The Mom Test: How to Ask Questions That Get Honest Answers
The Mom Test is a simple rule: ask people about their own life and past behavior instead of your idea, so they can’t lie to you even if they want to be nice. If a question is so grounded in what someone actually did that even your own mom couldn’t give you a false “that’s great, honey” answer, it passes. If your mom could dodge it with a compliment, it fails, and you rewrite it.
The rule comes from a short book called The Mom Test by Rob Fitzpatrick. You do not need to read the book to use it, but you do need to understand why the people you talk to keep lying to you, and how to stop asking questions that beg them to.
Why do people lie in customer interviews?
They are not really lying. They are being polite. A customer interview is a conversation where you learn how someone handles a problem, and the trouble is that almost everyone you talk to wants to be nice to you, especially when you’re a teenager who worked up the nerve to ask.
So when you say “I’m building an app that helps students study, would you use it?”, they hear “please tell me my baby is beautiful.” And they do. They say “yeah, that sounds useful,” you leave glowing, and you’ve learned absolutely nothing true.
Three kinds of answers feel like validation but are worthless:
- Compliments. “That’s a cool idea” costs the person nothing and tells you nothing.
- Hypotheticals. “I would definitely use that” is a promise about a future that hasn’t happened. People are terrible at predicting what they’ll do.
- Generic claims. “I always struggle with studying” is vague enough to mean anything.
The Mom Test exists to strip all three out of your conversations. If your friends and family loving your idea already convinced you it’s good, read why “my friends love it” is not validation first, because that feeling is the exact trap this whole method is built to break.
What makes a question pass the Mom Test?
A good question asks about the person’s real life and their past, not about your idea and their future. That’s the entire test. Truth lives in the past, because the past already happened and can’t be flattered.
Watch how one bad question becomes three good ones.
Bad: “Would you use an app that organizes your homework?” That’s about your idea and a hypothetical future. Mom says yes.
Good: “Walk me through how you kept track of your homework last week.” Now they have to describe a real thing that happened. Maybe they say “I just use the Notes app and I forgot two assignments.” That’s gold. That’s a real, specific, painful moment you did nothing to invent.
Good: “The last time you forgot an assignment, what happened?”
Good: “What do you use right now to stay on top of due dates?”
None of those mention your product. That’s on purpose. The second you describe your solution, the person stops reporting their life and starts reacting to your thing, and reactions are just politeness with extra steps.
The three-column rule for rewriting questions
Here’s the pattern in one table. The left column is what beginners ask. The middle explains why it fails. The right is the fixed version.
| What you want to ask (fails) | Why it fails | Ask this instead (passes) |
|---|---|---|
| “Do you think this is a good idea?” | Begs for a compliment | ”Tell me about the last time this problem came up." |
| "Would you pay $10 for this?” | Fake future money | ”What have you already paid to solve this?" |
| "Would you use a tool that did X?” | Hypothetical | ”How do you handle X today?" |
| "Don’t you hate when your planner is a mess?” | Leading; hands them the answer | ”What’s the most annoying part of staying organized?" |
| "Would this feature be helpful?” | Compliment bait | ”What did you do the last time you hit that wall?” |
The whole left column is about the future and about your idea. The whole right column is about the past and about their life. That is the only pattern you have to memorize. When you catch yourself writing a left-column question, flip it.
How to run a conversation that passes
You don’t schedule “a Mom Test.” You have a normal conversation and quietly follow the rules. Here’s a five-step shape you can use on a 15-minute call.
- Open with their world, not your pitch. “I’m a high school student trying to understand how people deal with [problem]. I’m not selling anything, I just want to hear your experience.” That lowers their guard and signals you want the truth, not a sale.
- Ask for the last specific time it happened. “When did you last run into this? Walk me through it.” A specific story is impossible to fake the way a general opinion is.
- Dig into what it cost. Time, money, stress, a missed grade, a lost customer. If the honest answer is “eh, it wasn’t a big deal,” you just learned the problem might be too small. Painful is good news for you.
- Find the current workaround. “What do you do about it now?” People duct-taping together group chats, spreadsheets, and sticky notes are showing you exactly where a product could live. If the answer is “nothing,” the problem probably doesn’t hurt enough.
- Shut up and let silence work. If you’re talking more than a third of the time, you’re doing it wrong. Ask, then wait. People fill silence with the real stuff.
Notice you never had to pitch. If they ask what you’re building, say “I’ll show you at the end, but I want your honest experience first,” and get back to their story. Your job here is detective, not salesperson.
What if they ask about my idea anyway?
Sometimes you can’t avoid it, or you genuinely need a reaction to a specific thing. When that happens, make the feedback expensive. Cheap feedback is a compliment. Expensive feedback costs the person something real, which forces honesty.
- Ask for a commitment, not an opinion. Instead of “would you use this,” say “can I email you when it’s ready?” or “would you put down $5 to reserve a spot?” A yes that costs time, money, or reputation means far more than a verbal one. This is the logic behind a presell before you build, where you ask for real money before writing a line of code.
- Ask for an intro. “Who else deals with this that I should talk to?” A warm intro costs the person social capital, so a real intro means they take the problem seriously.
- Watch behavior, not words. If they say “love it” but won’t give you their email, believe the empty inbox, not the compliment.
When someone gives you a commitment, that’s a signal you can actually trust. When they give you a compliment, thank them and mentally weight it at zero.
The signals to actually trust
Not all answers are equal. Rank everything you hear on this ladder, from ignore-it to trust-it.
| What they say | What it really is | How much to trust it |
|---|---|---|
| ”That’s a great idea” | Politeness | Ignore |
| ”I’d totally use that” | A future promise | Ignore |
| ”This happens to me constantly” | A real, frequent pain | Note it |
| ”Right now I do [messy workaround]“ | Actual behavior | Trust it |
| ”I already pay for [X]” or “Email me when it’s live” | Money or a commitment | Trust it a lot |
You’re mining for the bottom two rows. Anything a person already spends time or money on is a problem they’ve voted for with their behavior. Everything above is a wish. If you run ten honest conversations and nobody has a workaround, nobody’s paid for anything, and nobody can name the last time it hurt, that’s not a failed study. It’s a successful one. You just saved yourself weeks building something nobody needs.
Putting it to work this week
The Mom Test isn’t a personality you have to fake. It’s a filter you run over every question before it leaves your mouth: is this about their past and their life, or my idea and their future? Keep the first kind, delete the second.
Start small. Write down five questions you want to ask, then rewrite any that a polite mom could dodge with a compliment. Book five to ten conversations with people who plausibly have the problem, and if you’re not sure where those people are, here’s where to find people to interview. Take notes in their exact words, because “I spent an hour every Sunday copying grades into a spreadsheet” is evidence, while “user frustrated with process” is a paraphrase that throws the evidence away.
When you’re ready to turn these rules into a full call, pair this with our beginner’s guide to customer interviews for a complete script, and figure out how many customer interviews are actually enough so you don’t interview forever to avoid building.
At batch0, the first of our four one-week sprints is Validate, and this is exactly the muscle we drill: talking to real humans and hearing the truth before you spend a dollar building. If you want a live cohort and mentors pushing you to get honest answers instead of nice ones, take a look at the program. Applying is free, and you only pay tuition if you get in. Ask better questions, and the answers will tell you what to build.