Test a business idea before building the whole product

Test a business idea by identifying the assumption most likely to make it fail. It may be the problem, the buyer, the delivery method, or the willingness to pay. Choose a small test that produces evidence about that assumption rather than building every feature first.

Separate the assumptions

A fictional document service assumes freelancers struggle with project briefs, will share a brief for review, and will pay for a clearer scope. Those are three different questions. A polished landing page can test communication or interest, but it does not establish whether the service delivers a useful result.

Design a bounded test

The numbers are an authored test design, not completed research. Define your own decision rule before starting and record changes to it.

Give AI a testing brief

Identify the main assumptions in this idea.
Choose one that could change the decision to build.
Propose a small, reversible test with an honest description of what exists.
Define observations, costs, time limit, and what would change the plan.
Do not invent users, results, demand, or statistical confidence.
Idea and available resources: [facts]

Keep the offer truthful

If the test is a manual service, say so. If a prototype cannot perform an action, do not imply it can. A waitlist is interest in receiving information, not a paid customer count. A click is not a completed purchase.

After the test, write what happened, what you learned, and what remains uncertain. A useful result may be narrowing the audience, changing the output, or stopping the idea. The test should give you permission to revise the plan, not become a ritual that always concludes “build more.”