Minimum Viable Product (MVP)
A minimum viable product (MVP) is the smallest version of a new product that lets a team test its riskiest assumptions with real customers. Eric Ries defined it as the version that allows a team to collect the maximum amount of validated learning about customers with the least effort. Its purpose is learning, not shipping a cut-down first release.
How a Minimum Viable Product Works
Frank Robinson of the consulting firm SyncDev coined the term in 2001, describing a product sized to be big enough to drive adoption and sales without becoming bloated and risky. Steve Blank and Eric Ries later made it central to the Lean Startup movement, where the MVP is the first turn of a build, measure, learn loop.
An MVP starts from a hypothesis, not a feature list. The team names the assumption that would sink the idea if it were wrong, decides what customer behavior would confirm or reject it, and builds only what is needed to observe that behavior. Depending on the risk, an MVP can take very different forms:
- Landing page or smoke test: describes the product and measures sign-ups or pre-orders before it exists.
- Concierge MVP: the team delivers the service by hand to a few customers to learn what they value.
- Wizard of Oz MVP: customers use what looks like a finished product while people do the work behind the scenes.
- Single-feature product: one working capability that solves the core problem, with everything else left out.
"Viable" matters as much as "minimum." The MVP has to deliver enough value that real customers use it, or pay for it, in a real setting. Otherwise the result says more about the poor quality than about demand.
Why a Minimum Viable Product Matters
Most product risk sits in assumptions about customers: whether the problem is painful enough, whether they will switch and whether they will pay. An MVP tests those assumptions before a team commits months of engineering and runway. It also changes the conversation inside a team. Instead of debating opinions about what users want, the team agrees on a test and lets the results decide whether to persevere, change course or stop.
Minimum Viable Product Example
A founder believes independent landlords will pay for a tool that collects rent and chases late payments automatically. The riskiest assumption is willingness to pay, not technical feasibility. Rather than building the full platform, she launches a simple page offering the service for a monthly fee, then runs reminders for the first ten landlords with a spreadsheet and a messaging tool. Success is defined upfront: at least six of the ten pay for a second month. Seven do, and two ask for tenant screening, which goes on the list of next assumptions to test.
For keeping early scope this tight, see how pre-seed founders cut speculative features to protect runway.
Minimum Viable Product vs. Prototype
A prototype is a model used to explore or test a design or technical idea, often with test users in a controlled setting. An MVP is used by real customers in real conditions to measure real behavior, such as signing up, returning or paying. A prototype can tell you whether people understand a design. An MVP tells you whether they want the product enough to act. A successful series of MVPs is one of the main routes toward product-market fit.