Bootstrapped SaaS to $1M ARR in fintech embedded payments

The hardest moment was a regulatory scare in October 2024. A state-level money-transmitter regulator in one of the southern states sent us a letter alleging that our payment-spread model required a money-transmitter license. We were not a money transmitter — Stripe Connect was the underlying processor and they held the licenses — but the letter was real, the language was scary, and our outside counsel quoted us $48,000 to draft a formal response.

We didn't have $48,000. We had about $19,000 in the bank.

§Reading the statute ourselves

I spent two days reading the actual statute and three nights drafting a response myself. Anika reviewed it. We sent it. The regulator wrote back six weeks later acknowledging that our model fell under the existing exemption for payment facilitators with a sponsor processor relationship. No further action.

If I'd had the $48,000 I would have happily paid it. The fact that I didn't forced me to learn the actual rules. I now have a stronger working understanding of payment-facilitator regulation than most people I worked with at my previous fintech.

The second-hardest moment was a near-divorce of cofounders. Anika and I had a four-day stretch in February 2025 where we couldn't agree on whether to add invoicing — a feature 30% of customers were asking for that I felt would distract us from the core loop. She felt I was being controlling. We worked it out by writing a one-page document together listing the exact next four features we'd ship and the exact criteria for adding a fifth.

The unlock came from a single integration: QuickBooks. In May 2025 we shipped a one-click QuickBooks sync that pushed every transaction, including the payment spread, into the customer's accounting software with the right categorization. Most of our customers had been hand-entering transactions monthly because their accountant required QuickBooks data.

The sync took us four engineering weeks to build. It was the single feature that turned us from "a nice-to-have" into "the system of record." Churn dropped from 4.1% monthly to 1.2% monthly within ninety days of shipping it. Average revenue per customer ticked up 18% because operators were now using the app for every transaction instead of just card-present payments.

As of April 2026 we have 1,140 customers, gross margin around 78% (after Stripe interchange), and net revenue retention of 118%. We've never raised outside money. We've turned down two acquisition conversations.

§Crossing $1M ARR with eleven people

The six months after we shipped the Stripe Connect Custom migration tool were when growth went non-linear. Embedded-payments customers — vertical SaaS companies who'd built on Stripe Standard and outgrown it — could move to Connect Custom with us in three days instead of the typical six-to-eight weeks. Net revenue retention hit 156%. We crossed $1M ARR in February 2026, eleven months later than I'd planned and at a contribution margin (87%) that bought us our pick of how to grow.

We made our first revenue-side hire that quarter: Ines, a payments product manager from Adyen, who owned the Connect Custom migration product end-to-end. Her existence let me get back to writing code two days a week, which had been the thing I'd missed most through the first eighteen months of selling. Average implementation time fell from forty-one days to twelve. Trial-to-paid lift went from 14% to 39%. The vertical-SaaS bootstrapping motion overlaps closely with the legal-tech Casefile story in this cohort — different vertical, same mechanic of choosing a niche the platforms above you cannot serve well.

1. 1
   The wedge is the vertical, not the payments.

   Embedded-payments revenue follows the vertical SaaS, not the other way around. Win the operating system; the payments come along for the ride.

2. 2
   Boring categories pay best.

   Mobile dog grooming is not glamorous. It also has fourteen thousand operators, no real competition, and customers who pay invoices on time.

3. 3
   Read the statute yourself.

   A $48,000 legal quote can hide knowledge you actually need to operate the business. You will hire lawyers eventually. Be the kind of founder who knows what they're paying for.

4. 4
   The unsexy integration is the real moat.

   QuickBooks sync changed our churn rate more than any new feature. Listen to "I have to enter this somewhere else" complaints.

5. 5
   Cofounder fights need paper.

   A one-page written agreement on the next four features cost us less than one more day of arguing.

What this looked like in practice.

The thing nobody tells you about *embedded payments* is that the platform fees are the smaller revenue line. The bigger one is the float on the merchant ledger, and you only get to that line if you have the right banking partner. We spent four months on partner due diligence with Cross River Bank and Evolve Bank & Trust before picking one. That choice has been more load-bearing than any product decision.

The thing nobody tells you about *Stripe* is that they are the best partner and the best competitor simultaneously. Our entire wedge exists because Stripe Connect Custom is too complex for most vertical SaaS teams to implement well. The day Stripe ships a managed migration tool is the day we lose 60% of our pipeline. We have eighteen months, maybe twenty-four, to entrench.

The thing nobody tells you about *bootstrapping in fintech* is that the regulators talk to each other. One FinCEN filing screwup costs you years. We hired a fractional General Counsel — Aliyah, a former OCC examiner — at $2k of revenue per month, before we even had product-market fit. Cheapest insurance we have ever bought.

The thing nobody tells you about *87% contribution margins* is that they let you say no to bad customers. We have turned away three deals worth more than $300k each because the merchant risk profile would have raised our chargeback ratio above the Visa TC40 threshold.

The thing nobody tells you about *PMs from Adyen* is that they have already solved the problem you are about to encounter, twice. Hiring Ines compressed our 2026 roadmap by eight months.