How to Validate an MVP Before You Build It

To validate an MVP, test the riskiest assumption with real people before you build anything: do they have the problem, do they care enough to act, and will they pay or commit. You can learn this with conversations, a simple page and a manual version of the service, often within a couple of weeks.
It is Step 1 of our MVP development guide.
Start with your riskiest assumption
List what must be true for the business to work. Test the assumption that is both most important and least proven.
Seven ways to validate
- Customer interviews. Talk to 10 to 15 people who have the problem. Ask about past behaviour, not hypothetical interest.
- A landing page. Describe the product and measure how many people sign up or join a waitlist.
- Fake door test. Offer a feature that does not exist yet and see who clicks.
- Concierge MVP. Deliver the service by hand to a few customers.
- Prototype tests. Show a clickable prototype and watch people use it.
- Pre-sales or deposits. The strongest signal is someone paying or committing.
- Competitor and search research. Check what people already use and search for.
How to read the results
- Compliments are not demand. Action is.
- Count commitments: sign-ups with a real email, bookings, deposits.
- Pay attention to why people say no.
Set success criteria before you test
Write down what result would make you proceed, change direction or stop. It stops you from interpreting every result as a win.
Common mistakes
- Asking friends, who will be kind.
- Building first and testing after.
- Testing too many ideas at once.
Using AI to speed it up
AI can help draft a landing page, interview scripts and a clickable prototype quickly. It cannot tell you whether people want the product. See vibe coding in production: what breaks for why a prototype is not the product.
When you are ready to build
If the results are positive, define the smallest version that delivers the core value. Our web app and MVP development service turns validated ideas into a first release. Share your findings.
Frequently asked questions
How many interviews are enough? Patterns often appear after 10 to 15 conversations, though it depends on the market.
Should I charge before building? If you can, yes. It is the clearest evidence.
What if the test fails? Good. You learned cheaply. Adjust the idea or move on.