How to Validate a Startup Idea Before You Build Anything

Name the riskiest assumption

Every idea rests on a stack of assumptions: that the problem is painful, that your solution fixes it, that people will pay, that you can reach them. Only one of those is most likely to kill you. Write them all down, rank them, and test the top one first.

For most consumer ideas the risk is demand. For most B2B ideas it is willingness to pay and access to buyers. Test the one that applies to you, not the one that is easiest.

Cheap tests that produce real signal

Validation does not require software. It requires evidence that someone will change their behaviour.

• →Twenty problem interviews where you ask about the last time it happened, not about your idea.
• →A landing page with a real offer and a measured conversion rate from paid or community traffic.
• →A concierge test: deliver the outcome manually for five customers before automating it.
• →A pre-sale, deposit or signed letter of intent — the only signal that survives contact with a spreadsheet.

Reading the results honestly

Compliments are not validation. Look for behaviour: money moved, time committed, an existing tool abandoned, a colleague introduced. If nobody does anything after saying they love it, you learned something valuable and cheap.

When two or three signals line up, stop researching and build the smallest version that delivers the outcome for real. That is the moment an MVP earns its cost.