How to run customer interviews that get honest answers

Level

Foundational

Reading time

8 minutes

Updated

November 3, 2026

Sources

2 cited

Useful for

All business types

What a customer discovery interview is

A customer discovery interview is a structured conversation with someone who might become your customer. The goal is to learn about their life: the problem, how often it happens, what it costs them, and how they deal with it today. It isn't a sales call or a focus group for your idea. The U.S. National Science Foundation's I-Corps program, which trains scientists and engineers to explore the commercial potential of their research, is built around the customer discovery process.^[1]

Why most interviews go wrong

Most bad interviews fail the same way. The founder describes the idea, the other person says it sounds great, and the founder leaves encouraged and no wiser. People aren't trying to mislead you. They're being kind, and they're guessing. Ask "Would you use this?" and you're asking someone to predict their own behavior toward something they've never seen. Rob Fitzpatrick's The Mom Test is a whole book about this trap: how to get useful answers from customer conversations when people tend to give polite ones instead of true ones.^[2]

The fix is simple to say and takes practice to do: talk about their past, not your future. What someone did last Tuesday is a fact. What they say they'd do next month is a guess.

Who to talk to

Interview people who closely match your customer description and have dealt with the problem recently, so they remember the details. Don't rely on friends and family. They want you to succeed, which makes honesty hard for them.

Start with about ten interviews. If you're still hearing new problems or workarounds after ten, keep going until the answers start repeating.

How to run the interview

1. 01

   Pick one thing you need to learn

   Before you schedule anything, write down the most important question this round of interviews has to answer. For example: "How do clinic front desks handle double-bookings today, and what does each one cost them?" Everything you ask should serve that question.

2. 02

   Ask for 20 minutes, not feedback

   Reach out with a short, honest request. Say you're researching a problem, not selling anything, and ask for 20 minutes. Something like:

   I'm researching how small clinics handle scheduling problems. I'm not selling anything. Could I ask about your experience for 20 minutes this week?

   People are usually more willing to talk about their own work than to review a stranger's idea.

3. 03

   Open with context, not a pitch

   Explain who you are and why you're asking, in two sentences, and ask permission to take notes. Then go straight to their experience. If you describe your solution now, everything after becomes a reaction to it.

4. 04

   Ask about the last time it happened

   Anchor the conversation in a specific, recent event: "Tell me about the last time this happened." Then follow the story. What were you doing? What went wrong? What did you do next? Who else was involved?

   Specific stories surface details that general questions miss, like a workaround they've never thought of as a workaround.

5. 05

   Find out what it costs and what they've tried

   Ask how often it happens, what it cost in money, time, or stress, what they've tried, and what they dislike about what they use now. If they've never tried to solve it, ask why. That answer is often the most important one.

6. 06

   Listen more than you talk

   Aim to talk for less than a third of the conversation. When an answer is vague, ask for an example. When something surprises you, ask "Why?" Don't rush to fill silences; people often fill them with the most useful detail.

7. 07

   Ask for a small next step

   Near the end, ask for something that costs them a little: an introduction to someone else with the problem, a follow-up conversation, or permission to show them an early version. Whether they say yes, and follow through, tells you more than anything they said about your idea.

8. 08

   Write up your notes within the hour

   Details fade quickly, and it's easy to remember a conversation as more positive than it was. Write down facts, exact quotes, numbers, and commitments the same day, kept separate from your own interpretation. The customer interview guide includes a note-taking structure you can copy.

Questions to ask, and what to ask instead

Most weak questions ask for an opinion or a prediction. Most strong ones ask for a story about something that already happened.

A worked example

Common mistakes

How interviews differ by business type

Turn your notes into patterns

After five to ten interviews, look for what repeats: the same problem, the same workaround, the same cost. AI tools can help you sort notes quickly, as long as you check the summary against what people actually said. Remove names and other personal details before you paste notes into any AI tool.

Your next step

Write down your one learning goal, then use the customer interview guide to book and run your first five conversations.

When a problem keeps repeating, move on to asking for a real commitment, covered in how to validate a business idea before spending money. To understand why people choose the workarounds they do, read about jobs to be done, and to prepare research before your first call, see how to research a business idea with AI.