0 to 1 Product Development
0 of 9
Workbook
Module 2 · Steps 3 and 5 · about 90 min

Talk to them: story-based interviews

By the end of this moduleThree research questions, five interview questions, a list of eight people to interview, and interview snapshots for the ones you have done.

Primary market research is talking to and observing your target customers. Nothing replaces it.

Reports tell you what an industry thinks about itself. Interviews tell you what one person did last Tuesday. The second is worth more, because products are used on Tuesdays.

You are looking for pattern recognition, not one great quote. After 5 to 8 interviews with the same persona a pattern shows up. Before that, every interview feels like a contradiction of the last one. That is normal. Keep going.

Research questions are not interview questions

A research question is what you want to learn: when, where and why do people binge-watch? An interview question is what you ask to get a story that answers it: tell me about the last time you watched a series. Write the research questions first, then design questions that elicit stories, never questions that ask for opinions.

The tell of a good interview question is the word last. Tell me about the last time you looked for a coworking space. Not: is location important to you when choosing a coworking space? The first gets a story with a place, a time and a decision in it. The second gets a shrug and a yes.

Rules I do not break

  • Ask about the past, not the future. Last time, not next time. People predict badly and remember well.
  • Never pitch. The moment you describe your product, every answer becomes politeness.
  • Never ask would you use this. The answer is yes and it means nothing.
  • Anchor and dig: how much time do you spend on it each week? What are you already doing to fix it? Which tools do you pay for? Tell me more about that.
  • Twenty to thirty minutes. Record it, or bring a note-taker. Write down direct quotes, observations and your interpretations in separate columns. They are not the same thing.
  • Recruit where they already are. B2B: LinkedIn. B2C: Facebook groups, Reddit threads, the places they complain.

The interview snapshot

One page per interview, written within an hour of the call. Three things: what they said, verbatim; what you observed, including what they did not say; what you think it means. Keep the third column honest. Most bad product decisions are interpretations promoted to quotes.

After eight snapshots, lay them side by side. The pattern you are looking for is a job, a struggle, and a workaround they already pay for in time or money. That is the opportunity. If there is no workaround, there is usually no problem.

Worked example

What churned users told elyps

The analytics said 80% of new users churned on one screen, the one right before identity verification. Industry reports on neobank onboarding said the ID scanner was the pain point. Simplify the scanner, they said.

We interviewed churned users instead. Same story, interview after interview: I was on the subway and was not comfortable pulling out my ID. I was at work and did not want my boss seeing me do personal stuff. The problem was the moment, not the step. No report could have told us that, because no report was on the subway.

Exercise · 60 min to plan, then 5 to 8 interviews over two weeks

Interview plan and snapshots

Plan first, then go. Come back and fill a snapshot after every interview. The worksheet holds eight.

Saved in this browser
What I want to learn
1
2
3
Tell me about the last time you...
1
2
3
4
5
Name or roleWhere I found themStatus
1
2
3
4
5
6
7
8
WhoDirect quoteObservationMy interpretation
1
2
3
4
5
6
7
8
The job, the struggle, the workaround they already pay for. If you cannot write this yet, do three more interviews.
Feeds your pitch deck: Problem slide
Checkpoint

Tick only what is true. Nobody is grading this. The next module assumes it.