Glossary · Product and traction · Updated
An MVP, or minimum viable product, is the smallest version of a product that lets you learn something decisive from real users. It is not a degraded version of the final product: it is an experiment designed to confirm or kill a specific hypothesis as fast as possible.
The term is badly misused. An MVP is not "the product, but worse". It is a measuring instrument. The right question is not "what do we cut?" but "which hypothesis would kill the project if it were false, and what is the fastest way to test it?".
The word "viable" does all the work. The product has to be complete enough for usage to be real: if users drop out because of a bug or slowness, you measured nothing at all, you merely tested your own execution. Cut scope, never reliability of what you ship.
An MVP is also a cash decision. Every week spent building burns runway without producing learning. That is why experienced teams fix a date before they fix a scope, then cut scope to hold the date.
Example: scoping an MVP
A team wants to check whether recruiters will pay for an application-screening tool.
| Hypothesis under test | A recruiter will pay €50/month to halve screening time |
|---|---|
| In scope | CV import, automatic ranking, export of the shortlist |
| Out of scope | Multi-user accounts, billing, integrations, mobile app |
| Payment | A manual bank transfer, no billing module |
| Build time | 6 weeks |
| Users recruited | 20 recruiters contacted by hand |
| Validation signal | 10 of 20 use it 3 weeks running and 5 agree to pay |
| Cost | 1 developer, €0 marketing budget |
Handling payment by manual transfer is deliberate: a billing module would have cost two weeks to test a hypothesis nobody doubts, namely that people can pay online.
The common mistake
Confusing an MVP with a prototype. A prototype shows an idea; an MVP measures a behaviour. If nobody actually uses it, in their real work, with their own data, you built a demo and learned only what people say, not what they do.
Frequently asked questions
How long should an MVP take to build?+
Ask it the other way round: fix a duration first, usually four to eight weeks, then cut scope to hold it. An MVP that takes six months is no longer an MVP, it is a first release, and you burned runway before learning anything.
Should an MVP charge money?+
If your hypothesis is about willingness to pay, yes: it is the only test that counts, and a verbal yes proves nothing. If the hypothesis is about usage or retention, free is fine, as long as you do not then conclude the product is sellable.
Related terms
Prepare your raise with the right documents
Onefive brings the dataroom, the investor network and the startup profile into one workspace, built for European teams from pre-seed to Series A.
