You’ve just had a customer call. They said the idea was great. They said they’d definitely use it. They asked when it would be ready. You hang up feeling validated 😊 You shouldn’t 😬 Rob Fitzpatrick wrote The Mom Test to solve exactly this problem. The title comes from a simple observation: even your mum will lie to you about your business idea. Not because she’s dishonest - because she loves you and doesn’t want to hurt your feelings. Most customers do the same. They’re polite. They want to be helpful. And so they tell you what you want to hear. The Mom Test is a set of rules for asking questions in a way that makes it impossible for people to lie to you - not because you’ve forced them to be honest, but because you’ve stopped asking questions that invite dishonesty in the first place.Documentation Index
Fetch the complete documentation index at: https://userstory.zeroemdashes.com/llms.txt
Use this file to discover all available pages before exploring further.
The three rules
Talk about their life, not your idea The moment you describe your idea, you’ve invited an opinion. And opinions are useless. “Would you use something that did X?” is a bad question because it’s hypothetical - people are terrible at predicting their own future behaviour, and they’ll almost always say yes to avoid awkwardness. Instead, ask about what they actually do today. “How do you currently handle X?” “Walk me through the last time that happened.” “What did you try? How did it go?” You’re collecting facts about behaviour, not forecasts about intent. Ask about specifics, not generalities “Do you often have this problem?” is a weak question. “Tell me about the last time you had this problem” is a strong one. Specifics reveal truth. Generalities invite the answer they think you want. When someone says “I always struggle with X” - dig in. When did it last happen? What did you do? How did it turn out? The story behind the claim is where the real insight lives. Listen for things that matter, not compliments Compliments feel good but teach you nothing. “That sounds really interesting” is not a data point. What you’re looking for is:- Pain - Do they describe a problem that actually costs them time, money, or frustration?
- Workarounds - Have they already tried to solve it? Built something themselves? Paid for something that doesn’t quite work?
- Commitment - Have they done anything about it? Would they switch today if something better existed?
What bad questions look like
Fitzpatrick is blunt about this, and it’s useful to see the pattern:- “Do you think this is a good idea?” - You’re asking for an opinion. You’ll get a kind one.
- “Would you pay for something that did X?” - Hypothetical. Meaningless.
- “How much would you pay for this?” - Also hypothetical, and now you’re anchoring.
- “Would you use this if we built it?” - Yes. They will always say yes.
The questions that actually work
- “Tell me about the last time you dealt with [problem].”
- “What did you do? How did that go?”
- “Have you looked for solutions? What did you find?”
- “What’s the hardest part of your current process?”
- “How much of your time does this take up?”
- “Who else has this problem on your team?”