What is an MVP?
MVP stands for minimum viable product: the smallest thing you can put in front of real people to learn one specific thing. Minimum means you cut everything that doesn’t help answer the question. Viable means it still answers it. An MVP isn’t a worse version of your product. It’s an experiment that happens to look like one.
One question per MVP
Before you build anything, name the one thing you need to learn. Will students pre-order lunch at all? Can people figure out the checkout without help? Can the sensor read through a sock? Each of those is a different MVP, and trying to answer all of them with one build is how an MVP turns into a six-month project.
Four kinds of fake
Paper, for usability.
Sketch the screens and watch someone try to use them. You learn where they get stuck before anyone writes code.
An engineer’s prototype, for feasibility.
Rough code that proves the hard part can be done. Ugly is fine. Nobody outside the team sees it.
Real but rough, for demand.
Something people can sign up for, pay for or use, built just well enough to count who shows up.
A person behind the curtain, for a service.
The customer sees a finished service. Behind it, someone does the work by hand. It doesn’t scale, and it isn’t supposed to.
The crew listed twelve features for the lunch app. The question they need answered first: will students pre-order lunch at all? Cut the list down to what it takes to learn that, and nothing else.
This is the same exercise that’s in the app.
The one question
Will students pre-order lunch at all?
Decide what would kill the idea, before you run it
Write down the result that would make you stop. “If fewer than 30 students sign up in a week, we don’t build the app.” If you pick that number after you see the results, you’ll pick whatever keeps the idea alive.
Cut isn’t killed
Everything you leave out of an MVP can still be a good idea. It just isn’t the question right now. Keep the list, answer the first question, and let the answer tell you what comes next.
A real one: a video of a product that didn’t exist
Before Owlet had a working baby monitor, the founders made a video for a class project showing what it could do. They called it “smoke and mirrors.” It leaked, got picked up by news outlets in 13 countries, and brought in more than 500 emails from parents asking to buy one. It cost them about $220 and two weeks, and it told them what a survey couldn’t: people wanted it badly enough to chase it.
Questions people ask
What’s the difference between an MVP and a prototype?
A prototype is any rough version you build to learn something. An MVP is a prototype aimed at one question, usually whether people want it, and put in front of real people rather than your team. Every MVP is a prototype. Not every prototype is an MVP.
Is an MVP the same as product-market fit?
No. An MVP tests one question, often in days or weeks. Product-market fit is what it looks like when the whole product works for a market: people keep using it, tell others and pay for it without you pushing. MVPs are how you get there.
Learn it by doing it
The app teaches this in Module 14, Fake It Before You Make It.
Product Management is a course for curious teens, in five-minute lessons with something to play with in every one. Part I is free on iPhone, iPad and Android. No account, no ads.
Keep going
What is a value proposition?
Why someone would pick you over what they do today, in one sentence.
What is a PRD?
What you’re building, what you’re not, and how you’ll know. On one page.
The baby sock that found its market, then ran into the FDA
A baby monitor that proved its demand for $220, then spent five years at odds with the FDA.