The brief
A solo founder came to us with a validated idea in the SMB accounting space, six weeks of runway, and a small list of prospects already asking to pay. The goal wasn't a polished product — it was to prove the core workflow, take money, and learn.
What we cut
No custom design system. We used a well-maintained component library and themed it in a day. No microservices — a single Next.js app with a Postgres database. No CI/CD pipeline beyond preview deployments. No feature flags, no A/B testing, no analytics dashboards. Just the shortest path to a working product.
The hardest cuts were the ones that felt like fundamentals. Onboarding was a single email with a magic link. Billing was a Stripe Checkout redirect, not an in-app flow. Admin tools were direct database queries. Each of these would need to be rebuilt for scale, and that was fine.
What we kept
Testing on the core financial calculations — because getting money math wrong once destroys trust forever. Proper database migrations from day one. Structured logging, so we could debug production issues without shipping a new build. And a clear separation between the core domain logic and the framework, so the parts that mattered could survive a rewrite.
What we would do again
Weekly demos with the founder's three warmest prospects. Their feedback shaped the roadmap more than any user research could have, and it turned prospects into champions before launch. The product shipped on week six, took its first payment on week seven, and the founder raised a seed round three months later on the strength of the traction.
