By Tech Entrepreneur · 14 September 2026 · 2 minute read
Before you begin
Choose one user, one job and one end-to-end experience. Identify the assumption that could make the idea fail. A manual service, clickable prototype or small working feature may be enough to test it. Define what you need to observe before building more. Look at whether users complete the task and return, not just whether they like the presentation. Keep a record of the trade-offs made.
Test a complete user task
Imagine a booking idea for a small workshop. The first test might let one customer request an available slot and receive a manually confirmed booking. A large dashboard, loyalty programme and complex reporting may not be necessary to learn whether the basic flow helps. Define the task from start to finish, including what happens if the requested time is unavailable. A prototype that only shows the successful middle screen leaves important assumptions untouched.
Choose the least expensive credible test
A sketch can test understanding, a clickable prototype can test navigation, and a small working flow can test practical use. Choose the format based on the uncertainty you need to resolve. Be clear with participants about what is simulated and what actually works. Observe the task rather than explaining every step. Record where people hesitate, what they expect and what they cannot complete.
Use evidence to set the next scope
MVP Planning should produce a decision about the next test, not an automatic permission to build the entire wish list. Use Customer Discovery to check whether the task matters and Writing a Startup Product Brief to explain the implementation work. Keep excluded features visible so they do not quietly return as assumptions during design. Add scope when the evidence justifies it.
Put the notes to work
Prepare a brief with the free planning tool →