Chapter 1: The Mom Test

Core Idea

Don’t ask anyone whether your business is a good idea, because that question invites polite lies. The Mom Test is three rules for asking questions that even someone who loves you can’t lie about: talk about their life, ask about specific past events, and listen more than you talk.

Context (from the Introduction)

Learning from customers is fragile work. Heavy-handed questions (“do you think it’s a good idea?”) destroy the truth, and timid questions never reach it.
Bad conversations are worse than none. They produce false positives, which lead you to over-invest money, time, and team.
Scope: the book covers “customer conversation” (lowercase), meaning how to talk to customers and learn from them. It is not a retelling of Steve Blank’s Customer Development (uppercase). It assumes you already agree that you should talk to customers and focuses on how.

Frameworks Introduced

The Mom Test:
1. Talk about their life instead of your idea.
2. Ask about specifics in the past instead of generics or opinions about the future.
3. Talk less and listen more.
- When to use: every early customer conversation, and as a check on any question before you ask it.
- How: before asking, check whether the question is about your idea, about the future, or about generic habits. If it is, rewrite it to ask about something concrete they did recently. If the person leaves without knowing you have an idea, you did it right.
- Why it works: people rarely lie about specific things that already happened. They lie about opinions and the future, especially when they can see your ego is involved.
Good question / bad question: a drill of 14 questions, sorted by whether they pass the test (table below).
They own the problem, you own the solution: you aren’t allowed to tell customers what their problem is, and they aren’t allowed to tell you what to build. Gather facts about their problems, cares, constraints, and goals. Then make your own leap to a solution and confirm it through commitment (Ch 5).

Key Concepts

False positive: a signal that the idea is good when it isn’t. It is the most dangerous output of a bad conversation.
Useful conversation: one that gives you concrete facts about the customer’s life and world view, which in turn improve your business.
Failure points: most ideas can fail in several places (for example, the medium and the content). A good conversation probes each one.
Price anchor: what the customer pays today for a workaround, which tells you the range of value you are competing in.
Budget owner: in B2B, the person whose budget pays. This is often not the person you are talking to.

Mental Models

Treat every customer conversation as bad by default. Fixing it is your job.
The most common mistake is mentioning your idea too soon, not too late. Leaving the idea out is the easiest and biggest improvement you can make.
A vague idea (“old people like cookbooks and iPads”) always seems good. Only specifics expose the hard questions, such as which recipes to include or how people will hear about it.
Prefer watching over asking. Seeing someone do the task beats any description. If you can’t watch, “talk me through the last time” is the next best option.
Decide which alternative you want to be compared with. A customer may pay £100/month for a workaround app while also paying £120k/year to an agency. You can present yourself as a replacement for either.

Anti-patterns

“Do you think it’s a good idea?”: only the market can judge an idea. Unless you’re talking to a deep industry expert, the answer is opinion noise.
Hypothetical purchase questions (“Would you buy…”, “How much would you pay…”, “Would you pay X for Y?”): people are optimistic about their future and want to please you. Adding a number makes the answer feel rigorous without making it true.
Collecting feature requests as a to-do list: customers don’t know how to solve their problems, and you aren’t building the product by committee.
Leading people into emotion: it’s easy to get someone to say untied shoelaces are “the worst”. If they really cared, they would already tie a double knot.

Rules of Thumb (author’s wording)

Customer conversations are bad by default. It’s your job to fix them.
Opinions are worthless.
Anything involving the future is an over-optimistic lie.
People will lie to you if they think it’s what you want to hear.
People know what their problems are, but they don’t know how to solve those problems.
You’re shooting blind until you understand their goals.
Some problems don’t actually matter.
Watching someone do a task will show you where the problems and inefficiencies really are, not where the customer thinks they are.
If they haven’t looked for ways of solving it already, they’re not going to look for (or buy) yours.
People stop lying when you ask them for money.
While it’s rare for someone to tell you precisely what they’ll pay you, they’ll often show you what it’s worth to them.
People want to help you. Give them an excuse to do so.

Worked Example

The iPad cookbook, failed and then passed. The idea is a $40 cookbook app for the iPad, tested on mom.

Failing version: the son announces he has a business idea, which puts his ego on the table. He asks whether she likes her iPad (a leading question) and whether she’d buy a cookbook app (a hypothetical). He skips past her lukewarm “hmm” and adds the price, recipe sharing, a shopping-list app, and celebrity-chef videos until she says it sounds amazing and asks whether it will have pictures. That is a compliment plus a feature request made to seem engaged. He takes it as validation, repeats this a few times, quits his job, and nobody buys.

Passing version:

Question	What the question does	What it revealed
“What do you usually do on it?”	Generic (weak)	News, sudoku, friends. Low value.
“What’s the last thing you did on it?”	Specific past event	She planned a trip, so she uses it for practical tasks too.
“Did you use an app for that?”	Slight nudge toward the topic	No, she used Google and doesn’t browse the App Store. Marketing channel risk.
“How did you find the apps you have?”	Digs into a surprise	A weekly newspaper column on apps. Traditional PR might reach people like her.
“Where did those new cookbooks come from?”	Probes the second failure point (content)	Gifts she never opened, and she doesn’t need generic recipes. Possible gift market; younger cooks may be a better segment.
“What’s the last cookbook you bought yourself?”	Attacks a generic answer	A vegan cookbook three months ago. Experienced cooks still buy niche books.

Outcome: two candidate segments, each with a key risk. One is niche recipes for experienced cooks, reached through PR. The other is generic recipes for younger cooks, reached through the App Store, where the open question is whether they’ll pay a premium. She never learned there was an idea.

Good Question / Bad Question

Question	Verdict	Why, or how to fix it
“Do you think it’s a good idea?”	Bad	Ask them to show how they do it now, what they love and hate, what they’ve tried, and where they lose money. Then judge the idea yourself.
“Would you buy a product which did X?”	Bad	Ask how they solve X now, what it costs in money and time, and what happened last time. If they haven’t solved it, ask why not.
“How much would you pay for X?”	Bad	Ask what the problem costs them, what they pay now, and what budget exists.
“What would your dream product do?”	Sort-of-okay	Useful only if you follow up on why they want each feature.
“Why do you bother?”	Good	Gets from the perceived problem to the real motivation.
“What are the implications of that?”	Good	Separates problems they’d pay to solve from mere annoyances. Gives a pricing signal.
“Talk me through the last time that happened.”	Good	Shows you behaviour instead of opinion.
“Talk me through your workflow.”	Good	Reveals their day, tools, contacts, constraints, and what you’d need to integrate with.
“What else have you tried?”	Good	Reveals current spending, what they hate, switching cost, and whether they even searched.
“Would you pay X for a product which did Y?”	Bad	Ask about current behaviour, or if you’re far enough along, ask for a deposit or pre-order.
“How are you dealing with it now?”	Good	Gives workflow detail and a price anchor.
“Where does the money come from?”	Good (must-ask in B2B)	Finds the budget owner and who else can block the deal.
“Who else should I talk to?”	Good	End every conversation with it. A refusal means you were too pitchy or formal, or they don’t care.
“Is there anything else I should have asked?”	Good	Lets them correct your line of questioning. You’ll outgrow it as you learn the industry.

Key Takeaways

Never ask for an opinion about your idea. Ask about the customer’s life.
Turn every generic or future question into “the last time” or “what did you do”.
Probe every failure point of the idea (channel, content, price), not only the one you find interesting.
Use what they spend on workarounds today as your price anchor.
End conversations with “Who else should I talk to?”
Deciding what to build is your job. Confirm it later with commitments, not opinions.

Connects To

Ch 2: what to do when bad data (compliments, fluff, ideas) gets in anyway.
Ch 3: apply the test to important questions, not trivia.
Ch 5: commitment and advancement confirm the solution you chose.
Contextual inquiry (UX research): “talk me through the last time” is close to observing real tasks.