The Mom Test in Practice
A tactical framework for early-stage founders who need to validate (or invalidate) their bets before burning their runway
Why This Framework Exists
Founders traditionally stink at customer interviews. Not because they’re bad at talking to people, but because they ask the wrong questions and get misleading answers.
When this happens, they build something nobody wants, burn months of runway, and launch to crickets.
This framework teaches you how to have conversations that reveal truth that helps you move your business forward, instead of polite lies that delay, or worse distract.
Based on: Rob Fitzpatrick’s “The Mom Test” + real-world application early-stage founders and their customers
The Core Problem
People lie to you. Not maliciously - they’re trying to be supportive.
When you ask: “Would you use an app that helps you plan your trips?”
They hear: “My friend is excited about something. I should be encouraging.”
They say: “Oh yeah, that sounds great! I’d totally use that.”
What they mean: “I have no idea if I’d use that, but I don’t want to hurt your feelings.”
So, three months later: You’ve built the app. They don’t use it (none-the-less pay for it). You’re confused because they said they would.
They were never lying about the future. They were just wildly optimistic about what they’d do. Everyone is.
Therefore …
The Mom Test Rules
These three rules (and examples) transform bad conversations into useful ones:
1) Talk About THEIR Life, Not Your Idea
Bad: “I’m building an app that helps digital nomads plan trips. What do you think?”
Good: “Tell me about the last time you planned a trip while working remotely. Walk me through what that looked like.”
Why it works: They describe reality (past behavior) instead of imagining hypotheticals (future wishes).
2) Ask About Specifics in the PAST, Not Generics About the Future
Bad: “Would you pay $29/month for a productivity tool?”
Good: “What productivity tools are you currently paying for? How much? When did you last cancel one and why? When did you start a new subscription for one and why?”
Why it works: Past behavior predicts future behavior. Hypothetical statements predict nothing. Asking “why” is to understand their likes, dislikes, needs, wants, and importantly their future desired state.
3) Talk Less, Listen More
Bad: 60% of the conversation is you explaining your product
Good: 80% of the conversation is them talking about their life, problems, and past attempts to solve them
Why it works: The more you talk, the more you bias them. Your job is to learn, not to sell (yet).
Now, here’s how this works:
Track What You Learn
Customer interviews aren’t just about having conversations. They’re about extracting patterns that help you make decisions.
What to track after EVERY interview:
- What problem did they describe? (in their words, not yours)
- How are they solving it now? (current alternative)
- What’s it costing them? (time, money, frustration)
- Have they tried to solve it before? (past attempts = seriousness of problem)
- Would they pay for a solution? (look for behavior, not promises)
The Weekly Review:
After 5-10 interviews, block 30 minutes to look for patterns:
- What problems came up in 80%+ of interviews? (strong signal)
- What solutions did NO ONE mention? (weak signal, deprioritize)
- What did people say they’d pay? What did past behavior suggest they’d ACTUALLY pay?
- Are you talking to the right people? (or do you need to adjust your ICP?)
The goal isn’t to interview forever. The goal is to interview until you see clear patterns, then make a decision: build, pivot, or keep learning.
Warning
Don’t skip the review process. Interviews without synthesis = noise. You’ll “feel” like you’re making progress, but you won’t know what to build next.
1) Before the Interview
Who to Interview
GOOD CANDIDATES:
- People who’ve actively tried to solve this problem in the last 6 months
- People currently using a competitor or alternative solution
- People in your ICP who’ve spent money/time on this problem area
BAD CANDIDATES:
- Friends and family (they’ll lie to protect your ego)
- People who “love all new ideas” (like, cough cough, Product Hunt and the like) (they’ll give false validation)
- Random people who don’t actually have the problem
- People who’ve never thought about this problem before
- Yourself because you “know the space well”, or are “an expert” in it, or “have the problem” yourself. You are one data point. You are not the aggregate of the market.
How to find 10 good candidates:
- Leverage existing communities (Reddit, Facebook groups, Slack channels, Discord)
- Cold outreach to people who’ve posted about the problem in those communities
- Warm intros through your network (“Do you know anyone who…” describe your ICP)
- Partner channels (talk to service providers who serve your ICP) (🔥 tip: this is a great one for many reasons you may tap into later in the journey!)
- Your own network (not friends/family, but weak ties who fit the ICP)
Tip
Notice above we are always wanting to talk to our ICP. That’s key.
This doesn’t mean they’re the people to sell to yet. You’re going to be focused on the problem WAY longer than the solution when you’re interviewing. That means don’t sell, no matter how tempting it may be!
Now, if you’re just starting out, you might know know who your ICP is. That’s ok, the term “ideal” is semantics at that point. Make up a theoretical one and test it out. If you’re truly listening to the interviews you’re doing, they’ll tell you if (and when) they’re the ideal customer. Let them lead this process.
Sample Outreach Messages
For communities (non-salesy):
Hey [name],
I saw your post about [specific problem they mentioned]. I'm researching [problem space] and would love to learn more about how you currently handle this.
Would you be open to a quick 30-minute chat? Happy to share what I'm learning if helpful.
I promise, I'm not selling anything. :)
- [Your name]
For cold outreach:
Hi [name],
I noticed you [specific behavior: posted about X, use tool Y, work in Z space].
I'm doing research on [problem area] and talking to people who [description of ICP]. You seem like you'd have some great insights, and if you're ok sharing them, I'd love to learn more.
Would you be open to a 30-minute conversation? I'm happy to share the results of my research if you're interested in this space.
I promise, I'm not selling anything. :)
Thanks,
[Your name]
For warm intros:
[Mutual connection] suggested I reach out. I'm researching [problem space] and they thought you'd have valuable perspective given your work with [relevant experience].
Would you be open to a brief conversation? I'd love to learn about how you currently approach [problem].
Best,
[Your name]
Key elements:
- Mention specific, relevant detail (shows you did homework)
- Stroke their ego, just a little bit, if you can
- Frame as “research” not “sales”
- Offer value back (sharing insights)
- Keep it short (30 minute max commitment, super short message shows you’re not wasting their time already)
- No mention of your product/solution
- Follow up a few times.
Your Top 3 Questions
Your top 3 questions should be your “company killer assumptions”. This phrase comes from Customer Development, if you’re not familiar with it.
TL;DR: Before every interview, identify the 3 assumptions that, if wrong, kill your business. These are the most important questions to get answers to. Getting info on just one of them is worth gold.
Example:
For a SaaS scheduling tool for small home service professionals:
- Scheduling has to be one of the top 3 problems of small home service professionals. “What is the biggest pain in the butt you’re faced with daily, or regularly?” After they answer, if scheduling isn’t it, ask what the second biggest pain is, and the third. If it’s not in the top 3, they’re not thinking about it enough to make this an important problem to solve - read: they probably won’t pay for your solution.(Problem existence)
- Small home service professionals have to be trying to solve it with a shared Google Calendar, and they hate it with a passion. “How have you solved, or are trying to solve, the scheduling issues you’re experiencing today? Or are you even trying to solve it right now?” And then, why aren’t they trying to solve it right now, if not? This might lead to more important problems than the one you’re focused on. (Problem urgency, competition, problem identification)
- They have to hate Google Calendar enough to switch and pay within a day, even though Google Calendar is free. “Let’s just say I could fix this problem for you (I’m not saying I can, but what if)… what would have to be true for you to switch to it today?” (Switching cost tolerance, objections, price points, value and delight opportunities/features)
Your turn: Write your 3 company killer assumptions here before your next interview.
2) During the Interview
Opening
Set context and build rapport.
Script template:
Thanks for taking the time. As I mentioned in my message, I'm doing research on [problem space] and talking to people like you who [description of why they're relevant].
I have a few questions that should take 30 minutes. I'm genuinely trying to learn about how you currently handle [problem], not pitch you on anything.
Is all that good with you?
(Wait for their response)
Great. Before we dive in, quick background - what do you do and how does [problem space] show up in your work/life?
Why this works:
- Sets expectation (research, not sales)
- Gets permission from them to continue
- Gives them permission to be honest
- Gets them talking about themselves first
Question Templates
Use these as your foundation. Adapt to your specific problem space.
1) The “Walk Me Through” Question
Walk me through the last time you [dealt with problem]. Start to finish - what did that look like?
Example
“Walk me through the last time you planned a work trip abroad. Where did that process start for you?”
Why: Gets concrete, historical behavior, not hypothetical wishes.
2) The “Current State” Question
How are you currently handling [problem]?
Example
“How do you currently track your job applications?”
Why: Reveals existing solutions (your real competition could just be pen and paper, or other online tools, or indifference).
3) The “What Have You Tried” Question
What solutions have you already tried for [problem]?
Example
“What tools or services have you tried for scheduling home service appointments?”
“What did you land on?”
“When was the last time you looked for another solution to this?”
Why: Shows problem urgency. If they haven’t tried anything lately, the problem probably isn’t urgent enough.
4) The “Why Did You Stop” Question
You mentioned you tried [solution]. Why did you stop using it?
Example
“You said you tried Calendly. What made you switch to something else?”
Why: Reveals actual pain points, dealbreakers, what they liked, and usually they’ll complain about the solution they landed on too, which gives you differentiators.
5) The “Money Question”
What are you currently paying for to solve [related problem]?
Example
“What tools do you currently pay for to help you plan your trips?”
Why: Shows willingness to pay and budget allocation.
6) The “Magic Wand” Question
If you could wave a magic wand and fix one thing about [current solution], what would it be?
Example
“If you could wave a magic wand and improve how you find childcare, what would you change?”
Why: Reveals core frustration without them needing to imagine your product.
7) The “Show Me” Question
Can you show me how you currently do [task]?
Example
“Can you show me your current job search spreadsheet?”
Why: Seeing actual behavior beats hearing descriptions. Dig into why they do things that are new to you, or don’t make sense to you. Ask “why” they do what they do often.
8) The “Last Time” Question
When was the last time [problem] caused you a real headache? What happened?
Example
“When was the last time scheduling a service appointment caused you real frustration? Can you tell me more about how that came about and what it cost you?”
Why: Gets emotional, specific stories (where truth lives). Could show value-based pricing opportunity. Opportunity for you to write down “emotional conversion words” to use in marketing, landing page, and email copy later.
9) The “What Would Have To Be True” Question
What would have to be true for you to switch from [current solution] to something new?
Example
“Let’s just say I could fix this problem for you (I’m not saying I can, but what if)… what would have to be true for you to switch to it today”
Why: Reveals switching costs and requirements, objections, price points, value and delight opportunities/features, all without pitching your product.
10) The “Who Else” Question
Who else do you know who deals with [problem]? Would they talk to me?
Example
“Who else do you know who’s hiring a nanny right now? Could you introduce me?”
Why: Gets warm referrals. One “yes” from each person can lead to many more conversations (much easier than doing cold outreach too).
Emotional Conversion Words
When people get emotional, that’s a big signal.
Write down verbatim when they say:
- “I’m so frustrated with…”
- “I hate that…”
- “The worst part is…”
- “I wish someone would just…”
- “It drives me crazy when…”
- “I’d pay anything if…”
- “I’m desperate for…”
Examples:
Weak Signal 👎:
“Yeah, job searching is kind of annoying.”
Strong signal 👍:
“I literally cried last week because I’d sent 200 applications and couldn’t even remember which companies I’d applied to. I felt like such a mess.”
The emotional statement tells you:
- The problem is REAL (they felt pain recently)
- The problem is URGENT (happened last week)
- Current solution SUCKS (manual tracking failed them)
- They’re a BUYER (frustrated enough to pay for solution)
Hot tip
Saving these in text form allows you to bring them back later (probably very soon) and use them in all kinds of marketing copy including landing pages, email copy, direct messages, social posts, and more.
When you use these words in your copy, your prospects and customers will say things like “It’s like you were reading my mind.” No kidding. Don’t be surprised when it happens.
Types of Bad Data
1) Compliments
What they say: “That’s a great idea! I love it.”
What it means: “I want to make you feel good.”
How to redirect:
"Thanks! So to make sure I understand, tell me about the last time you dealt with [problem]. What was that like?"
2) Hypothetical Fluff
What they say: “I would definitely use that!” or “I’d pay for that!”
What it means: “In some imaginary future where everything is perfect, maybe.”
How to redirect:
"Interesting. Have you looked for solutions to this before? What did you find?"
Note
When you’re closer to having a solution, you could ask “how much” and keep a note of it for very loose pricing data. But, don’t put a ton of weight into it until you ask them to pay for the solution - that’s where you’ll find out the truth.
3) Feature Requests (Wishlists)
What they say: “You should add [feature X]!” or “It needs to do [Y] and [Z]!”
What it means: “I’m trying to help you by imagining features.”
How to redirect:
"That's an interesting idea. Help me understand - when's the last time you needed [feature]? What were you trying to accomplish?"
Critical point:
Features suggested in interviews should be noted but not built. You’ll only build what shows up repeatedly AND maps to urgent, past behavior.
Good vs Bad Interviews
BAD INTERVIEW:
You: “I’m building an app for digital nomads to plan their trips. Do you think that’s a good idea?”
Them: “Oh yeah, that sounds really useful! I’d definitely use that.”
You: “What features would you want?”
Them: “Hmm, maybe flight tracking? And packing lists? Oh, and budget tracking would be cool.”
You: “Would you pay $20/month for that?”
Them: “Yeah, probably!”
→ Result: You leave feeling great. You build it. They don’t use it. You’re confused.
GOOD INTERVIEW:
You: “Tell me about the last trip you took while working. Walk me through planning that.”
Them: “Oh man, it was chaos. I had to juggle time zones for meetings, find wifi-friendly places, figure out visa requirements. Took me probably 20 hours of research.”
You: “What did you use to organize all that?”
Them: “Honestly, just a mess of Google Docs, browser tabs, and messages to friends who’d been there.”
You: “Have you looked for tools to help with this?”
Them: “Yeah, I tried Notion and Airtable but they were too complex. I just need something simple that doesn’t require me to build templates.”
You: “What would it need to do for you to actually use it instead of your current system?”
Them: “It would have to auto-populate the basic stuff I always forget - visa requirements, wifi speeds, time zones. If I have to input all that myself, I’ll just stick with Google Docs.”
→ Result: You learn they’ve tried solutions, why they failed, and what actually matters (automation, not more templates).
Interview Logistics
Duration: 15-30 minutes
- First 5 minutes: Build rapport
- Next 10-20 minutes: Questions
- Last 5 minutes: Ask for intro to others
Notes:
- Take notes (less formal than typing)
- OR better, record with permission (“Mind if I record this so I can focus on our conversation?”) so you can stay in the moment and be curious and natural with them
- 🔥 Go back and take notes IMMEDIATELY after the call so you don’t forget
- Note verbatim emotional quotes
- Note surprised looks/reactions and what was being said at the time
Two-person interviews (if possible):
- One person asks questions
- Other person takes detailed notes + watches for signals
- Debrief immediately after and consolidate / edit / combine notes and insights
3) After the Interview
Immediate Debrief
This part is no joke. Don’t wait. Memory fades fast. People fail here often.
Capture:
- Top 3 learnings - What surprised you?
- Emotional quotes - Write the exact phrases they used when emotional
- Next steps - Did they offer intros? Agree to pilot? Ask to buy?
- Contradictions - What didn’t match your assumptions?
Template:
Interview - [Name] [Email] - [Date]
TOP 3 LEARNINGS:
1.
2.
3.
EMOTIONAL QUOTES:
- "[exact quote]"
- "[exact quote]"
NEXT STEPS / COMMITMENTS:
-
CONTRADICTIONS / SURPRISES:
-
Pattern Analysis
Don’t make decisions based on less than 5-10 interviews. I highly recommend 10+.
Questions to ask yourself:
Problem Validation:
- Did the same problem come up in 5+ out of 10 interviews?
- Is it painful enough that they’ve tried to solve it?
- Is it recent (happened in last 3-6 months)?
Solution Validation:
- What are they currently using? (Your real competition)
- Where did you hear about the solution you’re using (Potential watering holes or partners)
- Why did previous solutions fail (and how is the current one falling short)?
- What would it take to make them switch TODAY?
Willingness to Pay:
- What are they currently paying for in this problem space?
- Did anyone ask “When can I buy this?”
- Do they want it even if it’s imperfect? (How desperate are they?)
ICP Validation:
- Is everyone you’re talking to actually the same type of person or can you find patterns in the type of person they are to segment and niche down your ICP even more?
- OR are you getting mixed signals because your segment is too broad?
- Should you narrow or pivot your ICP?
Signals to Persist vs Pivot
STRONG SIGNALS TO PERSIST:
✅ Problem exists and is urgent
- 8/10 people described the same pain
- They’ve tried multiple solutions
- Problem happened in last 30 days
✅ Willingness to pay
- Currently paying for adjacent solutions
- Multiple people asked “When can I buy this?”
- Offered to pilot even if it’s basic
✅ Clear path to them
- You know where they hang out (you can find more like them)
- They introduced you to others like them
- Repeatable channel is emerging
WARNING SIGNALS TO PIVOT:
⚠️ Problem exists but isn’t urgent
- They acknowledge problem but haven’t tried to solve it
- “Yeah, that would be nice to have”
- No current budget allocated to this
⚠️ Solution exists and is “good enough”
- Current solution works fine for them
- Switching cost is higher than pain
- “Maybe I’d try it if it was free”
⚠️ Can’t find consistent ICP
- Interview #1 wants X, Interview #2 wants opposite
- No pattern emerging after 10 conversations
- Everyone’s problem looks different
STRONG SIGNALS TO PIVOT:
❌ No problem urgency
- “I’ve never really thought about this (problem)”
- Haven’t tried any solutions
- Problem hasn’t occurred in 6+ months
❌ Solution not differentiated
- “That’s just like
[existing tool]” - Can’t articulate why they would or what would need to be true to switch
- Existing solution is free/easy/“good enough”
❌ Wrong ICP
- People you’re talking to don’t actually have budget
- OR don’t have authority to buy
- OR aren’t in pain enough to act
Feedback → Action
Not all feedback is created equal. Here’s how to translate what you hear into what you do.
| What They Said | What It Means | What You Do |
|---|---|---|
| ”This solves a real problem I have” + gave recent example | Core problem validated | Keep going, talk to 10 more |
| ”You should add feature X!” | Feature request | Note it. Don’t build yet. |
| ”That’s cool! I love it!” (but no specifics) | Compliment (worthless) | Ignore completely |
| ”When can I buy this?” | Strong buying signal | Get them to pilot ASAP |
| ”I tried 3 other tools but they all suck because…” | Validated pain + competition insight | Understand why others failed |
| ”I’ve never thought about this problem” | No problem urgency | Wrong ICP or non-urgent problem |
| ”I’d use it if it was free” | Not willing to pay | Nice-to-have, not must-have |
| ”That’s exactly like [competitor]“ | Not differentiated | Either pivot or find unique angle by asking what bothers them when using it |
Special Scenarios
B2B Interviews
Critical additional question:
“Where does the money come from?”
Why: The user and the buyer are often different people.
Example:
- Individual contributor uses your tool (USER)
- But manager approves budget (BUYER)
- And IT/Legal controls access (GATEKEEPER)
You must talk to all three:
- User: Does it solve their problem?
- Buyer: Does it fit their budget/priorities?
- Gatekeeper: Does it pass security/compliance?
Part of your job in this case is to figure out what the power dynamic is.
Interview sequence:
- Start with users (validate problem)
- Move to buyers (validate willingness to pay)
- End with gatekeepers (validate procurement process)
Technical Products
This is for plugins, dev tools, APIs, etc.
1. Watch them use their current solution
Screen share or in-person observation is gold.
You’ll see:
- What they actually do vs what they say they do
- Workarounds they’ve created
- Pain points they didn’t mention (because they’re so used to them)
2. Demo AFTER they describe or show their workflow
Don’t lead with your product. Lead with understanding theirs.
Good sequence:
- “Show me your current workflow”
- “What’s the most frustrating part of it?”
- “How are you solving that today?”
- “Okay, based on what you showed me, let me show you something…”
Tip
Obviously, if your solution doesn’t solve their most frustrating part, there’s some things to consider changing in your business or product, or they aren’t your ideal customer. In that case DO NOT DEMO. Just cut this interview short. Everything that’s said after this will be a waste of both your and their time.
3. Ask about switching cost
Some technical products can have high switching costs (learning curve, workflow change, data portability, etc).
Ask: “What would this need to do for you to actually switch from [current tool]?”
4. Understand their buying process
Ask:
- “What’s the last
[category of solution]you bought? Why that one?” - “What’s your typical budget for this?”
- “Walk me through when you bought it from start to finish. How did you become aware of it?” (then walk through step by step to get potential marketing channels, impulsiveness, price sensitivity, potential objections, things that made them buy faster than you would have thought, etc.)
Promoting in Communities
Join communities and give value BEFORE pitching anything at all (including conversations)
You can’t just show up and ask for something.
Process:
- Join groups, provide value (answer questions, share knowledge, earn trust)
- Weeks later, start some research posts: “I’ve heard some people are frustrated with
x, why is that?” (basically spark your ICP’s emotions and get them talking about the problems). - A week or two later, DM people who engaged in the post with one of the sample outreach messages for communities.
Common Mistakes & How to Fix Them
1) Asking Permission to Build
❌ What you’re saying:
“Would you use an app that does X?”
Why it fails:
People are overly optimistic about the future. They’ll say yes to be nice.
✅ Fix:
“Walk me through the last time you dealt with X. What did that look like?”
Now you’re getting: Past behavior (truth) not future wishes (lies).
2) Pitching Instead of Learning
❌ What you’re doing:
Spending 60% of the conversation explaining your product
Why it fails:
The more you talk, the more you bias them. They start imagining your solution instead of describing their reality.
✅ Fix:
Spend 80% listening, 20% asking questions. Only describe your product at the end if they ask/beg.
3) Seeking Validation Instead of Truth
❌ What you want to hear:
“That’s a great idea! I’d definitely use that!”
Why it fails:
You’re looking for confirmation, not information. You’ll unconsciously steer the conversation toward positive responses. DON’T DO THIS!
✅ Fix:
Go in hoping to hear “No, I wouldn’t use that because…”
Bad news is good news. It saves you months of building the wrong thing. And remember, the scientific method only invalidates - that’s it’s purpose. It’s not to validate (despite the words we use in startups).
4) Interviewing the Wrong People
❌ Who you’re talking to:
- Friends and family (they’ll lie to protect you)
- People who don’t have the problem
- People who’ve never tried to solve it
Why it fails:
They have no relevant past behavior to share. Everything they say is hypothetical.
✅ Fix:
Only interview people who:
- Have the problem RIGHT NOW
- Have tried to solve it in last 6 months
- Fit your ICP profile
5) Taking Feature Requests as Gospel
❌ What you hear:
“You should add [feature X]!” → You go build it
Why it fails:
They’re being helpful consultants, not revealing their real behavior. Most feature requests won’t move the needle unless they surprise and delight, or are table stakes you’re missing. You will only know this when a person uses strong emotional conviction and conversion words.
✅ Fix:
Ask: “Why is that important to you? When’s the last time you needed that?”
If they can’t give you a recent, specific example, note it but don’t build it.
6) Stopping After 2-3 Interviews
❌ What you do:
Talk to 2 people, both say it’s a great idea, you go build.
Why it fails:
You need patterns, not individual opinions. 2-3 interviews can be outliers.
✅ Fix:
Minimum 10 interviews before making major decisions.
Pattern shows up in 8+ of 10 conversations → That’s strong signal.
Pattern shows up in 5-7 of 10 conversations → That’s signal.
Pattern shows up in 2-3 of 10 → That’s noise.
7) Not Getting Commitments
❌ How interviews end:
“Thanks for your time! This was really helpful.”
Why it fails:
You learned something, but the relationship stops there.
✅ Fix:
Always ask for a commitment:
Soft commitment:
“Who else do you know dealing with this? Could you introduce me?”
Medium commitment:
“Would you be open to trying a rough prototype when I have something?” (ONLY ask this to your ICP who has budget and urgency)
Hard commitment:
“Based on what you shared, I think I could solve [specific problem]. Will you pay today for a pilot?” (“Will”, not “would”.. big difference in this one word.)
Resources
Original source:
“The Mom Test” by Rob Fitzpatrick
Related FounderLabs frameworks:
- Customer Development → How to structure discovery overall
- Your First Funnel → What metrics to track post-interview
- Persist or Pivot Framework → How to decide when to change course
Additional reading:
- “Talking to Humans” by Giff Constable (shorter, more tactical)
- Jobs to be Done framework (why people “hire” products)
Common Questions
Q: How many interviews before I can make a decision?
A: Minimum 10. Patterns show up in 5+ conversations. Less than that, you’re guessing.
Q: What if I can’t find people to interview?
A: If you can’t find people to TALK to, you won’t find people to BUY from. Distribution channel IS your company killer assumption. Focus on test that now, not later.
Q: Should I show them my product during the interview?
A: Not unless they ask. Your job is to learn about THEIR life, not pitch YOUR product. Save the demo for after you’ve learned about their behavior.
Q: What if they say my competitor is “good enough”?
A: That’s useful data. Ask: “What would have to change for you to switch?” If the answer is “nothing really,” you have a differentiation or importance problem.
Q: Can I do this async (email surveys, forms)?
A: Not at this level of depth (and it’s 20x slower). You lose all the nuance, emotion, and ability to dig deeper. The conversation IS the data collection method. 30 minutes on Zoom beats 100 survey responses.
Q: What if people keep asking for features I don’t want to build?
A: That’s fine. Note them. But only build what shows up in 7+ interviews AND maps to urgent, recent behavior. Most requests are noise.
Q: How do I handle introverts or shy interviewees?
A: Ask about specific past events, not hypotheticals. “Walk me through…” gets even shy people talking. They’re experts on their own experience.
Q: Should I pay people for interviews?
A: No. Offering to share research findings is enough incentive. People who take money for interviews are not incentivized to solve the problem you’re asking about. They’re incentivized to tell you whatever will get them the money.
Framework Checklist
Print this out. Check it off as you go.
Before Interview:
- Identified 3 company killer assumptions
- Found 10 potential interviewees who fit ICP
- Sent outreach messages (not salesy)
- Scheduled interview
- Prepared top 3 core questions
During Interview:
- Set context (research, not sales)
- Asked about their life, not my idea
- Asked about past actions/history, not future hypotheticals
- Listened more than I talked
- Captured emotional quotes verbatim
- Redirected compliments and hypotheticals
- Asked “who else?” for warm intros
After Interview:
- Immediate debrief (top 3 learnings, etc)
- Documented emotional quotes in a place I will use later
- Noted next steps/commitments
- Identified contradictions/surprises/learnings
- Thanked interviewee for time and warm intros (in advance)
After 10 Interviews:
- Analyzed patterns (what showed up 5-7+ times?)
- Validated problem urgency
- Validated willingness to pay
- Validated distribution channel
- Made persist/pivot decision
Remember:
The Mom Test isn’t about being nice or mean. It’s about asking questions that reveal truth instead of polite lies.
Good customer interviews feel like conversations, not interrogations. You’re genuinely curious about their life. They leave feeling heard. You leave with data that changes your business.
Now go talk to 10 people.