MVP vs prototype vs proof of concept

The short answer

A proof of concept tests whether something can be built. A prototype tests how it should look and work, usually with a few people. An MVP tests whether real customers want it enough to use it or pay. You only need the first two when building it or using it is the risk; for most ideas, the risk is demand.

The difference at a glance

Each one is a way to reduce one kind of risk before you spend more. The question it answers is what sets them apart.

	Proof of concept	Prototype	MVP
Question it answers	Can it be built?	How should it work?	Do people want it?
Who sees it	Your team, sometimes investors	Testers and a few likely users	Real customers
What it looks like	A technical demo, often rough	Sketches, screens or a clickable model	A working product or a real offer
What it proves	Feasibility	Usability	Demand
Typical time	Days to weeks	Days to weeks	Days to a few weeks

What a proof of concept is

A proof of concept, or POC, is a small piece of work that shows an idea is technically possible. It answers one question: can this be done at all? A POC might be a script that proves two systems can talk to each other, a model that hits the accuracy you need, or a hardware rig that shows a sensor works in the field.

Nobody outside the team is meant to use it. It's usually rough, with no design and no edge cases handled, because its only job is to remove doubt about feasibility. If nobody doubts that your idea can be built, which is true of most software ideas today, you don't need one.

What a prototype is

A prototype shows how a product might look and work, so you can test the experience before building it properly. It ranges from paper sketches to a clickable model in a design tool to a high-fidelity mockup that looks real but has nothing behind it.

Prototypes are for learning about usability: can people find their way, does the flow make sense, what confuses them? You test them by watching a handful of people use them. A prototype can't tell you whether anyone will pay, because the people testing it were invited and nothing was at stake for them.

What an MVP is

A minimum viable product, or MVP, is the smallest version of a product or offer that real customers can use or buy. Eric Ries defined it as the version that brings the most learning about customers for the least effort. It can be a basic app, a service delivered by hand, or simply a landing page with a price.

What makes it an MVP is who it's for and what it measures: real customers, deciding for themselves whether to sign up, use it or pay. That's the only one of the three that tests demand.

What is an MVP? The full guide

Which one do you need?

Ask what's most likely to sink the idea, and test that first.

If you're not sure it can be built: Build a proof of concept. Keep it small and throw it away once it has answered the question.
If you're not sure people can use it: Build a prototype and watch five people try it. Fix what confuses them before writing real code.
If you're not sure people want it: Test demand with an MVP, or before that with a landing page and a price, shown to people who have the problem.

The mistake most founders make

Most software ideas don't fail because they couldn't be built or because the buttons were confusing. They fail because not enough people wanted them. Startup post-mortem studies, such as CB Insights' analyses, regularly list “no market need” among the top reasons startups fail.

Yet founders often start with a proof of concept or a polished prototype, because building feels like progress. They spend weeks proving something nobody doubted, then find out about demand at launch. If demand is the biggest risk, test it first. It's also usually the cheapest test of the three.

Test demand before you build any of them

You can test demand without a POC, a prototype or a product. Find people describing the problem, then put a landing page with a price in front of them and count how many sign up or click a plan. If they do, you've learned the most important thing first. If they don't, you've saved yourself all three builds.

How to run a landing page test

Questions

What is the difference between a prototype and an MVP?

A prototype shows how a product might look and work, and it's tested with a few invited people to improve the design. An MVP is something real customers can use or buy, built to find out whether they want it. A prototype tests usability; an MVP tests demand.

What is the difference between a POC and an MVP?

A proof of concept shows that an idea can be built, usually for the team or investors, and nobody is meant to use it. An MVP is a small product or offer that real customers use or pay for. A POC answers “can we build it?”; an MVP answers “do people want it?”.

Which comes first, a prototype or an MVP?

Whichever tests your biggest risk. If people might struggle to use the product, a prototype comes first. If the main question is whether anyone wants it, which is true for most ideas, test demand first with a landing page or a simple MVP, and prototype once you know people care.

Is a landing page an MVP?

It can be. A landing page with a price and a sign-up form, shown to people who have the problem, is one of the simplest MVPs: it tests whether the promise alone makes people act. It doesn't test whether the finished product keeps them, which comes later.

Do I need a proof of concept for a software idea?

Usually not. Most software ideas use technology that already works, so feasibility isn't in doubt. Build a proof of concept only when there's a real technical question, such as whether a model can hit the accuracy you need, and keep it as small as the question allows.

See who's already asking

Describe your idea and see real posts from people asking for a fix, for free. If you want, Ghostlaunch runs the 7-day test for you.

Try the free idea validator