Idea validation

How to test a startup idea cheaply

You do not need funding or a finished product to find out whether an idea has legs. You need a few cheap experiments that produce evidence instead of encouragement.

Guide · 6 min read

Testing cheaply is a discipline: spend the least amount of time and money required to learn the one thing that would make you stop. Here are the experiments, roughly from cheapest to most involved.

1. The 10-conversation test (free)

Talk to ten people who have the problem. Do not pitch — ask how they handle it today, what they have tried, and what it costs them when it goes wrong. If the problem is not painful enough to describe vividly, that is your answer.

2. The fake-door test (a few dollars)

Put up a one-page description of the solution with a clear call to action ("Join the beta", "Get early access"). Share it where your audience already is. Real clicks and emails are far stronger signal than verbal interest.

3. The concierge test (your time)

Deliver the outcome manually for one or two people before automating anything. If you cannot get someone to want the result when you do all the work by hand, software will not save it.

4. The clickable prototype (small, fixed cost)

When the question is "will people actually use this workflow," a clickable prototype gives you behavior-based evidence without a full build. It is the step that turns vague interest into a real usage test — and it should cost a few hundred dollars, not tens of thousands.

What to skip while testing

• Logos, names, and brand polish.
• Accounts, billing, and infrastructure.
• Every feature except the single core workflow.

Put the experiments in order

Run the cheapest test that could kill the idea first. Only graduate to a prototype once conversations and a fake-door show there is real pull. For the full sequence, see how to validate a software idea before you build it.

← Back to Promptlaiy