MVP guide

What is an MVP — and how do you define the right one?

An MVP (Minimum Viable Product) is the smallest version of a product that delivers enough value to attract early customers and validate your core assumptions. It is not a half-baked product — it is a fully functional product with a narrow, specific scope.

What an MVP is NOT

The word "minimum" is widely misunderstood. Here is what an MVP is not.

It is not a prototype or demo

A prototype explores a concept. An MVP delivers real value to real users.

It is not a product with all the bugs removed

Removing bugs is QA, not scope definition. An MVP is about what you build, not how polished it is.

It is not 50% of all the features

Cutting features by half is not a strategy. The 'minimum' in MVP means minimum to test your most critical hypothesis — not minimum to be acceptable.

The MVP decision framework

Four steps — in order. Do not skip to step 3 before you can answer step 1.

1
Write your riskiest assumption
'We believe [user type] will pay for [solution] because [reason].' One sentence. If you cannot write it in one sentence, you do not know what you are testing yet.
2
What is the smallest experiment that tests this assumption?
Not the smallest product — the smallest experiment. Sometimes that is a landing page. Sometimes it is a manual process. Sometimes it is a video. Build the experiment, not the product.
3
What must be in the product to run this experiment?
That is your MVP scope. Everything else is a should-have, could-have, or will-not-have. Write the scope boundary down explicitly.
4
Define success before you build
What would success look like? A specific number, a specific behavior. If you wait until after you ship to define success, you will rationalize whatever you see.

Famous MVP examples

Three companies that validated billion-dollar ideas with almost no product at all.

Dropbox

Instead of building the product first, Drew Houston made a 3-minute demo video explaining what Dropbox would do. 75,000 sign-ups overnight validated the idea before a line of code was written.

Zappos

Instead of building inventory and logistics, Nick Swinmurn took photos of shoes at local stores and posted them online. When orders came in, he bought the shoes at retail and shipped them. Validated demand without a warehouse.

Airbnb

Founders rented air mattresses in their own apartment. Validated that strangers would pay to sleep in someone else's home before building the platform.

MVP scoping in practice: the MoSCoW method

A structured way to force every feature into a category — and make the trade-offs explicit.

Must have

Without this, the product does not work at all.

Should have

Important but the MVP works without it.

Could have

Nice-to-have for later.

Will not have (this time)

Explicitly out of scope.

Common MVP mistakes

Four ways teams get MVP wrong — and how to avoid them.

Goldilocks problem

Too big (scope creep) or too small (cannot test anything meaningful). The MVP must be large enough to actually run the experiment — no more, no less.

Building for edge cases in v1

99% of users will never hit them. Edge cases can wait. Scope them out explicitly so the team is not tempted to add them.

Skipping the learning

MVP is only valuable if you define metrics and actually measure them. A ship without a success metric teaches you nothing.

MVP as an excuse for poor quality

Poor quality gives you wrong signals. If users churn because the product is buggy, you cannot tell whether the idea failed or the execution did.

Next steps

Learn PM fundamentals in the Product Manager track

MVP definition is one skill in a full PM toolkit. The product manager track covers discovery, prioritization, roadmapping, stakeholder communication, and more.

Explore the PM track