MVP in 8 Weeks: The Scope-Cutting Framework That Actually Ships

How do I build an MVP in 8 weeks?

You build an MVP in 8 weeks by cutting features, not by hiring more engineers. The teams that ship fast and the teams that ship late are not separated by skill — they're separated by scope discipline. An 8-week MVP has 8 features, one core user flow, and a clear hypothesis. A failed MVP has 40 features, three half-built flows, and a vague aspiration to "build a platform". The four-stage framework — Validate, Core, Cut, Ship — exists to keep the scope honest. We've shipped 50+ MVPs using it. Every single one that launched in 8 weeks followed the framework. The ones that drifted to 16 weeks broke a stage. Usually Stage 3.

Why most MVPs ship late (or never)

The pattern is always the same. A founder starts with a clear hypothesis. By week 3, an advisor suggests "you should also build X". By week 5, a beta tester says "if only it had Y". By week 8, the team is building admin dashboards and onboarding flows for users who don't exist yet. The original hypothesis is buried under feature creep. The MVP didn't fail to ship — it failed to stay small. Scope discipline is uncomfortable because every cut feels like saying no to a future user. But every feature you ship is a feature you have to maintain, support, and migrate when you pivot. The discipline isn't "say no to everything" — it's "say no to everything that doesn't help test the hypothesis."

The 4 stages — validate, core, cut, ship

Stage 1: Validate. Before code, run 8–15 customer interviews. Look for whether the problem you imagined is the problem people actually have. If it isn't, change the hypothesis. We've covered this discovery sprint approach in detail in our product strategy consulting work.

Stage 2: Core. Identify the one critical user flow. Sign up → do the thing → see the value → convert. Everything else is a branch off this flow; if it's not on the main path, it's a Stage 3 cut.

Stage 3: Cut. Take your 100 ideas, group them into 24 themes, prioritize to 12 features, ruthlessly cut to 8. The cuts are not negotiable. We've had founders push back here every time — and every time we've shipped on schedule, the cuts held. The features that survived were the ones that mattered.

“”

Stage 4: Ship. Four sprints. Sprint 1: discovery + wireframes. Sprint 2: build core flow. Sprint 3: auth + billing + polish. Sprint 4: QA, deploy, launch. The schedule is fixed; the scope flexes only by cutting (never by adding).

What does an 8-week MVP cost?

A typical 8-week MVP runs $35k–$60k. That's the real range for what we ship at MVP development — not a salesy "starts at $15k" number. The variance comes from four things: whether you need a dedicated designer in the sprint, your stack choices (familiar stacks ship faster), whether you have compliance requirements (GDPR, HIPAA, SOC2 add weeks), and how many third-party integrations you need (Stripe is fast; custom auth flows or legacy API integrations are slow). What doesn't move the cost much: how big your eventual market is, how impressive the design needs to look at launch (we use systems that look good by default), or whether you have a tech background. We've shipped MVPs for non-technical solo founders and for ex-Google engineers; the cost is the same.