What we got wrong about the free tier 5 invoice cap
Three weeks after launch, a plumber in Manchester sent us a message. He'd hit his monthly invoice limit on a Tuesday. Not because he was trying to break our system. He just had a busy August.
The number made sense on a spreadsheet
When we were designing Invoicr's free tier, five invoices per month felt like a reasonable gate. It was generous enough to let someone try the app without hitting a wall immediately. It was low enough that we could confidently afford to serve that user for free, long-term, without watching the cost of payment processing and server time spiral.
The logic was sound. The math worked. We even had a little internal celebration about it. Five invoices, three customers. A perfect starter tier that wouldn't drain us and wouldn't frustrate someone genuinely evaluating the app.
Then real people started using it.
The plumber working from his van
That plumber's message stuck with me. He wasn't angry. He was just pointing out that he'd issued five invoices in three days because he'd finished five jobs. A good week. In his world, a good week is five jobs across maybe three different customers. He wanted to keep using Invoicr. He actually liked it. But he was now either stuck waiting until September 1st or switching back to his old spreadsheet system.
We had dozens of similar messages over the first month. Electricians sending invoices to multiple sub-contractors on the same job. A cleaner with a corporate contract issuing invoices for different properties. A gardener who did three quotes that all turned into billable work in the same week.
The cap wasn't protecting against abuse. It was just getting in the way of people doing ordinary work.
What the cap was actually supposed to solve
Here's what we were really worried about, if I'm honest: we didn't want someone running a full invoice service for 500 clients on our free tier, just bleeding us dry on payment processing fees. We wanted to avoid the free user who was actually a business, extracting value at scale without contributing anything back.
But a five invoice limit doesn't really stop that person anyway. Someone serious about running 500 clients for free would just set up multiple accounts. It's not a technical barrier. It's an annoyance, and annoyances hit your actual customers a lot harder than they hit the people you're trying to exclude.
We were trying to solve a hypothetical problem by making a real problem for the people we actually cared about.
What we changed, and why it still matters
We didn't remove the cap entirely. That would have been naive. But we lifted it to something that actually reflects how the tradespeople using Invoicr work. The free tier now covers unlimited invoices per month, but keeps the three customer limit. That's the real throttle. You're not running a factory for free. But you're also not blocked from finishing a job on a Tuesday.
The Pro tier picks up where that gets tight. Unlimited invoices, unlimited customers, plus the stuff that busy tradespeople actually asked for: payment reminders so you're not chasing late payers, WhatsApp delivery so you're not relying on email, quotes so you can lock in a price before you start the work.
The money side of it changed, too. When we did the maths with a higher free invoice limit, we realised the payment fees weren't actually the constraint we thought they were. Most free users never make large-value invoices. The real cost was the infrastructure. And the infrastructure scaled fine. We were just being overly cautious.
The lesson was about listening, not metrics
Here's what I got wrong: I treated the free tier cap like a pure product decision, when it was really a question about who we were building for. I was optimising for a number - the cost to us per free user - when I should have been optimising for whether the app actually solved someone's problem on day one.
If Invoicr blocks a plumber on Tuesday because he's done too many jobs, we've failed him. It doesn't matter how cheap he is to serve. We've just told him the app doesn't work for his actual life.
The second thing I got wrong was assuming the cap would be self-selective. I thought people who hit the limit would naturally upgrade. Some did. Most just left. They figured they'd come back when they had a bigger business. Except when someone quits at that point, there's no relationship to build on. They're not your customer. They're a person who tried your app and it didn't fit.
We still cap the free tier now. But it's a cap that aligns with how people actually work, not with what makes our spreadsheet look healthiest. Does your free tier actually solve someone's problem on day one, or does it just gate access while you wait for them to convert?
Ready to try Invoicr by MRVL?
One tap to download. No sign-up wall.