🪐 MVP = smallest test of a real hypothesis, not 'v1 but worse'

PM MVP Guide
(2026 Edition)

5 things MVPs actually are, 5 things they aren't, 5 scoping questions, and 5 situations where skipping the MVP is the right call.

Build PM Scoping Skills Daily — Free →

What an MVP Actually Is

Smallest version that tests your core hypothesis with real users

Functional end-to-end — users can complete the core job, not a screenshot

Learning-oriented — primary output is signal, not scale

Ugly is fine, broken is not — polish can wait, correctness can't

Scoped ruthlessly — 1 core flow, not 5 'starter' features

What It Isn't

A prototype — prototypes don't ship to real users

A beta — betas are fuller versions; MVPs are thinner

A launch — MVPs test; launches scale

'Everything we eventually want, but worse' — that's underbuilt, not minimal

Meant to be scalable — expect to rebuild post-learning

5 Questions for Scoping an MVP

1.

What's the ONE hypothesis we're testing?

2.

What's the minimum user flow that tests it end-to-end?

3.

What can we fake (Wizard-of-Oz, manual backend, mocks) for now?

4.

What would we learn in 4 weeks that we can't learn from a 12-week MVP?

5.

What MUST we get right for the signal to be valid?

5 Situations to Skip the MVP

1.

Well-understood problem space with clear solution — just build

2.

Existing product adding a well-understood feature — not an experiment

3.

High-stakes launch where 'broken MVP' damages brand more than delay

4.

Regulatory/compliance-heavy products — can't ship partial

5.

Platforms or infrastructure — usually iterative, not MVP-oriented

FAQ

What's the biggest MVP mistake PMs make?

Including too much. 'Minimum' gets compromised to avoid user complaints, edge cases, or stakeholder asks. The discipline: if your MVP takes more than 6 weeks to build, it's probably not an MVP — it's a v1 with optimistic labeling. Cut scope ruthlessly; you'll be surprised what users tolerate for genuine new value.

How do PMs decide if an MVP was successful?

Pre-commit to learning criteria before building. 'If 40% of users complete the flow and 25% come back within 7 days, we'll invest further.' Without pre-committed criteria, you'll rationalise any outcome into 'validation.' PMs who pre-commit make sharper kill/invest decisions and learn faster.

Build PM Scoping Muscle Daily

Daily scenarios on MVP scoping, hypothesis-testing, and minimum-viable-learning.

Start Free Trial →