MVP - Minimum Viable Product: What It Really Is, and Why It's So Easy to Get the Definition Wrong
MVP - Minimum Viable Product: What It Really Is, and Why It’s So Easy to Get the Definition Wrong
An MVP is not a “first version.” It’s not “something temporary.” And it’s certainly not an excuse for low quality.
An MVP is a minimal product that can survive in reality.
The Precise Definition
An MVP is the smallest implementation of a solution that reliably allows a business or product assumption to be tested.
Three necessary components:
- Minimum - minimum capabilities, not minimum thought
- Viable - works, is stable, and is used by real users
- Product - something that delivers value, not just a technical demo
If even one of the three is missing - it’s not an MVP.
What Makes a Product Viable?
Viable doesn’t mean “bug-free.” But it does mean:
- Predictable behavior
- Failures that don’t cause total collapse
- A usage experience people can actually live with
- The ability to measure usage, success, and failure
An MVP that can’t be trusted doesn’t generate learning. And learning is the entire point.
An MVP Is an Experiment, Not a Prototype
A prototype exists to show an idea. An MVP exists to prove or disprove an assumption.
That’s why an MVP:
- Must meet real users
- Must run under real conditions
- Must produce real data
If there’s no decision that’s supposed to be made based on it - there’s no reason to build an MVP.
What’s the Smallest Thing That Lets Us Learn the Most Important Thing?
A good MVP starts with this question.
Not:
- What’s easy to build
- What we already have
- What we can “improve later” without thinking
But rather:
- What’s the most critical assumption right now
- Which solution tests that truth in the cleanest way
An MVP doesn’t require a perfect architecture, but it does require:
- Clear boundaries
- Explicit assumptions
- A structure that can change
It is minimal in scope, not minimal in responsibility.
The Question That Defines Whether It’s Really an MVP
Before building an MVP, there must be a clear answer to two points:
- What signal will count as success, and what assumption does it validate
- What failure will count as useful information, and what decision should it change
If there are no clear criteria for success and failure - no real learning can be produced.
And without learning that leads to a decision, it’s not an MVP. It’s just a start with no direction.