Definition, with the versions that hold up in practice
What a minimum viable product is, and the four kinds that actually work
A minimum viable product, or MVP, is the smallest version of a product that lets you test the riskiest assumption in your business with real users. Viable means it delivers genuine value to someone and can take their money, not that it is a rough draft of the full product. The word people get wrong is minimum: an MVP is narrow in scope, not shoddy in quality, and the part it does must actually work.
- ◆MVP stands for minimum viable product
- ◆Purpose: test the riskiest assumption, not to launch a small version of everything
- ◆Viable means it delivers value and can charge for it
- ◆The four working types: concierge, single feature, piecemeal, and pre sale
The term comes from lean product thinking and has since been stretched to mean 'whatever we shipped first'. That vagueness is expensive, because it lets teams call a broad, half finished product an MVP and then learn nothing from launching it.
Here is the definition that is actually useful, and the four shapes an MVP takes when it works.
The workbook version of this guide
This page explains the thinking. The Launch Workbook makes you do it: seven modules with fields you fill in, saved as you go.
The definition that is worth using
An MVP is the least amount of built product required to get a truthful answer to your riskiest question. Start from the question, not from the product. If your riskiest question is whether people will pay, the MVP needs payment and can fake almost everything else. If your riskiest question is whether the technology can work at all, the MVP is a technical spike with no interface.
Two words carry the weight. Minimum means narrow scope. Viable means the part you built genuinely delivers value and someone can complete a real transaction. Narrow and excellent, not broad and broken.
Type one: the concierge MVP
You deliver the outcome manually while the customer experiences a product. They pay, they get the result, and behind the interface it is you doing the work by hand.
This is the fastest and most underrated version. It tests willingness to pay with almost no build, and it teaches you the edge cases before you encode them in software. It fails only when the value depends on speed or scale you cannot fake.
Type two: the single feature MVP
You build one feature properly, to production quality, and nothing else. No settings, no roles, no dashboard. Users get one thing and it works every time.
Use this when you already know the buyer and the problem, and the open question is whether your particular solution is good enough to switch to.
Type three: the piecemeal MVP
You assemble existing tools into a working service: a form, a spreadsheet, a payment link, a scheduling tool, a model API. No custom software at all.
It looks unglamorous and it answers the demand question in days. If people pay for a duct taped version, they will pay for a real one. If they will not pay for it duct taped, building it properly will not change their mind.
Type four: the pre sale MVP
You sell it before it exists: a page, a specific promise, a price, and a real payment or a signed commitment with a delivery date.
This is the strongest signal available and the most uncomfortable to run, which is exactly why it is honest. A waitlist is curiosity. A deposit is demand.
The four ways an MVP proves nothing
Most MVPs fail as experiments rather than as products. They launch and the team cannot say what they learned.
- ·It is free, so it measures interest rather than demand
- ·It is broad, so when it does not work you cannot tell which part failed
- ·The riskiest assumption was never written down, so any outcome can be rationalised
- ·The success threshold was decided after seeing the numbers
Which type fits your riskiest question
| Your riskiest question | MVP type | Time to answer |
|---|---|---|
| Will anyone pay for this at all | Pre sale | Days |
| Is this outcome valuable enough to buy | Concierge | One to two weeks |
| Can existing tools already do this well enough | Piecemeal | Days |
| Is my solution better enough to switch to | Single feature | Two to six weeks |
| Is this technically possible | Technical spike, no interface | Days |
Common questions
What is a minimum viable product?
The smallest version of a product that lets you test your riskiest assumption with real users, while still delivering genuine value and being able to take payment.
What does MVP stand for?
Minimum viable product. In business and software it means the narrowest build that produces a truthful answer about demand.
What is the difference between an MVP and a prototype?
A prototype demonstrates how something could work and is usually not used by real customers. An MVP is used by real customers who can pay for it, which is why it produces evidence a prototype cannot.
Does an MVP have to be low quality?
No. Minimum refers to scope, not craft. The narrow thing your MVP does should work reliably, because an unreliable MVP tests your bugs rather than your idea.
Should an MVP be free?
Usually not. A free MVP measures curiosity. Charging, even a small amount, is the only way to separate people who like the idea from people who have the problem.
You can read this, or you can do it with someone who has done it three times.
Zero to Entrepreneur is an eight week live cohort with 12 seats. You finish with a product live, real customers, and the numbers to decide what happens next.