Messy search page
What should I ask AI to validate a startup idea?
The weak prompt is 'is this a good idea?' The better prompt forces the AI to separate assumptions, evidence, buyer behavior, and decision gates.
Use these terms
Riskiest assumption
Finds the belief that would kill the idea fastest if it were false.
Assumption mapping
Ranks beliefs by importance, uncertainty, evidence, and next test.
Evidence standard
Defines how much proof is enough before continuing, changing direction, or stopping.
Demand signal
Separates polite interest from real behavior, urgency, budget, or active workaround.
Fake door test
Measures demand before building the full product.
Willingness to pay
Tests whether the pain is strong enough for someone to spend money, time, or political capital.
Problem-solution fit
Separates the customer pain, current alternatives, proposed solution, and evidence quality.
Prompt pattern
Weak ask
Validate my startup idea and tell me if it will work.
Exact Terms ask
Evaluate this startup idea by identifying the riskiest assumption, mapping assumptions, defining an evidence standard, naming demand signals, designing a fake door test, testing willingness to pay, and assessing problem-solution fit.
Common questions
Why is validation language dangerous?
It can invite approval-seeking. Research and evidence language pushes the AI toward falsification and decision quality.
What should founders test first?
Start with the riskiest assumption and the smallest honest test that could disprove it.