Mom Test

Thinking Tools3.1 Finding the Current FactsAlso known as The Mom Test

Originated by Rob Fitzpatrick (The Mom Test, 2013).

Customer conversations built on three rules: talk about their life, ask about specific past behavior, and listen more than talk, so that politeness cannot pass for evidence.

Five questions to a restaurant owner, and what each one draws out

At each of the founder's turns, pick a question: the owner's answer shows, with a verdict on whether it is a fact about her past or a compliment about the idea. The owner's turns say what to write down. Illustrative.

Founder

Restaurant owner

Founder

Restaurant owner

Select a stage to read it, or play the sequence.Opening. The founder is building an app that tracks kitchen stock. He has twenty minutes with the owner of a sixty-seat bistro.Pick a move to see the reply.Oh, for sure. That sounds really useful.A compliment: the pitch told her what the founder hopes to hear.Sunday night I count the walk-in, then text three suppliers. Thursday I top up whatever ran low.A fact: her actual routine, with no idea on the table to be polite about.Her routine. Note the steps in her words: a Sunday count, texts to three suppliers, a Thursday top-up. Her routine is the baseline any product has to beat.The problem. The founder suspects running out of stock mid-service is the pain. He needs to know if it has happened, and when.Pick a move to see the reply.Two Fridays ago. Short ribs, at eight o'clock. We comped a table and pulled the special.A fact: a date, an item, and a cost.Sometimes, I guess. Everyone does.A generality: easy to agree with, and it says nothing about how often or what it costs.The cost. A specific night with a specific loss is one data point. Ten owners with nights like it would be a pattern.The workaround. A problem people have tried to fix is a problem they care about. One they have shrugged off may not be worth solving.Pick a move to see the reply.Yeah, that would've been great.Agreement with the founder's solution: it costs her nothing to say yes.I had my sous chef hang a clipboard on the walk-in door. It lasts about a week, every time.A fact: she built a fix, and it keeps failing.Her fix. A homemade workaround that keeps breaking is strong evidence: the need is real, and what she has does not meet it.The money. Price is the question founders most want answered and customers can least answer about the future.Pick a move to see the reply.Maybe fifty a month? Hard to say.A guess about the future: people are poor judges of what they would pay for something they have never bought.A consultant built me an inventory spreadsheet for $1,200. Nobody used it after the first month.A fact: money already spent, and why the last fix failed.Money spent. Money already spent sets a floor on what the problem is worth to her, and the failed spreadsheet says what to avoid.The close. The end of the conversation is where a founder is most tempted to ask for approval.Pick a move to see the reply.Sure. Come at nine, and bring coffee.A commitment: she is giving up a Sunday morning, which a polite yes never costs.Honestly, I love it. You should go for it.A compliment: pleasant, and worth nothing as evidence.A commitment. Time, money, or an introduction given up for the next step is the signal. Count these across interviews, and ignore the praise.

Facts and commitments drawn out: 0 of 5Facts and commitments drawn out: 1 of 5Facts and commitments drawn out: 2 of 5Facts and commitments drawn out: 3 of 5Facts and commitments drawn out: 4 of 5Facts and commitments drawn out: 5 of 5

  1. Opening (Founder). The founder is building an app that tracks kitchen stock. He has twenty minutes with the owner of a sixty-seat bistro.
    • I'm building an app that tracks your stock. Would you use it? Oh, for sure. That sounds really useful. A compliment: the pitch told her what the founder hopes to hear.
    • Walk me through how you order produce and meat each week. Sunday night I count the walk-in, then text three suppliers. Thursday I top up whatever ran low. A fact: her actual routine, with no idea on the table to be polite about.
  2. Her routine (Restaurant owner). Note the steps in her words: a Sunday count, texts to three suppliers, a Thursday top-up. Her routine is the baseline any product has to beat.
  3. The problem (Founder). The founder suspects running out of stock mid-service is the pain. He needs to know if it has happened, and when.
    • When did you last run out of something during service? Two Fridays ago. Short ribs, at eight o'clock. We comped a table and pulled the special. A fact: a date, an item, and a cost.
    • Do you ever run out of things? Sometimes, I guess. Everyone does. A generality: easy to agree with, and it says nothing about how often or what it costs.
  4. The cost (Restaurant owner). A specific night with a specific loss is one data point. Ten owners with nights like it would be a pattern.
  5. The workaround (Founder). A problem people have tried to fix is a problem they care about. One they have shrugged off may not be worth solving.
    • Wouldn't a phone alert have saved you that night? Yeah, that would've been great. Agreement with the founder's solution: it costs her nothing to say yes.
    • What did you do about it afterward? I had my sous chef hang a clipboard on the walk-in door. It lasts about a week, every time. A fact: she built a fix, and it keeps failing.
  6. Her fix (Restaurant owner). A homemade workaround that keeps breaking is strong evidence: the need is real, and what she has does not meet it.
  7. The money (Founder). Price is the question founders most want answered and customers can least answer about the future.
    • How much would you pay for something that fixed this? Maybe fifty a month? Hard to say. A guess about the future: people are poor judges of what they would pay for something they have never bought.
    • Have you ever paid for anything to fix this? A consultant built me an inventory spreadsheet for $1,200. Nobody used it after the first month. A fact: money already spent, and why the last fix failed.
  8. Money spent (Restaurant owner). Money already spent sets a floor on what the problem is worth to her, and the failed spreadsheet says what to avoid.
  9. The close (Founder). The end of the conversation is where a founder is most tempted to ask for approval.
    • Could I come on Sunday and watch you do the count? Sure. Come at nine, and bring coffee. A commitment: she is giving up a Sunday morning, which a polite yes never costs.
    • So, do you think it's a good idea? Honestly, I love it. You should go for it. A compliment: pleasant, and worth nothing as evidence.
  10. A commitment (Restaurant owner). Time, money, or an introduction given up for the next step is the signal. Count these across interviews, and ignore the praise.

An example

A founder building a stock-tracking app for independent restaurants had spoken to five owners, and all five said they would use it. He had pitched the app in the first minute of every call. For the next seven conversations he kept the app out of it and asked only about the last month of ordering: how the count happened, the last time something ran out, what they did about it, and what they had spent on it.

The picture changed. Five of the seven had run out of a protein on a weekend night in the past month, and four had built a workaround, mostly clipboards and group texts, that kept breaking. Two had paid for a fix that nobody used. Produce and dry goods never came up. One owner let him watch her Sunday count. He narrowed the product to weekend protein stockouts, dropped the produce module he had planned, and set up a pilot with the two owners who had already paid for a fix that failed.

Why it works

A question about the future asks for a guess, and a guess offered to someone hopeful comes out kind. A question about the past can only be answered with what happened, and few people will invent a Tuesday that never happened to spare a founder's feelings. Keeping the idea out of the conversation removes the thing to be polite about, so the owner talks about her week and the founder learns where the problem actually sits.

The second half of the move is how the answers are counted. Praise costs nothing, so it carries no information. A workaround someone built, money they spent, or a morning they hand over all cost something, and only someone who has the problem pays those costs. Counting costs and ignoring praise turns a pile of pleasant conversations into evidence a decision can rest on.

The thinking behind it

Rob Fitzpatrick Rob Fitzpatrick, a founder, set the method out in The Mom Test (2013), a short practitioner's guide to talking with customers. The name carries its premise: even a founder's own mother will give a kind answer about the idea, so asking whether it is good teaches nothing. His fix is a conversation that passes even her test: questions about the customer's life and what they have actually done in the past, and none about the idea. It is a guide drawn from practice, without controlled studies behind it.

How to run it

  1. Pick the behavior to learn about. Name the part of the customer's life the idea touches, such as weekly ordering, and leave the idea itself off the list.
  2. Write questions about the past. Phrase each one around the last time something happened: when it was, what they did, and what it cost.
  3. Keep the idea out of the room. If asked about the product, say the aim is to learn how the work gets done today, with more to share at the end.
  4. Press for specifics. When an answer turns general, ask for the most recent example and walk through it step by step.
  5. Ask what they have tried and spent. Workarounds and money already spent show how much the problem matters to them.
  6. Close by asking for a cost. Ask for a next step that takes their time, money, or reputation, such as a visit or an introduction.
  7. Count across conversations. Write up facts and commitments the same day, and look for a pattern only after five or more interviews.

Where it goes wrong

The most common slip is the pitch that sneaks back in. A founder who keeps the idea out for ten minutes and then describes it, hoping for a reaction, gets the compliment the method was built to avoid. If the notes from a call hold the word love, the idea was in the room.

The second is treating every answer as a fact. A story about the past can still be vague; without a date, an amount, or a name it is an opinion with a past tense. Ask for the most recent instance until one appears.

The third is stopping at three friendly conversations. A handful of interviews with warm contacts will agree with each other for the wrong reasons. Talk to people who owe the founder nothing, and wait for a pattern before deciding.

When to use something else

  • Customer Discovery. When the conversations are one step in testing a whole business claim, from who has the problem to whether they will commit before anything is built.
  • Jobs to Be Done. When the problem is confirmed and the question is what progress the customer is trying to make and what they would switch from.
  • Empathy Map. When a team already has interview notes and needs to share one picture of what the customer says, thinks, does, and feels.

Where it sits in SOLVE eX

Go to the source

Back to the catalog