How to Run Customer Discovery Interviews

You can read every forum and review in your market and still be guessing about one thing: whether a specific person will actually change their behavior and pay you. Customer discovery interviews answer that. Ten honest conversations will teach you more than a hundred page market report, but only if you run them in a way that gets the truth instead of a polite lie.

Talk about their life, not your idea

The biggest mistake is pitching. You get excited, describe your product, and ask "would you use this?" People say yes to be nice, you walk away thrilled, and the feedback was worthless. Compliments are not commitments.

So do not talk about your idea at all in the early interviews. Talk about the person's life and the problem. Ask what they did, when, and what happened. Past behavior is fact. Opinions about the future are guesses. "Would you buy a tool that does X" is a guess. "Tell me about the last time you dealt with X" is a fact you can learn from.

Ask about the last time it happened

The most useful question in discovery is some version of "walk me through the last time you faced this." It forces specifics:

• When did it happen and what triggered it
• What did you do step by step
• What did it cost you in time, money, or stress
• What did you try before that, and why did it fail
• How did you feel about the whole thing

Specific stories are reliable. Generalizations like "I usually handle it fine" hide the truth. If they cannot remember a recent time it happened, that is a strong sign the problem is not painful enough to build a business on.

Listen far more than you talk

A good interview is the other person talking most of the time. Your job is to ask a short question and then get out of the way. Silence is your tool. When someone pauses, do not rush to fill it. Wait, and they will often add the most honest part.

A few habits help:

• Ask open questions that cannot be answered with yes or no
• Follow up with "why" and "tell me more about that"
• Never finish their sentence or suggest the answer
• Resist defending your idea, because you are here to learn, not to win

Watch for real signals of demand

Words are cheap. Look for signals that cost something to give:

• They have already paid for a partial solution
• They have built a workaround that took real effort
• They get visibly frustrated when describing the problem
• They ask you, unprompted, whether something exists to fix it
• They offer to be a first user or to pay before you even ask

These cost the person credibility, money, or effort, so they are far more trustworthy than a friendly "that sounds useful." If nobody shows any of these signals across ten interviews, the problem is probably not urgent, and that is a finding worth having.

Recruit the right people

Interview people who actually have the problem, not your friends and not anyone willing to chat. The fastest way to find them is to go where they already gather: the forums, communities, and review threads where they complain about the problem. Reach out, say you are researching the problem and not selling anything, and ask for fifteen minutes.

Aim for ten to fifteen conversations with people who fit your target. That is usually enough to hear the same themes repeat, which is the moment you know you have a real pattern and not a sample of one.

Take notes you can compare

During each call, write down quotes word for word, especially the emotional ones and the descriptions of current workarounds. After the call, jot the three things that surprised you. Surprises are where your assumptions were wrong, and that is the whole point of doing this.

After all the interviews, lay the notes side by side and look for:

• Problems that came up again and again
• The same words used by different people
• The workarounds everyone shares
• The moments that triggered the most frustration

Those patterns become your positioning and your offer, built on what real people told you rather than what you hoped was true.

Customer discovery is not a sales call and not a survey. It is a structured way to learn what people actually do and feel about a problem. Run it honestly, listen more than you speak, and let the truth change your plan. That is the entire value of doing it.

If you want a running start before the calls, a DemandSonar scan mines the real demand in your market and surfaces the recurring problems and the language people use, which gives you sharper questions and a list of the communities where you can find the right people to interview.