Sign inGet started
Skills / Product-management

The Mom Test

It's called The Mom Test because it leads to questions that even your mom can't lie to you about.

The Mom Test is Rob Fitzpatrick's prescriptive methodology for running customer interviews that surface truth instead of validation. Most founders unknowingly design interviews that invite polite agreement rather than honest feedback — The Mom Test provides a concrete ruleset for fixing that. By focusing on past behaviors instead of hypothetical opinions, and commitments instead of compliments, you learn what customers actually do rather than what they wish they might do.

By Rob Fitzpatrick · Free
Specimen 01 · Live diagnosisThe Mom Test
Input

“I think restaurant owners spend too much time manually tracking inventory and end up over-ordering or running out of popular items during service.”

Diagnosis
Let's build your Mom Test interview guide.
Full transcript ↓
Calibrated referenceagent-skills.ai
The gap

Ask questions so good even your mom can't lie to you

The Mom Test is built on a foundational distinction: opinions about the future are worthless, but past behaviors reveal truth. Fitzpatrick's Three Core Rules — talk about their life instead of your idea; ask about specific past behaviors, not hypothetical futures; talk less and listen more — provide a concrete filter for every question you design. The methodology distinguishes compliments (empty noise that feels like validation) from commitments (time, money, or reputation staked on the problem being real), and provides a five-part conversation structure for framing discovery sessions. Fitzpatrick also introduces a note-taking shorthand organized around facts, emotions, and commitments, and establishes a test for whether a conversation was valuable: did it end with an advancement or a next step?

The problem

Most founder customer interviews are fatally flawed: they ask hypothetical questions ('Would you use this?', 'Do you think this is a good idea?') that allow polite people — including your own mother — to validate bad ideas out of social kindness. The result is a pipeline of false positives: customers who seemed enthusiastic in interviews but won't pay when the product ships. Founders mistake compliments for signal and spend months building for a market that doesn't exist.

The solution

The Mom Test skill stress-tests every interview question against Fitzpatrick's rules before you walk into the room, so you come out with facts about what customers actually do — not opinions about what they wish they might do.

You bring
  • Your product hypothesis or the problem you believe exists
  • Draft interview questions you've written (optional, for auditing)
  • A transcript or notes from a past customer conversation (optional)
  • The specific customer segment you're targeting
You get
  • A complete Mom Test-compliant interview guide tailored to your specific hypothesis
  • A red/green audit of your draft questions with concrete rewrites for each violation
  • A commitment-tracking framework to evaluate real signal across multiple conversations
  • A diagnostic of where past interviews went wrong and what to ask instead
In action

Watch the methodology work.

Three specimens from a single real session: the same situation, unaided and calibrated, the full transcript, and the skill answering live in the channel where the work happens.

Fig.01The same situation, two ways
Specimens A · B
Specimen AVague
Unaided

“You've written ten interview questions: 'Would you use a tool that did X?', 'How much would you pay for this?', 'Do you think inventory management is painful?' You run five customer meetings, everyone says 'that sounds really interesting, keep me posted' — and you walk out convinced you've validated the idea.”

×Vague — nothing concrete to act on.
Specimen BCalibrated
With The Mom Test

“Every question in your guide is a behavioral probe: 'Walk me through the last time that happened.' After five interviews, you know exactly how three owners currently handle inventory, which one abandoned a spreadsheet system after a costly failure, and that two are willing to be paid pilot partners. You know which pain is real and which was politeness.”

Specific, grounded, and ends with one clear next step.
Fig.02A real session, unedited
In-app · assistant
The Mom TestClaude
The Mom Test
Let's build your Mom Test interview guide. Before I generate any questions, I need your hypothesis — not your solution, just the problem you believe exists. Describe the pain in one or two sentences from the customer's perspective: who experiences it, when does it happen, and what does it currently cost them?
I think restaurant owners spend too much time manually tracking inventory and end up over-ordering or running out of popular items during service.
Good hypothesis. Here's a quick Mom Test diagnostic on how most founders approach this: they walk in and ask 'How painful is inventory management for you?' or 'Would you use software that tracked inventory automatically?' Both questions are traps — they invite the customer to comment on your framing rather than describe their reality.
Instead, we want to reconstruct what actually happened the last time they faced this problem. Here are the first three questions I'd put in your guide:
1.
"Walk me through what happened the last time you ran out of something important mid-service."2. "How do you currently track what you have in stock? Take me through what that looks like on a typical Tuesday."3. "What have you already tried to make this easier? How did that go?"
Notice: no mention of your idea, no hypotheticals, no requests for opinions. Just their life. What customer segment are you targeting — fast casual, fine dining, something else?
Fast casual, 1-3 locations. But what if they say inventory isn't actually that painful? Like if they just say 'oh it's fine, we have a spreadsheet'?
'It's fine, we have a spreadsheet' is actually excellent data — if you probe it correctly. Most founders hear 'it's fine' and either move on or get defensive. Fitzpatrick flags this exact moment: the question isn't whether they have a system, it's what that system actually costs them.
Follow-up questions for the spreadsheet response:- "Who maintains that spreadsheet? How long does it take them each week?"- "When did the spreadsheet last fail you? What happened?"- "Have you ever had a situation where the spreadsheet said you had something and you didn't?"
If they answer with shrugs and 'not really,' your hypothesis may be wrong for this segment — and that's valuable information. What you're looking for is emotional intensity: the owner who sighs when you ask about the spreadsheet, or who leans forward and says 'Oh god, let me tell you about last Friday.' That's signal. Flat responses are data too — just different data.
That makes sense. What about the end of the conversation — how do I know if the interview was actually good?
Fitzpatrick's test is direct: did the conversation end with an advancement? A real advancement requires the customer to commit one of three currencies — their time, their money, or their reputation.
- Time: 'Can we do a follow-up call so you can meet my co-founder and see an early prototype?'- Money: 'Would you be willing to pay for early access and shape what we build?'- Reputation: 'Could you introduce me to two or three other owners who deal with this?'
If the conversation ended with 'This sounds cool, let me know when it launches' — that's a compliment. It felt good but it's not signal. It costs nothing to say. The closing question I'd put in your guide: 'Based on what we've talked about, would you be willing to be one of our first pilot restaurants? We'd work directly with you and your feedback would shape the product.' Watch what they do next. That's your answer.
Perfect. I want to run 10 interviews this week. Can you give me the full guide?
Here's your complete Mom Test interview guide for fast casual restaurant owners (1-3 locations):
Opening Frame (say this at the start — you're not pitching, you're learning)'I'm not here to sell anything — I'm trying to understand how restaurants like yours actually handle the back-of-house side of things. Would you mind walking me through how it currently works?'
Core Behavioral Questions1. Walk me through a typical week — when do you think about what you have in stock?2. How do you currently track inventory? Take me through what that actually looks like.3. Walk me through the last time you ran out of something important mid-service. What happened?4. Have you ever over-ordered and been stuck with product you had to throw out? Tell me about that.5. What have you tried to make the back-of-house run more smoothly? How did that go?
Intensity Probes (use when someone sighs, leans in, or shows frustration)- Tell me more about that.- How often does that happen?- What did that cost you — in time, money, stress?- What made you try that approach?
Closing Sequence'Who else do you think I should talk to — other owners dealing with this kind of challenge?' [reputation commitment]'Based on what we've discussed, if we built something that addressed this, would you be interested in being an early pilot partner and giving us direct feedback?' [time/money commitment]
After each interview, log: What facts did you learn about their current behavior? Where did you see emotional intensity? Did you get a commitment of time, money, or reputation — or just a compliment?
Reply to The Mom Test
What the skill is doing
01Interview Question Auditor
02Mom Test Interview Guide Generator
03Commitment Signal Evaluator
Fig.03The same skill, in the channel you already use
WhatsApp · live
9:41
TMThe Mom Testtyping…
Messages are end-to-end encrypted. No one outside of this chat can read them.
Today
I think restaurant owners spend too much time manually tracking inventory and end up over-ordering or running out of popular items during service.9:14✓✓
Let's build your Mom Test interview guide.9:14
Before I generate any questions, I need your hypothesis — not your solution, just the problem you believe exists.9:15
Describe the pain in one or two sentences from the customer's perspective: who experiences it, when does it happen, and what does…9:15
Message

The same skill, where the work happens.

No new app to learn. The methodology runs over the WhatsApp Business API, so the answer lands as a reply in the thread you’re already in — same rigour, zero context-switch.

Reads the situation, names the pattern, returns one concrete next move.
Delivered in seconds, inside a conversation that already exists.
Specimen · WhatsApp Business API · live
Capabilities

What it does, specifically.

Each capability is a distinct move drawn straight from the source methodology — not a generic assistant guessing.

CapabilityC-01

Interview Question Auditor

Submit your draft interview questions and receive a line-by-line audit against The Mom Test's Three Core Rules. Each question is flagged as compliant or violating, with a specific explanation of what rule it breaks and a rewritten version you can use instead.

Based on Fitzpatrick's good/bad question taxonomy: opinion-seeking questions ('Would you...', 'Do you think...', 'How much would you pay for...') are rewritten as behavioral, past-tense probes ('Walk me through the last time...', 'How do you currently handle...', 'What have you already tried?').
CapabilityC-02

Mom Test Interview Guide Generator

Provide your hypothesis and target customer segment, and receive a complete, structured interview guide: conversation framing language, sequenced behavioral questions, probes for emotional intensity signals, and a closing sequence that solicits commitment advancement and referrals.

Follows Fitzpatrick's five-part conversation structure — vision framing, hypothesis framing, weakness demonstration, pedestal building (invite them to teach you about their life), and referral request — producing a guide that stays in learning mode throughout and never tips into pitching.
CapabilityC-03

Commitment Signal Evaluator

After a customer conversation, describe what happened and receive an assessment of whether you got real signal. The evaluator distinguishes the three commitment currencies from compliments and fluff, and tells you whether the meeting constituted a genuine advancement.

Based on Fitzpatrick's commitment framework: only three things constitute real signal — the customer commits time ('Let's schedule a follow-up with my team'), money ('I'll pre-pay for early access'), or reputation ('I'll introduce you to three colleagues'). Everything else, including 'This sounds great, keep me posted,' is noise.
CapabilityC-04

Fluff vs. Signal Roleplay

Practice your interview in roleplay mode, where the skill plays a realistic customer who gives plausible but often empty responses. Learn to identify genuine pain signals — emotional intensity, unprompted returns to a topic, sighs — versus polite noise that feels like validation but predicts nothing.

Trains the pattern recognition Fitzpatrick describes: watching for leaning-in moments, emotional language, and topics customers return to unprompted (high-signal), versus the 'fluff' of generic positivity ('Oh yeah, that sounds interesting') that sounds encouraging but carries zero predictive weight.
CapabilityC-05

Transcript Debrief & Diagnosis

Paste a transcript or notes from a real customer conversation and receive a structured diagnostic: where did you slip into opinion-fishing? Where did a customer signal real pain and you moved past it? What follow-up probes did you miss? What did you actually learn vs. what felt like learning?

Applies Fitzpatrick's evaluation criteria for a valuable conversation: Did the founder speak less than 50% of the time? Did high-intensity moments get probed with 'Walk me through the last time...'? Did the conversation end with an advancement (time, money, or reputation committed) or a compliment?
Tested

Graded before it shipped.

Every skill is scored against independent scenarios for methodology fidelity before it goes live — not vibes, a rubric.

What it produces
OutputD-01

Mom Test Interview Script

A complete, hypothesis-specific interview guide with framing language, sequenced behavioral questions, emotional intensity probes, and a closing sequence that solicits commitment and referrals. Ready to use in your next customer conversation.

OutputD-02

Question Audit Report

A red/green analysis of your draft questions — each flagged as Mom Test-compliant or violating, with the specific rule being broken and a concrete rewritten version you can use instead.

OutputD-03

Commitment Signal Log

A structured record of commitments gathered across multiple customer conversations, categorized by currency type (time / money / reputation) — making the real signal in your discovery campaign visible and comparable at a glance.

OutputD-04

Behavioral Fact Sheet

A structured notes template using Fitzpatrick's facts/emotions/commitments shorthand, capturing what customers actually do (not what they say they'd do), how they feel about it, and what they're willing to stake on the problem being real.

The source

Grounded in the original work.

Every answer traces back to a real source and the practitioner who wrote it — not a secondhand summary. Here is the source of record.

Source authorA-01

Rob Fitzpatrick

Rob Fitzpatrick is a serial entrepreneur and startup educator whose customer development framework has become foundational curriculum at Y Combinator, startup accelerators, and entrepreneurship programs worldwide. He developed The Mom Test after observing that most founders conducted interviews specifically designed — if unintentionally — to confirm their assumptions rather than challenge them. The book is now widely regarded as the definitive guide to non-biased customer conversations.

Status · Inspired by Rob Fitzpatrick’s work — not yet claimed. Are you Rob Fitzpatrick?
Primary sourceS-01

The Mom Test: How to Talk to Customers & Learn If Your Business Is a Good Idea When Everyone is Lying to You

by Rob Fitzpatrick

Serial entrepreneur; author of The Mom Test; customer development curriculum used at Y Combinator and leading global accelerators.

Read the original ↗
Citationmomtestbook.com
Get started

Put it to work.

agent-skills runs as MCP — connect it once and The Mom Test works in Claude, ChatGPT, Cursor, or any MCP-compatible client. Sign in to see your config and accept the terms.