Skip to content

Part I: Coming up with an idea

6. Talking to people

Ask about what actually happens, and use what you learn to check your assumptions.

Talk to the person you named earlier. You have a problem sentence and some assumptions to check. The conversation is a chance to find out where those guesses fit their experience and where they fall apart.

It can feel easier to show someone your idea and ask whether they like it. But a kind response gives you very little to build on. Someone can like your idea without having the problem, wanting a solution, or ever using what you make.

Ask about their experience

Rob Fitzpatrick's The Mom Test is a useful guide to these conversations. Its advice is to focus on the other person's life, ask about things that have happened, and give them room to speak. That helps you learn even when someone wants to be encouraging.

In our hypothetical designer example, “Would you use a tool that makes client feedback easier?” invites a prediction. Try asking about the last project they worked on:

  • How did you ask for feedback on the draft?
  • What happened after you sent it?
  • How did you decide what to change?
  • Was anything holding up the work?
  • What did you do next?

Leave room for the answer to be that everything went well. You want to learn whether your problem description fits their experience.

If something was difficult, follow it. Ask what made it difficult and what the consequences were. If they mention a way around it, ask how that works. A real example can take the conversation somewhere you hadn't thought to ask about.

Ask about things that have happened, and give them room to speak.

Make it easy to talk

A short invitation helps someone understand why you're asking. You could adapt this message:

I'm trying to understand how freelance designers get client feedback and finish projects. Could I ask you about how it worked on a recent project? Would you have 15 minutes for a chat?

Use your own words and describe what you're actually trying to learn. Keep the time you ask for, and be clear if you're taking notes or want to record the conversation.

Ask about their work before showing your proposed solution. Have your important questions nearby, but follow useful details when they come up.

Before you finish, ask whether they know someone else who has dealt with the same situation. That can help you find people beyond your immediate circle.

Watch what you do with the answers

You can accidentally steer someone towards the answer you want. “How frustrating is it when clients don't reply?” assumes both a delay and frustration. “What happened after you sent your request for feedback?” gives them room to describe what actually happened, including a prompt reply.

You can also steer your own notes. If someone likes the idea but says their existing process works well, record both. Don't turn that into a note that says they're interested and leave out the part that challenges your plan.

Give short answers room to develop. You don't need to fill every pause with another explanation of your idea. If you're unsure what someone means, ask them to describe a specific occasion or show how they handled it.

Separate evidence from encouragement

The Mom Test also emphasises commitments as a way to understand customer interest.1 Notice what people have done and what they actually follow through on.

A recent example tells you about the problem. A workaround shows you how they currently cope. Paying for an existing service tells you something about what they value, though it doesn't establish that they'd pay for yours. Agreeing to a trial is a useful next step; taking part in it tells you more.

Keep each piece of evidence attached to the assumption it bears on. Someone introducing you to a colleague helps you find another conversation. It doesn't prove that your product solves their problem.

For the designer, you might discover that clients reply promptly, but their comments conflict. That would change your explanation of the delay. Record it and return to your problem sentence before deciding what to build.

Update your notes, then choose the next step

After each conversation, write down what happened in the examples they described, what surprised you, and which assumptions became more or less convincing. Keep your interpretation separate from their words. An unanswered question can stay unanswered.

Start with the person you named, then talk to others in a similar situation. Look for repeated problems and differences that matter. A useful next step becomes clearer when you can explain who has the difficulty, what it costs them, and what you still need to learn. A conversation count alone can't tell you that.

AI can help you prepare questions or organise your notes. If you ask it to pretend to be a customer and answer your questions, treat those answers as guesses to investigate. They don't tell you what a real person has experienced or will do.

If the evidence changes your mind, change the person, outcome, or explanation in your sentence. You can also set the idea aside. If you have a problem worth pursuing, the next chapter helps you choose a small response to it.

Before you move on0 / 4

Footnotes

  1. Rob Fitzpatrick's teaching guide for The Mom Test identifies recognising biased answers and using commitments to assess interest as core topics. The designer questions and invitation here are examples written for this handbook. ↩