What It Is
The Mom Test is a set of three rules for conducting customer conversations that produce reliable information. Named after the observation that even your mother will give you useful data if you ask the right questions, it shifts focus from pitching your idea to understanding the other person's real behavior, problems, and context. It was introduced by Rob Fitzpatrick in The Mom Test (2013).
When to Use It
- Any time you are talking to a potential customer, user, or stakeholder to learn about their needs.
- During early-stage discovery when you have an idea but have not yet validated the problem.
- When you notice conversations keep producing encouraging but vague feedback ("That sounds great!").
- Before building anything — and continuously after, whenever you need to learn rather than sell.
How It Works
Rule 1: Talk About Their Life, Not Your Idea
Ask about what they already do, what problems they encounter, and how they currently solve them. Do not describe your product and ask if they would use it.
- Bad: "Would you use an app that tracks your groceries?"
- Good: "Walk me through how you decided what to buy last time you went shopping."
Rule 2: Ask About Specifics in the Past, Not Generics or the Future
People are unreliable predictors of their own future behavior. Opinions and hypotheticals are worthless. Specific past actions are gold.
- Bad: "How often would you use this?"
- Good: "When was the last time this happened? What did you do?"
Rule 3: Talk Less, Listen More
Your job in a learning conversation is to extract information, not to impress. Every minute you spend talking is a minute you are not learning. If you are talking more than 30% of the time, you are doing it wrong.
Key Principles
- Compliments are noise. "That's a cool idea" teaches you nothing. Deflect compliments and dig into behavior.
- The facts are in the past. "I would definitely pay for that" is a future promise and nearly meaningless. "I paid $200 last month for a workaround" is a bankable signal.
- Bad data is worse than no data. A conversation that produces false confidence (because you pitched instead of listened) actively harms your decision-making.
- You do not need to mention your idea at all. The best discovery conversations often end without the other person knowing what you are building.
- Anyone can give you good data. The rules are about your questions, not their expertise. Apply them with domain experts, casual users, and yes, your mom.
Common Mistakes
- Fishing for compliments disguised as questions. "Don't you think it would be great if..." is a pitch, not a question. Rewrite it as a question about their life.
- Accepting hypothetical commitments as validation. "I'd totally buy that" is not a purchase. Push for a real commitment: a pre-order, a letter of intent, time on their calendar for a pilot.
- Talking about the solution too early. Once you reveal your idea, the conversation shifts from learning to selling. Hold it back until you have extracted the information you need.
Source
Rob Fitzpatrick, The Mom Test (2013), Chapters 1-3. The three rules are introduced in Chapter 1, with Chapters 2 and 3 expanding on how to avoid bad data and how to structure conversations.
