Six weeks is enough to ship a real product, but only if the scope is honest. Most MVPs miss the date because the team quietly kept adding, not because engineering was slow.
Weeks one and two: decide what not to build
The first job is subtraction. We write the one sentence a user should be able to complete, and cut everything that does not serve it. A settings page that no one needs on day one is a week you do not have.
- One core flow, end to end
- One integration, not five
- Auth and billing only if the flow truly needs them
Weeks three and four: build the spine
We ship the core flow behind a flag and use it ourselves every day. Dogfooding surfaces the real gaps far earlier than a spec review ever will.
Quality is not the thing you add at the end. It is the corner you refuse to cut in week two.
Weeks five and six: harden and launch
The last stretch is polish, error states, and the empty screens everyone forgets. A production-grade MVP is small, but every path a user can take actually works.
