Book

The Mom Test

Rob Fitzpatrick

Awesome. Recommended to me by Gabe Ragland (Founder of DivJoy). This book gives you clear and practical advice that will immediately improve the quality of feedback from customer development interviews. You will not be able to lie to yourself if you follow it.

The major lessons for me were:

People stop lying when you ask them for money
Don’t ask customers hypotheticals, keep the conversation grounded on their actual workflow to uncover interesting talking points.

Example: “How often do you train?” = Bad; “When did you last train?” = Good. “What kind of workouts do you do?” = Bad; “What was your last workout?” = Good.
People exaggerate pain points, see if they’ve actively tried to solve it.

Example: “Not having a rest timer really bugs me!” - ask them what they use to time their rest, if nothing, or some other solution, like a watch or the timer on their phone, unlikely they’ll switch apps just because you have a rest timer.
If you think you’re on to something, don’t take “yeah sounds great, I’ll try it when it’s out” as a positive. Make them commit one of time/money/reputation to gauge if they’re genuinely interested.

Points

Asking anyone if your business “is a good idea” is a bad idea and invites lies.
People may put in feature requests to appear engaged.
False positives are super dangerous, they cause you to invest time/money/reputation in the wrong direction.
Instead of asking a question like “What type of training do you do?” Be specific and ask “What was the last workout you did? When? Where did you get it from? How did you record your results?”
In general: specific questions > general questions.
How are people finding solutions to the problem you solve? Some people might use Google, some might use App Store searches. This may vary by stratum (e.g. geography, age) and is essential info for finding reliable marketing channels.
A useful conversation gives you concrete facts about a customers’ life and world views.
Without concrete facts, you make generalisations. Generalisations make it hard to answer difficult trade-offs that come up during product development.
You can get all this information without ever mentioning your product. Mentioning your product too soon is a mistake.
“Who else should I talk to?” Is a great question because it opens up new leads for you.
If they claim something is annoying but haven’t looked for any solutions to fix it, chances are, it doesn’t move the needle for them.
A feature request may actually obfuscate the actual problem. Take someone requesting messaging in spreadsheets. They want the feature so they can make sure they’re on the right version. The real problem here is versioning strategies.
Learn from people’s actions, rather than their opinions.
Watching someone do a task shows you where the problems/inefficiencies really are, not where the customer thinks they are.
Learn where the budget comes from (more relevant in B2B).
“Is there anything else I should have asked”?
Customers own the problem, you own the solution.
People are wildly optimistic about what they would do in the future.
People describe themselves as who they want to be, not who they actually are. This ties into the marketing line: “People don’t buy a set of features, they buy a better version of themselves.”
In all discovery calls, anchor them on the life they already lead and the actions they’re already taking.
If you get a feature request from a user, take a moment to understand the motivations behind it. Ideas and feature requests should be understood, not obeyed.
When a interviewee says something like “So it’s similar to x…” and is off the mark, instead of correcting them, dig further into the specifics of what they think is similar. It gives you a glimpse of how they perceive.
You can identify important questions by assessing how significantly the answers change your business assumptions. You should be terrified of at least one of the questions you’re asking in every conversation.
There are certain types of question where the answer is obvious and doesn’t reveal useful information:

Example: “My marketplace connects you with ‘thing’ so you can get more business and save on agent’s fees”. More business in most circumstances is always good, so an interviewee responding positively isn’t valuable. In these types of situations the risk is in kickstarting the marketplace, not the end result.

The above lets us conclude that there are two types of risk: Product risk - can you build it, can you grow it?; Customer risk - Do they want it? Will they pay for it? Are the lots of them? Customer interviews are only concerned with the second type.
As far as possible keep these conversations casual. It speeds up the flow. If you’re starting the conversation with something like “Thanks for this meeting”, it’s too formal!
When you do have a product to demo, you need to rapidly work out if someone is a real or zombie lead. You can get avoid zombie leads by giving users a clear chance to commit or reject.
Symptoms of rubbish pitches: no useful information (compliments are not useful), no clear next steps, the other side has given up nothing of value [time, money, reputation].
Pitches either succeed or fail. They don’t just ‘go well’. It’s not a real lead until you’ve given them a concrete chance to reject you.
If you start too generic, everything is watered down. Your marketing message is generic. You suffer feature creep. By serving too many segments you run into three problems: You get overwhelmed by options and don’t know where to start; You aren’t moving forward but can’t prove yourself wrong; You receive mixed feedback and can’t make sense of it.
When serving segments that are too generic, you end up with feature bloat because a feature advocate can always make a reasonable case by saying “Well, those guys would love it.” You can’t prove yourself right or wrong, so end up being held hostage by competing interest groups over middling features.
How finely can you slice your customer segment? Strava took it really to the bone, people who could afford a bike over £2k, that cycled over 100km/week, who had a specific Garmin cycling computer. A slice like “Finance professionals in the city aged 20-35” doesn’t cut it.
Your segment should be a subset of people you enjoy being around. in an industry where you enjoy being around. If you don’t, or you are cynical, it can be a real grind getting to know them.
Make sure you write up your conversations.