What Is a Minimum Viable Product (MVP)? Definition, Types & Examples

TL;DR

- A minimum viable product (MVP) is the simplest version of a product you can release to real users to test a core idea with the least effort.

- MVP stands for Minimum Viable Product. The term was coined by Frank Robinson in 2001 and made famous by Eric Ries in *The Lean Startup* (2011).

- The point is validated learning, not a cheap product. You build just enough to find out whether people actually want what you are making, before you spend big.

- An MVP is not a broken or half-finished product. It is a complete, working slice that does one thing well and delivers real value.

- Dropbox, Airbnb, Zappos, and Uber all started as MVPs. Some were not even software at first. Dropbox's first MVP was a three-minute video.

What is a minimum viable product (MVP)?

A minimum viable product (MVP) is the simplest version of a product that can be released to real users to test a core hypothesis with the least possible effort. It has just enough features to deliver value and gather validated learning, so a team can decide whether to continue, change, or drop an idea before investing in a full build. The term was coined by Frank Robinson in 2001 and popularised by Eric Ries in *The Lean Startup*.

What is a minimum viable product?

A minimum viable product is the smallest version of your product that still works and still delivers value. You give it to real users, watch what they do, and learn whether your idea holds up. That is the whole job of an MVP: to turn a guess into evidence as cheaply and quickly as possible. For how founders use that instrument as a playbook, see MVP development for startups.

The keyword is *viable*. An MVP is minimal, but it is not broken. It is a real, functioning product that solves one problem for one type of user. If it does not work, or does not deliver value, it is not an MVP. It is just an unfinished product.

Eric Ries framed it as the version of a product that lets a team collect the most validated learning about customers for the least effort. The emphasis is on learning, not on shipping features. You are running an experiment, and the MVP is the experiment. Atlassian and the Agile Alliance both define it the same way: enough to learn, not a half-built mess.

What does MVP stand for, and who coined it?

MVP stands for Minimum Viable Product. The term was coined in 2001 by Frank Robinson, co-founder of the product-strategy firm SyncDev, as part of a method for finding the sweet spot between what a company can build and what a customer will pay for.

It reached the wider world through two people. Steve Blank built the idea into his customer development method, and his student Eric Ries put it at the centre of *The Lean Startup* in 2011. Ries paired the MVP with the build-measure-learn loop: build the smallest thing, measure how real users respond, learn from the data, and repeat. That loop is why the MVP became the default way modern startups test ideas.

Why build an MVP?

Because building the full product first is the most expensive way to find out you were wrong. An MVP flips the order: learn first, build big later, only if the learning says so.

• Validate demand before you spend. Find out if people want the product before you pour months and a large budget into it.
• Reach the market faster. A smaller build ships sooner, so you start learning and, sometimes, earning earlier.
• Cut wasted cost. You do not pay to build features nobody uses. Most of a bloated first product is exactly that.
• Get real user feedback. Opinions from a survey are weak. Behaviour from real users of a real product is strong.
• Attract investors with proof. Traction from a live MVP is far more convincing than a pitch deck describing a product that does not exist yet.

The failure statistics are the reason this matters. The most common reason startups fail is building something the market does not want. An MVP is the cheapest insurance against that specific mistake.

What an MVP is not

The word *minimum* trips people up, so here is the line. An MVP is not a low-quality product, a beta full of bugs, or version one with half the features missing and broken. Those are just bad products.

The classic illustration is transport. If the goal is to help someone travel, a minimum viable product is a skateboard, then a bike, then a motorbike, then a car. Each one actually moves the person. What an MVP is *not* is a single wheel, then two wheels, then a chassis. Those are parts. Nobody can use them, and nobody can give you feedback on them. An MVP is always something a real person can actually use, even if it is basic.

MVP vs prototype vs proof of concept vs MMP vs MLP

These terms get mixed up constantly. They describe different things built at different stages for different reasons.

The short version: a proof of concept asks "can we build it?" A prototype asks "what will it look like?" An MVP asks "do people want it?" Only the MVP is a live product in real users' hands. That is what makes it the tool for validating a business, not just an idea. For the full comparison, costs, and which one you need, see MVP vs prototype vs proof of concept.

The types of MVP

An MVP does not have to be code. Some of the most famous ones were barely software at all. The type you choose depends on what you need to learn.

Landing page MVP

A single page that describes the product and asks visitors to sign up or pre-order. It measures interest before anything is built. Buffer started this way, testing demand and pricing with nothing but a page.

Explainer video MVP

A short video showing how the product would work. Dropbox is the textbook case: founder Drew Houston posted a three-minute demo video in 2008, and the beta waiting list jumped from 5,000 to 75,000 people overnight, before the product was fully built.

Concierge MVP

You deliver the service manually, by hand, to early users. There is no automation behind the scenes, just people. It proves the value before you spend on building the technology to scale it.

Wizard of Oz MVP

The product looks fully automated to the user, but humans are doing the work behind the curtain. Zappos began like this: founder Nick Swinmurn photographed shoes in local stores and only bought them when someone ordered, testing whether people would buy shoes online at all.

Single-feature MVP

You build only the one core feature and nothing else. This is the most common software MVP, and it is exactly the discipline a fixed-scope build enforces.

Famous MVP examples

The biggest names almost all started small and unglamorous.

• Airbnb. In 2007 the founders rented air mattresses in their own apartment and built a basic site to test whether strangers would pay to stay. That was the MVP for a company now worth tens of billions.
• Dropbox. A three-minute explainer video, not a finished product, validated the demand and filled the waiting list before the heavy engineering began.
• Zappos. Photos of shoes from local stores tested online demand with no inventory and no warehouse.
• Uber. The first version, UberCab, only connected riders with black cars in San Francisco. One city, one feature.
• Instagram. It launched as photo filters and sharing, stripped down from a cluttered check-in app called Burbn once the founders saw which feature people actually loved.

The pattern is the same every time. Prove the one core thing, with the least you can build, then expand. For the mobile-specific version of this, see MVP app development.

How do you build and test an MVP?

The process is short to describe and hard to do well. Define the problem and your target user. Prioritise a tiny set of must-have features and cut everything else. Build that slice. Put it in front of real users. Measure what they actually do, not what they say. Then decide: keep going, change direction, or stop.

That decision, made on evidence instead of opinion, is the entire return on building an MVP. For the full step-by-step method, read our companion guide on how to build an MVP. If you want to know what that build tends to cost in 2026, see MVP development cost.

Where the 21-day model fits

The hardest part of an MVP is not agreeing what it is. It is holding the line on *minimum* once the build starts, because every stakeholder wants to add one more feature. A fixed scope and a fixed timeline are how you protect the discipline the concept demands.

That is the thinking behind 21-Day MVP Development: one core workflow, a capped three-week timeline, and a real, working product at the end. It turns the theory on this page into a shipped thing. For the exact calendar, see the 21-day MVP development process. We used it to launch Spellbook, an AI legal assistant, as a working MVP. See how we work for the full method, or the MVP development company guide if you are weighing up who should build yours. For the contract side of locking scope, see fixed-price vs time and materials.

Parallel Loop pricing (USD): 21-Day MVP from $5,000. MVP plus AI feature from $10,000. Free scoping call to size your idea.

Have an idea and want to know what its MVP would actually look like? Book a free scoping call and we will help you find the minimum, and the viable.

Frequently Asked Questions

What is a minimum viable product in simple terms?

A minimum viable product is the simplest working version of a product that you can release to real users to test whether your idea is worth pursuing. It has only the core features needed to deliver value and gather feedback. Think of it as the smallest real thing you can build to learn whether people actually want what you are making, before you spend the time and money to build the full product.

What does MVP stand for?

MVP stands for Minimum Viable Product. Minimum means the smallest amount of work and features possible. Viable means it still works and still delivers real value to a user. Product means it is a real, usable thing, not a mock-up or a plan. Put together, an MVP is the least you can build that a real person can actually use and give you honest feedback on.

Who invented the term MVP?

The term minimum viable product was coined in 2001 by Frank Robinson, co-founder of SyncDev. It was later popularised by Steve Blank through his customer development approach and by Eric Ries, who made it a central idea in his 2011 book The Lean Startup. Ries paired the MVP with the build-measure-learn loop, which is why the concept spread so widely through the startup world.

What is the difference between an MVP and a prototype?

A prototype is a model that shows how a product will look and feel, and it is often not fully functional. It is used internally to test design and gather early feedback. An MVP is a real, working product released to actual users to test whether there is genuine demand. The key difference is that a prototype answers 'what will it look like?' while an MVP answers 'do people actually want this?'

What is the difference between an MVP and a proof of concept?

A proof of concept (PoC) is a small internal test that proves something is technically possible to build. It usually never leaves the engineering team. An MVP is a complete, working product, however basic, that goes into the hands of real users to validate market demand. A PoC checks feasibility; an MVP checks desirability. Many products go through a PoC first, then an MVP.

Is an MVP a finished product?

No. An MVP is deliberately incomplete. It contains only the core features needed to test a specific idea, and it is meant to grow based on what you learn from real users. That said, an MVP must be viable, meaning the features it does have work properly and deliver genuine value. It is minimal but not broken. The full product comes later, shaped by MVP feedback.

What are the main types of MVP?

Common types include the landing page MVP (a page that tests interest), the explainer video MVP (a demo video, like Dropbox used), the concierge MVP (the service delivered manually by hand), the Wizard of Oz MVP (looks automated but humans do the work, like early Zappos), and the single-feature MVP (only the one core feature is built). The right type depends on what you most need to learn.

What is a famous example of an MVP?

Dropbox is the classic example. Before building the full product, founder Drew Houston released a three-minute video in 2008 showing how it would work, and the beta waiting list grew from 5,000 to 75,000 people overnight. Other famous MVPs include Airbnb, which started with the founders renting air mattresses in their apartment, and Zappos, which tested demand by photographing shoes in local stores.

How much does it cost to build an MVP?

It varies widely with complexity. A simple MVP can cost roughly $8,000 to $25,000, a standard SaaS MVP around $30,000 to $60,000, and a complex or AI-heavy MVP $60,000 to $150,000 or more in 2026. The main cost drivers are feature count, platforms, integrations, and who builds it. Our MVP development cost guide breaks these figures down variable by variable. Parallel Loop 21-Day MVPs start from $5,000.

What comes after the MVP?

After the MVP, you use what you learned to decide your next move: continue and add features, pivot and change direction, or stop if the idea did not validate. If it did validate, the product usually grows toward a minimum marketable product (MMP), the version polished enough to market and sell broadly. See our guide on what to do after your MVP launches for the full 90-day playbook.