mvp

The hard part of an MVP is not building it: it is deciding what to leave out. Almost everyone knows what they want in their product. Very few know the minimum they need to learn something real about their market.

An MVP is not a cheap, stripped-down version of the dream product. It is the smallest experiment that answers your biggest uncertainty. Framed that way, three weeks is enough to put something in real users' hands.

The question that defines the scope

Before you list features, answer this: which assumption, if false, makes the product pointless? That is your critical hypothesis. Anything that helps test it makes the cut; everything else waits.

If you are building a marketplace, the hypothesis is rarely "can I build a cart?" but "are there people willing to list and to buy?". The MVP should attack that, not the perfection of the checkout.

What fits in three weeks

• A single end-to-end core flow. The user arrives, performs the central action, and gets value. No detours.
• Simple authentication. Email and password, or a social provider. No complex roles yet.
• The essential data. Only the entities the core flow needs to work.
• A way to measure. If you cannot see what users do, the MVP teaches you nothing.

What stays out (on purpose)

• Elaborate admin panels: manage things by hand at first.
• Settings, preferences and personalisation.
• Performance optimisations for a scale you do not have yet.
• Rare edge cases that affect 1% of users.
• "Just in case" integrations nobody has asked for yet.

Every feature you add "just in case" is a week you delay the learning you actually need.

The three-week timeline

• Week 1 — Scope and skeleton. Define the hypothesis, the single flow and the minimal data model. Stand up the navigable base.
• Week 2 — The real flow. Build the core action end to end, with real data and basic analytics wired in.
• Week 3 — Polish and controlled launch. Fix what breaks the core flow, not the peripheral stuff, and open up to the first users.

After the MVP

The goal of these three weeks is not a finished product, but data. Do people complete the flow? Do they come back? Was your assumption true? With those answers, the next sprint decides itself, and you build on evidence instead of opinions.

A well-scoped MVP is not the small product. It is the shortest path to knowing whether the big product deserves to exist.