The Mom Test for Product Managers

If you haven't read The Mom Test by Rob Fitzpatrick, stop reading this and go buy it. It is the only book on customer conversations that actually matters.

The premise is brilliant: People lie to you.

They don't do it maliciously. They lie because they want to protect your feelings. If you go to your mom and say, "Mom, I have an idea for an app that organizes recipes. Do you think it's a good idea?" she will say, "Oh honey, that's incredibly smart! I would use that."

She is lying. She has never used an app in her life. She still has recipes written on index cards from 1982.

If your mom can lie to you, your users can too. Here is how you apply The Mom Test to product management so that lying becomes impossible.

The Three Rules of The Mom Test

To extract the truth, you must design questions that are immune to politeness.

1. Talk about their life, not your idea

The moment you introduce your idea into the conversation, the data is corrupted. They are no longer thinking about their actual workflow; they are thinking about how to react to your brilliance.

Never say: "We are thinking about building a feature that automates invoice reminders." Instead, ask: "How are you currently handling unpaid invoices?"

If they tell you they just wait for people to pay, then unpaid invoices aren't actually a burning problem for them. Don't build the automation.

2. Ask about specifics in the past, not generics or opinions about the future

We discussed this in the user interviews article, but it cannot be overstated. Future statements are fantasy. Past behavior is fact.

Never say: "Would you be willing to pay $10 a month for this?" Instead, ask: "What tools have you purchased in the last six months to solve this problem?"

If they say they haven't spent a dime solving the problem, they are absolutely not going to pay you $10 a month for it. People only pay to solve problems that are currently causing them active financial or emotional pain.

3. Talk less, and listen more

If you are talking for more than 20% of the meeting, you are pitching, not learning. You do not learn anything while your mouth is moving.

Spotting False Positives

When I was doing discovery for an enterprise feature earlier this year, I had a user tell me, "I love this idea. This would save me so much time. You guys should definitely build this."

A younger version of me would have taken that straight to the engineering team. But that statement is actually a false positive. It is a compliment.

Compliments are the enemy of truth. When someone compliments your idea, they are ending the conversation. They are giving you a gold star so you will go away.

When you get a compliment, you have to deflect it and anchor back to reality. User: "This is a great idea." You: "Thanks. But I want to make sure I'm not over-engineering. Talk me through the last time you actually had to deal with this issue. How long did it take you manually?"

If they can't remember the last time it happened, the compliment was fake.

Seeking Rejection

You should walk into every customer conversation actively trying to ruin your own day.

You want to find the fatal flaw in your logic before you spend three sprints writing code. If you are looking for validation, you will find it, because humans are polite. If you look for invalidation, you will find truth.

Ask questions that invite them to shoot you down. "We thought this was a major issue, but honestly, is this even in the top five problems you deal with on a Monday?"

Give them an out. Let them tell you your premise is wrong. If they say, "Actually, no, it's not a big deal," congratulate yourself. You just saved your company a month of wasted engineering effort.

The Physics of Commitment

The ultimate test of whether someone actually cares about your solution is if they are willing to give up something of value right now.

In The Mom Test, this is called advancing the commitment.

Value isn't just money. Value can be time, social capital, or reputation. If a user says they desperately need your feature, ask for a commitment.

"Great. I can get my engineers to start on a prototype next week. Can we schedule a 90-minute working session with your whole team next Thursday to map out the exact data schema we need?"

If they balk and say they are too busy, the problem isn't real. If a problem is actually bleeding them dry, they will give you the 90 minutes. Commitment is the only currency that matters in product discovery.

---

FAQ

Does The Mom Test apply to B2B and B2C products?

Yes. Physics is physics. Whether you are selling to a bored teenager scrolling TikTok or a Chief Financial Officer at a Fortune 500 company, human psychology remains the same. They will both lie to you to avoid awkwardness.

Can I use these techniques in surveys?

Surveys are inherently flawed because you cannot ask follow-up questions when someone drops a generic compliment. However, you can write survey questions that focus exclusively on past behavior ("How many times did you log in last week?") rather than future intent ("Will you log in next week?").

What if I don't have an idea yet, and just want to explore a space?

That is exactly what The Mom Test is designed for. Schedule a call, ask them what is currently broken in their workday, and pull the thread on whatever causes them the most obvious emotional distress. Build to solve the distress.