Review

The Mom Test is great if you’re new to conducting user interviews. It explains the hazards of user interviews, and will help you get the most out of yours. There’s an art to asking questions and this is the playbook. There’s enough examples and sample questions to give you confidence. I first read this book in 2015 and I keep coming back to it!

You Might Also Like…

Book Review and Summary: Just Enough Research by Erika Hall

Get a comprehensive review and summary of the best-selling book "Just Enough Research" by Erika Hall, the research and design expert. Learn about the key principles and practices for conducting effective and efficient research in the design process.

andrewclark.co.uk

Book Review and Summary: Talking to Humans by Giff Constable

Get a comprehensive review and summary of the best-selling book "Talking to Humans" by Giff Constable, the market research and product development expert. Learn about the key principles and practices for conducting effective customer research and using the insights to inform product development.

andrewclark.co.uk

Book Review and Summary: Testing with Humans by Ron Chernow

Get a comprehensive review and summary of the best-selling book "Testing with Humans" by Ron Chernow, the market research and product development expert. Learn about the key principles and practices for conducting effective user testing and using the insights to inform product development.

andrewclark.co.uk

Key Takeaways

The 20% that gave me 80% of the value.

• If you talk to customers before understanding the common pitfalls - you’re going to get misleading information and make bad decisions of the back of it
• The key challenge with user interviews is that you can’t just ask people what they think. People are bad at articulating their rationale. They’re also too nice - which results in false positives
• Every question can create bias and skew findings - so the art of asking the right questions is important. You can’t ask... “Is my business a good idea?”. Nobody is going to tell you the truth. Ask around your idea to get insights
• 3 Rules for Good Interviews:
• Talk about their life not your idea
• Talk about specifics in the past - not generics in the future
• Talk less, and listen more (stop pitching)
• The customer owns the problem - you own the solution. Don’t ask about what you should build - that’s your job not theirs. Instead gather as much information about them as you can.
• 3 Types of bad data:
• Compliments
• Generic Hypotheticals
• Feature Requests
• Ask important questions, escape the trivial. What would make your company fail?
• Don’t zoom in too early, and don’t ignore their indifference. Don’t ask them about the gym if they’re not trying to get/stay healthy
• 2 Types of risk:
• Product Risk - Can I build it? Can I grow it? (prototype)
• Customer and Market Risks - Do they want it? Will they pay for it? (conversations)
• Only once you’ve learnt about the customer - can you reveal your product or idea
• Then look for commitment - Give them a clear chance to reject or commit
• 3 Types of commitment
• Time (trial, training for colleagues)
• Cash (letter of intent)
• Reputation (boss, pilot, introductions)
• Qualification:
• People have the problem
• Know they have the problem
• Have the budget to solve the problem
• Have cobbled together a solution
• Don’t be a learning bottle neck - get the customer data to your team

Deep Summary

Longer form notes, typically condensed, reworded and de-duplicated.

Introduction

• Talking to customers can help reduce the risk that we’ll build something...
• nobody wants
• nobody wants to buy
• nobody want to use
• HOWEVER - Talking to customers is really hard. If you talk to customers before understanding the common pitfalls - you’ll likely make things worse. Doing it wrong is worse than doing nothing.
• We can’t just ask people what they think - people are bad at articulating their rationale
• People are generally nice - which results in false positives
• Every question can create bias and skew findings
• You need to take responsibility for crafting and asking good questions
• Navigate through noise and find out what they want

The Basics

• The social dynamics of sharing an idea aren’t conducive to getting a truthful response
• What is unsaid: I’m about to expose my ego, please don’t hurt my feelings when I share this with you
• So you can’t ask... “Is my business a good idea?”. Nobody is going to tell you the truth, it’s your responsibility to find it
• Example: If you’re researching around the idea of using an iPad for cookbooks?
‣

Bad Questions

Would you ever buy an iPad app for cookbooks?

• share recipes?
• Videos?
• To-Do list?

Response from Mom: It is cheaper than a hardback will it have pictures?

Questions like that are self indulgent noise, will likely result in false positives. Generally opinions aren’t worth much.

‣
Better Questions
• How is the iPad treating you?
• What do you usually do on it?
• What’s the last thing you did on it?
• Did you use an app for that?
• Response: I just use google, I didn’t know there was an app what’s it called?
• Learnt: Won’t discover through the App Store (risk) - needs word of mouth recommendations
• How did you find out about the other apps you have?
• App of the week from the news paper
• Weird question... I saw new cookbooks on the shelf, where did those come from?
• Got them for Christmas, I don’t need another lasagne recipe at my age
• Keep digging...What’s the last cookbook you did buy?
• A vegan one 3 months ago.
• Learning: Niche segments could be key
• Instead ask around your idea to get insights

Three Rules

• Talk about their life not your idea
• Don’t pitch the idea
• Don’t even mention it (or wait until the end)
• Talk about specifics in the past, not generics in the future
• Would you ever y? → When did you last x?
• Talk less, listen more (stop pitching)
• Get into a topic
• Keep them on track
• Shut up and take good notes

Good and Bad Questions

Every customer conversation is bad by default, your job to fix it with good questions.

• None of the good questions are asking about what you should build
• Deciding what to build is your job
• You should humbly and honestly gather as much information about them as you can.
• Take your own visionary leap to a solution (then confirm its correct and refine it)
• They own the problem / you own the solution

Avoiding bad data - How these learning conversations go wrong:

• We want to avoid false negatives and false positives

• Types of Bad Data - and how to avoid them
‣
Compliments - the fools gold of learning conversations
• If you ask for feedback - you are really pitching and asking for compliments
• Deflect compliments - they're probably lying. Even if not, so what, opinions are generally useless.
• Avoid by not revealing idea
• Compliments and stalling or expressing interest is a lethal cocktail
• Notice when you get a compliment
• Save the conversation → quickly get back to reality
• How do you solve the problem now? Talk me through the steps.
‣
Fluff - generic, hypotheticals, opinions about the future
• 3 types of fluff
• Generic claims: I usually, I always, I never
• Future tense promises: I would, I will
• Hypothetical maybes: I might, I could
• Terrible fluff generating questions that result in low value responses
• Don’t ask hypothetical question questions
• Would you ever?
• Do you think you?
• Might you?
• Turn fluffy questions into concrete ones ...
• When did it happen last?
• How did you try to solve it? How did you deal with it?
• Talk me through it? What they did step by step?
• Why didn’t they do it a different way?
• Example: How many emails are in your inbox right now?
• Push them to a decision - Why not try that solution now? If they don't want to - is it that bad?
• Often people complain - you highlight a solution and they ... aww its not a big problem for me
• Be skeptical
‣
Ideas, suggestions, feature requests
Feature requests should be understood, not obeyed
• Reject generic claims, anchor them back to their life and their actions
• Potential Customer: Are you going to add sync to excel?
• Respond with: why do you want to do that?
• If they’ve got a shitty costly workaround (it’s interesting)
• Why do they want the feature? how will it change their lives? Do we need to delay the launch for that? How would that fit into your day?
• Nudge them to elaborate
• Don’t mention the product - but give them a vague space.
• Talk less! Don't interrupt them (even if they've misunderstood or launch into objections)
• Don’t invite the bad data by asking the wrong questions
• Do not expose your ego or your product

Asking important questions

• Don't ask trivial things
• Find the best questions
• What would make the company fail?
• What would have to be true for the company to succeed?
• If I had a free researcher - what would you get them to look at?
• Look for questions and answers that could completely change or disprove your business idea
• If they can’t - they might not be good questions
• Ask unpleasant questions you might be dreading
• Learn to love bad news, it is getting you closer to the truth
• The premature zoom is a big source of false positives
• Don’t zoom in on unimportant problems
• Don't ignore their indifference - Don't ask them about the gym - if they’re not trying to get/stay healthy
• What are you big problems in life?
• if they answer work, job, house, family (thats a red flag for your fitness app)
• Save it by asking ... Is getting healthier on that list?
• You can assume they care only if the need is obvious (you can assume a small business wants more business)
• Risk Types:

Type	Questions	Tool
Product Risk	Can I build it? Can I grow it?	Prototype
Customer and Market Risk	Do they want it? Will they pay me?	Customer Conversations

• You need both. If your product risk is high - make sure you have a prototype
• Prepare a list of 3 big learning goals for each type of person you’ll speak to
• {Customers, investors, industry experts, key hires}
• Update them as you learn more
• Helps you take advantage of bumping into customers
• Decide on the tough questions with your team

Keep it casual

• Separate your meeting types - know what one you are in
• Learning about your customer and their problem (keep casual)
• Learning about how your product performs
• Selling your product to a new prospect
• Avoid premature zoom. could be too much work
• Friendly first contacts are a good place to start
• Learning customer problems is best done in a quick and less formal chat
• Conversations can be fast (5 minutes to find out if there’s a problem)
• Keep it casual
• Meetings have a formality and time overhead
• If you get bad news doesn’t mean you have to give up, just find a new way

Commitment and advancement

• Once you’ve learnt about the customer - you can reveal idea and show the product
• Moving to the next step in sales is advancement
• You can end up with zombie leads that go nowhere, lots of meetings but no progress
• You don’t want to be rejected, so it goes nowhere
• You need to get to a clear point, give them a chance to reject or commit
• You have to pop the question
• Don’t try to convince everyone, try to find the people who get what you’re doing
• If you leave without certainty, you’re probably falling into one of these traps:
• Asking their opinion / fishing for compliments
• Failing to ask for a commitment or next step
• Getting rejected by a customer is not a negative thing, it’s learning
• The currencies of commitment, giving up something they value
• Cash (letter of intent)
• Time (trial, training for colleagues)
• Reputation (testimonial, boss, pilot, introduction)
• Compliments aren’t data - they’re a red flag
• You are looking for:
• People who have the problem
• Know they have the problem
• Have the budget to solve the problem
• Have already cobbled together a makeshift solution

Getting in front of your potential customers

• Face to face better than a call - likely to get more nuance, better conversation and more likely to build a relationship.
• If B2B do as many as possible. You are learning, quickly and generating sales leads.
• If B2C these are less likely to be your customers, how many is enough? Keep talking to people until you hear you’ve heard everything before.
• Or set a goal/cap: x hours a week talking to customers about your product, listening to feature requests / bugs. Those hours are going to save you time

Choosing your customers

• Startups don’t starve they drown in ideas
• Choose a customer segment. If you don’t and you stay too wide you’re going to...
• Get mixed feedback: some love it, some hate it
• Be overwhelmed by options: don’t know where to start
• You can’t prove yourself wrong as there’s always more people to talk to
• If the customer segment is too broad, you’ll struggle to serve everyone
• People only ask to add features - they don’t ask for them to be removed
• The more people you talk to, the harder it gets (a sign segment is too broad)
• You want to have 10 conversations with the same same type of customer. Not 1 conversation with 10 different types of customer
• If this happens - create a taxonomy and segment

• Try and find customers you like

Have the right process

• Do the work before and after your conversations to make them successful
• Don’t be a learning bottle neck - get the customer data to your team
‣
1) Prepare
• Identify your big questions
• Avoid stuff that could be found out with desk research
• Do due diligence on their business.
‣
2) Review
• Review notes
• Update beliefs
• Update questions
• Get raw data out of your head and into something you can share with the team
• Key quotes and raw data.
‣
3) Take good notes
• exact quotes, summaries, add emotions, background
• Meetings go best with 2 people, one talks one takes notes
• Get involved - you can’t outsource customer learning
• You likely won’t believe it, you’ll end up doing it again anyway
• Feature requests, barriers, people, next steps, tasks
• Write it down quick and easy to assess
• Spreadsheets are good, sortable and filterable
• Take notes on a computer, sit at a 45 degree angle, keep eye contact
• Notes are useless if you don’t look at them afterwards
• Staying on Track
‣
Beforehand
• choose a focused findable customer segment, slice thin
• Talk about learning goals with the team
• Decide on next steps for customers
• Who I can talk to, where I can find them
• Try to answer questions other ways first - before scheduling conversations. Desk research, MVP
‣
In the conversation
• Frame it so they know it’s not about sales
• Vision → Framing → Weakness → Pedestal → Ask
• So they know it’s about learning and that they can help you
• Keep the conversation casual
• Ask good questions which pass the mom test
• Every question might not be perfect
• Always try to bring it back
• Deflect the compliments and the fluff
• Dig beneath signals
• Get emotional about anything, will ask you more
• Take good notes
• Give the chance to accept or reject a commitment / next step
‣
After the conversations
• Sit down with team and go through notes, put notes in storage somewhere
• Update beliefs, plan and questions
• New questions
• Plan for next week
• Max 2 meetings, prep on Monday, download on Friday
• Spread learning throughout team as quick as you can

Get Going Quickly

• Spend an hour then go talk to people.
• Don’t spend months only doing customer conversations - you can’t remove all risk and get to 100% security
• Customer conversations are a tool, when appropriate open up big learnings.
• Biggest mistake is to be too formal.