take a payment from a buyer.
- Split it and pay out to one or more of your own users, on a schedule, minus your cut.
That's it. You don't necessarily need to become the payments platform to do that.
The lighter path: sub-merchant payouts on a PayFac
A payment facilitator (PayFac) can onboard your users as sub-merchants and split each payment to them directly, with the regulated liability sitting on the institution rather than on you. In practice:
- Progressive KYB onboards your sellers in-app and lets them go live the same session with a cap, verification completing in the background - instead of a hard gate before anyone can transact.
- Real-time splits to unlimited sub-merchants, so a booking or sale pays the venue or creator and takes your fee in one move.
- The liability sits with the licensed institution, not your platform balance.
- You don't build Connect. Onboarding, splits and payouts are the provider's surface, not your roadmap.
| Stripe Connect (full) | PayFac sub-merchant (e.g. paas.build) | |
|---|---|---|
| Seller onboarding | You build the KYC flow; hard gate before payout | Progressive KYB in-app; live the same session with a cap |
| Who holds liability | Can land on your platform account | Sits with the licensed institution |
| Time to first payout | After each seller clears verification | Same session, capped until verified |
| What you build | Onboarding, splits, payout dashboard | Little - it's the provider's surface |
| Best for | Deep marketplace customisation, at scale | Platforms that just need to pay out their own users |
Honest limitations
- Region. paas.build serves builders in the UK, EU and US. Buyers can pay from anywhere, but if your platform entity is elsewhere, this doesn't unblock you.
- Tax. You and your sellers stay the merchants of record, so tax is yours (with tooling). If you want tax fully outsourced, that's a Merchant of Record - a different category, and not built for paying third parties anyway.
- Scale. If you're at Uber scale with bespoke flows, full Connect's flexibility earns its complexity.
When Stripe Connect is still the right call
Be honest with yourself: if you need deep marketplace customisation, you're already deep in the Stripe ecosystem, or you want to own the payments platform as a differentiator - use Connect. It's the most flexible tool in the category.
But if you reached for Connect just to pay out to your own users, and you'd rather ship your actual product, a PayFac with sub-merchant payouts is the lighter path. I wrote up the specifics for our version here: https://paas.build/stripe-connect-alternative
Top comments (1)
The progressive-KYB point is the real one here. The hard gate before payout kills more programs than the fee does.
One thing I'd add from the affiliate-payout side: a lot of platforms reach for real-time splits before they need them. Affiliate commissions usually sit behind a refund or clawback window, so you're paying on a 30 to 60 day delay anyway. At that timing a monthly batch through Wise or PayPal covers a surprising amount of ground before either Connect or a PayFac earns the integration. You move to sub-merchant onboarding when the manual side actually starts to hurt, not on day one.
(I build affiliate software on Stripe, so payout timing is a hill I think about a lot.)