The math that changed how we built Invoicr

The moment I understood the real cost

I didn't start MRVL Technologies to build invoicing software. I started it because I kept watching people I knew work themselves senseless and then lose money to payment processors they'd never really chosen.

A plumber, electrician, or decorator typically operates on a 10 to 15 percent margin if they're doing well. When a card processor takes 2 to 2.5 percent of every payment, that's real money. On a £2,000 job, you're looking at £40 to £50 per invoice. Do that twice a week and you're paying the processor nearly £4,000 a year. That's a week's profit, gone.

The reason card fees exist is partly legitimate. Card networks need fraud protection, chargebacks happen, there's infrastructure to maintain. But for a sole trader waiting five days to get paid anyway, that infrastructure cost shouldn't be priced the same way as a high-street shop taking contactless payments from 200 customers a day.

UK open banking changed that equation. Bank-to-bank payments are direct. No middleman network, no chargeback risk in the same way, no card processor sitting between you and your money. The actual cost to move money from one UK account to another is pennies. We charge a flat fee because that's what it costs us. On a £500 invoice, that's £4. On a £2,000 invoice, it's still £4.

Why we didn't go the easier route

Building on top of Stripe or Square would have been simpler. Those platforms handle everything. We'd integrate their payments, take a small commission, and be done. The feature set would be predictable, the support structure already exists, and we wouldn't have to maintain payment infrastructure ourselves.

But we would have inherited their fee structure. And then our product would have quietly become another layer that costs tradespeople money without them really noticing. We'd be complicit in the same problem we were trying to solve.

When we built Invoicr, we chose to own that relationship. We integrated directly with UK open banking. Your client sees an invoice with a payment link. They click it, log into their bank, and approve the transfer in seconds. Their bank handles the moving of money. We handle the rest. The fee is transparent, it's predictable, and it's roughly a third of what card processors charge.

That decision meant building more infrastructure ourselves. It meant navigating bank APIs, handling compliance differently, thinking carefully about fraud. It's messier than reselling someone else's payment product. But it's the right choice for people working on thin margins.

The numbers actually tell a story

Let's be concrete. A decorating company invoicing 15 clients a month at an average of £800 each is generating £12,000 in monthly invoice value.

With a card processor at 2.2 percent: that's £264 a month in fees, or £3,168 a year.

With Invoicr's bank-to-bank payment at a flat fee per transaction: that's roughly £60 a month, or £720 a year.

The difference is £2,448 a year. On a business with a typical 15 percent margin, that's like picking up an extra two weeks of work profit without doing anything differently.

I'm not saying that number works for everyone. A coffee shop taking fifty card payments a day isn't our customer. But if you're a sole trader, a small team of tradespeople, or a consultant sending invoices and waiting for bank transfers anyway, that number matters. It compounds.

We built Invoicr's free tier to let people test this without risk. Five invoices a month and three customers. That's enough to see whether bank-to-bank payments fit your workflow. If you're invoicing more than that, the Pro tier adds unlimited invoices, automated payment reminders, and the ability to send invoices via WhatsApp. You're paying £9.99 a month and getting back what card processors would have charged you by week two.

The thing we didn't expect

After we launched, something happened that we didn't design for. Clients started paying faster.

There's friction in card payments. A customer receives an invoice, they have to dig out their card, type details, worry about PCI compliance, wait for confirmation. With bank-to-bank, they log into something they already use and approve a transfer. It's the interface they trust.

We didn't add a "payment gamification" feature or a reminder sting. The speed came naturally. When payment is literally inside your bank app, you're less likely to misfile the invoice or forget about it. The friction that card processors created actually slowed down your cash flow.

That wasn't the main reason we chose bank-to-bank, but it's become one of the things users mention most. "My customers are paying me in three days instead of two weeks," someone messaged last month. Faster payments mean less time chasing debts. Fewer conversations about whether money's on its way.

Who this is actually for

Invoicr isn't for everyone, and we're okay with that. If you're a high-volume business taking payment from hundreds of different sources, you probably need a more complex payment infrastructure. If your customers expect to pay by card, we're not the fit.

But if you're a plumber in Manchester, an electrician in Edinburgh, a cleaner in London, a consultant sending three to twenty invoices a month, or a small team of tradespeople with a couple of people invoicing clients: we built this for you.

The free tier costs nothing. You get to test whether bank-to-bank payments work with your customers. If they do, and you need more than five invoices a month, the Pro tier is £79.99 a year, which is less than card processors would charge you in the first two months. If you're a team, the Business tier adds compliance for VAT and CIS, multiple seats, and accountant export.

We're mobile-first because you're probably invoicing from the building site, the van, the kitchen of someone's house. The app is native on iOS. It's built for UK regulations because we're UK-focused, not trying to be Stripe for the world.