MVP testing
MVP testing is using a minimum viable product with target customers to test a business assumption, such as whether people will pay for a solution or keep using it. The results guide what to build, change or stop.
There is no single standardized "MVP test." Unlike quality assurance, which checks whether a product works, MVP testing checks whether customers value what it does.
How it works
Choose the riskiest assumption, the customers who can test it and a measurable pass/fail rule before launching. Build only enough to run that test. A service delivered manually behind the scenes can qualify; automation can wait.
Eric Ries's The Lean Startup puts validated learning, rather than feature count, at the center of this approach.
Eight customers pay, but only five renew
You test a manually operated invoice-reminder service with 20 freelance designers who send invoices monthly. The price is $25 per month.
Before the test, you set two targets: at least eight customers must pay, and at least six must renew for a second month. These are your decision rules, not industry benchmarks.
| Measure | Calculation | Result |
|---|---|---|
| Initial purchase rate | 8 buyers ÷ 20 prospects | 40% |
| First-month revenue | 8 × $25 | $200 |
| Second-month renewal rate | 5 renewals ÷ 8 buyers | 62.5% |
The payment target passes; the renewal target fails. Investigate why customers left before spending money on automation. This small test provides useful evidence, not proof of product-market fit.
Questions people ask
How to validate an MVP?
Test it with people who match your intended buyers, not just friends. Look for behavior tied to your assumption: payment, completed tasks or repeat use. Positive survey answers alone do not establish demand.
What is MVP for agile?
In agile product development, an MVP is the smallest product that enables meaningful learning from customers. It is not automatically the first sprint's output. A team can deliver working software without testing whether anyone wants it.
