The Mom Test
by Rob Fitzpatrick
How to talk to customers and learn if your business is a good idea even when everyone is lying to you.
The short route — our review and key takeaways, 5 min read. The long route — buy the book on Amazon if you want to go deeper. Both routes work.
About the author
Rob Fitzpatrick
The short route
northstar's take on this book
The Mom Test is the shortest, sharpest, and most under-priced book in the PM canon. Rob Fitzpatrick wrote it in 2013 as a 130-page self-published manual, and it has quietly become the single book most experienced PMs and founders recommend to first-time founders without hesitation. The thesis fits in one sentence: stop asking your customers whether your idea is good — they will lie politely — and instead ask them about their lives, their problems, and the workarounds they already use.
Its central craft contribution is a small set of conversational moves: anchor every conversation in the customer's past behavior, not their future hypotheticals; deflect compliments back to specifics; and treat any commitment that doesn't cost the customer something (time, money, reputation) as worthless data. These rules sound obvious in the abstract and are unbelievably hard to apply in a live conversation — which is exactly why the book has staying power. PMs and founders re-read it before every new wave of customer interviews.
Timing-wise, the book has had a strange long tail. It came out in 2013, around the same time as The Lean Startup, but didn't really go viral until 2018-2020 when the indie hacker / bootstrapped founder community made it a foundational text. Y Combinator partners and prominent founders started recommending it publicly, and by 2022 it was on most founder reading lists despite zero marketing push from the author.
Newsletter
Like this review? Get the next book breakdown in your inbox.
One book or case study every few days — the short route, with the takeaways and our honest read. Free.
Free forever. Unsubscribe anytime. No spam.
The most common misreading is treating it as a script. PMs and founders sometimes try to literally copy the example questions in the book, deploy them in a customer interview, and feel disappointed when the conversation feels stilted. The book is teaching a posture, not a script — the posture is being genuinely more interested in what the customer already does than in what you want them to think. The questions are downstream of that posture and have to be improvised in the moment.
Its limitation is that it's written from a B2B founder's perspective. The examples are mostly about selling enterprise software, and the 'meaningful commitment' bar (a pre-order, an intro to a budget holder, a paid pilot) is calibrated for that world. For consumer products, the framework still applies but the commitment signals look different: deleting a competitor's app, repeatedly using a prototype, or sharing it with a friend without prompting are the consumer equivalents.
For Indian PMs and founders, the book is genuinely cross-cultural in a way most American PM books aren't. The 'people will lie politely' dynamic is, if anything, more pronounced in Indian customer conversations where direct critical feedback is socially expensive. Fitzpatrick's techniques work as well in Bangalore as they do in San Francisco — possibly better, because the politeness gap is wider. The book is a strong recommendation for any Indian founder doing problem discovery for a product idea.
Pair with Continuous Discovery Habits for the operating cadence (how often, with whom, in what structure), and read this one first. Mom Test gives you the in-conversation craft; Torres gives you the program around it.
Key concepts
- Anchor questions in past behavior — Ask about what the customer actually did last week, not what they'd hypothetically do with your product. Past behavior is the only data that doesn't lie.
- Deflect compliments to specifics — When customers say nice things about your idea, redirect to concrete behavior: 'when did you last try to solve this?' Compliments are social politeness, not signal.
- Meaningful commitment as the only real signal — Customers are only validating you if they part with something — time, money, reputation, or a hard introduction. Anything cheaper is polite noise.
- Bad data vs. no data — Badly run customer interviews produce false confidence, which is worse than no information at all. Better to skip the conversation than to mis-ask the question.
- The three deadly questions to avoid — Don't ask 'would you buy this?', 'do you like this idea?', or 'what features would you want?' All three reliably produce misleading answers.
Who should read it
Every first-time founder. Every PM about to run their first round of customer interviews. Senior PMs and founders revisiting customer discovery after a stretch of feature shipping. Possibly the highest ROI-per-page book on this entire list.
Frequently asked
4 questionsSee it in practice
Case studies that demonstrate the ideas
More in Product Management