What Eight Weeks Actually Buys You in an MVP

The question behind the question

When someone asks whether we can build their product in eight weeks, they are rarely asking about our typing speed. They are asking whether they can put something in front of a customer before their runway, their board meeting, or their competitor forces the decision for them.

So we answer the real question. Eight weeks buys one workflow, done properly, for one kind of user.

What fits

• One path through the product. Sign up, do the thing the product exists to do, see the result.
• Real data. A schema you will not have to throw away, with the constraints written down on day one.
• Production hosting. Deployed, monitored and backed up from the first week, not the last.
• A way to watch it. Analytics on the three events that tell you whether the workflow works.

What does not

Admin panels, role hierarchies, billing tiers, notification preferences, an onboarding tour, a second platform. None of these are hard. All of them are work, and none of them teach you anything you do not already know.

The fastest MVPs we have shipped were not the ones with the shortest feature lists. They were the ones where the founder could name, in one sentence, the thing they needed to find out.

How we split the scope

We run a two-day shaping session before anyone writes code. Every requested feature lands in one of three buckets: needed to learn, needed to launch, and needed eventually. Only the first bucket is in scope. The second gets a plan, the third gets a note.

That conversation is uncomfortable for about an hour and saves about three weeks.

The part nobody budgets for

Reviews, environments and the last ten percent. Assume two of your eight weeks go to things that are not features: access to a payment sandbox, a domain stuck with a registrar, a legal review of the terms page.

Budget for it and you ship on time. Ignore it and you ship in eleven weeks, which is what everyone else calls eight.