The Mom Test, explained: why your mother loves your idea and customers might not

The Mom Test in plain terms: ask about what people have actually done, distrust compliments and predictions, and judge interest by what they commit to.

Instant Expert EditorialPublished Updated 6 min read

The Mom Test is Rob Fitzpatrick's short book about one problem: people are too polite to tell you your idea won't sell. The book's fix fits in two sentences. Ask people about their own lives and what they have actually done, not about your idea or what they might do someday. Then judge their interest by what they commit to, not by what they say.

The subtitle promises to show you whether your business is a good idea "when everyone is lying to you" (The Mom Test). The author's site says it takes about two hours to read, and Fitzpatrick's site says a second edition is in progress. Below is the approach in our own words, with a made-up example.

"Is my idea good?" only has kind answers

Ask your mother whether your app idea is good and the answer will be yes. It would be yes for a subscription service for artisanal ice cubes. Fitzpatrick's point is that the question itself is badly framed, and anyone you ask it will give you some degree of misleading feedback (Unusual Ventures). People are kind, and predicting your own future behavior is hard.

The name sets the standard: questions so grounded in the other person's experience that even your mother couldn't mislead you with an answer.

The three rules

Paraphrased:

  1. Talk about their life, not your idea. Once you describe your product, the conversation becomes about being nice to you.
  2. Ask about specific things that already happened. "When did this last happen?" beats "Do you usually...?" and "Would you...?"
  3. Let them do most of the talking. Every minute you spend explaining is a minute they spend reacting to you.

A worked example: rewriting four bad questions

Here is a made-up example. A founder is considering software that fills last-minute cancellations at dental practices by texting patients on a waitlist. They have three calls booked with office managers at independent practices, and every question in their first draft invites a guess or a compliment.

First draftRewritten
Would you use an app that fills cancelled appointments automatically?Tell me about the last time a patient cancelled on the same day. What happened next?
How much would you pay for something like this?What does an empty chair cost the practice? Have you tried anything to fill them, and what did that cost?
Do cancellations bother you?How many same-day cancellations did you have last week? How do you know?
Don't you think patients would like getting a text?When an opening comes up today, how do you contact patients? Who does it?

The rewritten questions can produce bad news, which is what makes them useful. If the last same-day cancellation was filled within ten minutes from a paper list at the front desk, the founder has learned something important before writing any code. If the practice loses two or three a week and nobody has time to call around, ask who notices, what they have tried and who would pay for a fix. Customer interview questions has more questions built the same way.

Three kinds of answer to distrust

The book's second chapter covers what the author's teaching notes call the three major causes of bad and biased data (The Mom Test: teacher resources). In our words:

  • Compliments. "That sounds great" tells you the person is polite. Thank them and steer back to facts: "How do you handle that today?"
  • Generalities and predictions. "Usually," "always," "never" and "would" describe a pattern they believe in or a guess about the future. Unusual's summary of the advice is to anchor these back to specifics from the past.
  • Ideas, requests and strong opinions. A feature request is a clue to an underlying problem. Ask what the feature would let them do and why that matters.

A good conversation ends with a commitment

Pleasant and useful are different things. According to Unusual's summary, the book counts a meeting as a success when it ends with a commitment or a clear next step, where commitment means the other person gives up something they value: their time for a follow-up with a set purpose, their reputation by introducing you to a colleague or decision maker, or their money (Unusual Ventures). The more they give up, the more seriously you can take what they said.

In the dental example, strong outcomes after a first call might be:

  • The office manager agrees to share a count of last month's same-day cancellations, with no patient details.
  • The office manager introduces the founder to the practice owner, who would approve any purchase.
  • The owner agrees to pay for a one-month trial.

By that standard, "let me know when it's ready" is a polite way to end the call. It moves nothing forward. Market validation shows how to rank these signals before you build.

Prepare, keep it casual and write things down

Three habits from the book, as Unusual summarizes them:

  • Pick your three most important questions before each conversation. At least one should be a question whose answer could sink the idea. For the founder: "Do practices already fill most same-day cancellations without help?"
  • Keep it informal. If it feels like they are doing you a favor by talking to you, the setup is probably too formal.
  • Take notes and review them with your team promptly. Record exact words, strong emotions and specific names and companies, then update your three big questions. (How to take customer interview notes has a format.)

What the book won't tell you

The Mom Test helps you find out whether a problem exists and how people deal with it now. It won't tell you how common the problem is across a market or whether people can use what you build. Those need a count from a larger sample and a usability test with a working prototype.

Whom you ask matters as much as what you ask. Careful conversations with people who don't have the problem will still mislead you, so choose interviewees with care. And once you understand the problem, a solution interview is where your idea finally gets to come out.

Your next step

Write your three most important questions, including one that could prove you wrong. Rewrite every question that mentions your product or starts with "would you" so it asks about a recent, specific event. Then decide what commitment you will ask for at the end of each conversation.

If you can't find people to have these conversations with, describe them on Instant Expert, such as "office managers at independent dental practices," or start from operations professionals in dental services. For the founder's three calls, say you offer $100 each for 30 minutes: with our fee on top (20% of what you pay), each costs $125, charged only if the person books (how it works). The amount is fixed in the invitation before the call, so the office manager is paid the same whether they love the idea or take it apart. Pay for their time, not for a favorable answer: paying for participation without paying for positive feedback explains how.

If you're the one being interviewed

The most useful thing you can give a founder is what you actually did last week, including the parts that make their idea look unnecessary. An Instant Expert invitation shows the founder's note and the amount you would receive, which is yours in full, and nothing about it depends on what you say. You have 7 days to book a call, send a written or voice answer, or decline. Payment goes out 24 hours after the call, by direct deposit, a gift card or a donation to a charity you pick, which we make for you.