The 5% decision: why we built Seedr's fee structure the way we did

When a percentage doesn't feel like profit

The early versions of Seedr had no fee at all. We'd pull it from wherever we could. That lasted about three weeks into live usage. We learned quickly that no fee meant no incentive to fix anything fast, and creators started asking harder questions about where the money actually went.

By October, we had a spreadsheet. One column was what Stripe took (2.4% + 20p per transaction in the UK). Another was what we needed just to keep the lights on: infrastructure, monitoring, the weekly payout worker that runs every Monday morning, compliance tracking. The third column was blank, and I knew we had to fill it.

Most platforms we admired were taking 20-30%. Patreon takes 5% for monthly memberships. But Seedr isn't Patreon. We're about micro-tips during a moment. A £5 donation to a church live-stream. A £10 tip to a comedian on Giggl. The psychology is different.

I kept thinking about that pastor's question. Five years is long enough to stop being a startup and start being infrastructure. Infrastructure that takes 5% feels honest to me. Not lean. Not aggressive. Just honest.

The Foundr gravity well

Then we shipped Foundr, our maker platform. And suddenly we had creators who were already paying for Foundr Pro, who'd built an audience there, who trusted the ecosystem. For those creators, we knew the math was different. They'd brought users into MRVL. They'd invested.

So we built tiered fees into the same backend that handles everything else. If you're a Foundr Free maker, it's 1.5%. If you're Pro, it's 1%. The fee drops as your commitment to the ecosystem goes up. This sounds simple on paper, but it meant every schema decision from day one had to account for this. Every transaction record in Supabase needed to know which tier applied at that moment.

That's why we built the fees.ts file as a single source of truth. Not scattered across three services. One place. Auditable. When we go for FCA Payment Institution status in 2028, an auditor will be able to see exactly how every pence was routed, to the integer, with no ambiguity.

The pastor would have paid 5%. The Foundr Pro maker would have paid 1%. Both felt right.

Stripe Connect isn't our final answer

Here's what people misunderstand about Seedr. We're not FCA-authorised yet. We're not MRVL Pay. Not yet. Right now, Stripe Connect is holding the keys. It's the licensed bridge between us and the payment system. That's fine for now. It's also temporary.

By 2028, we want MRVL to be its own Payment Institution. That means MRVL Pay. That means we own the compliance, the auditing, the fee structure entirely. When we get there, the fee might change. It might not. But it will be ours to own, which means it will be ours to be accountable for.

The 5% we chose today has to survive scrutiny that a Payment Institution faces. Every percentage point has to map to a real cost or a real value we provide. That's not theoretical for us. It's built into how we've designed the ledger. INTEGER pence throughout. No rounding mysteries. No offshore fee shuffling. Just honesty.

A creator looking at 5% in 2025 should feel confident they're seeing the same structure we'll defend in 2028.

The micro-tip problem nobody talks about

Stripe takes 2.4% + 20p. At £5 (our minimum tip), that's 2.4% of a small number plus a fixed cost that stings. The 20p is brutal for small amounts. That's why we built the minimum at £5 Seeds in the first place. Below that, the fee math destroys the gesture.

Our 5% standalone fee was built to sit on top of Stripe's costs and still leave creators money that matters. A £5 tip, minus Stripe's 2.4% + 20p, minus our 5%, leaves £4.34 for the creator. That's not nothing. That's real.

Compare that to the Foundr Pro creator tipping at £20. Stripe takes their cut. We take 1%. The creator gets £19.12. Scale matters here. The bigger the tip, the better the ratio looks. We see people tipping £50, £100 during live events. At those sizes, our 1% fee for Foundr Pro creators is a whisper compared to Stripe's cut.

The 5% for standalone creators needs to hold its own on principle: we do real work to keep this running, and we should say so directly.

What the fee actually buys you

Last week, a faith creator on Streamr asked why the fee existed at all. I walked her through what she gets for it. The 3-line SeedrButton embed that took our team two months to build. The web profile at seedr.app/@handle that's on our servers. The creator dashboard that shows her where tips came from, when, and how much. The weekly Monday payout worker that never misses a cycle. The AML review at £10,000 that keeps both of us safe. The monitoring that wakes someone up at 3am if the transaction log gets weird. The Supabase schema that's ready for Payment Institution audit in 2028.

She asked if 5% was worth it. I said: it's not about worth. It's about whether you believe the infrastructure is real. If you do, 5% is the cost. If you don't, we're too expensive. That's fair.

She's still using Seedr.