How do you teach customer discovery in one class session?

Do not explain it first. Have students interview each other about a real problem using their own instincts, then show them what they did wrong using their own notes. A 75-minute structure works: ten minutes framing, fifteen minutes of unguided paired interviews, fifteen minutes of debrief where the failures surface, ten minutes teaching the technique, fifteen minutes of a second round, and ten minutes of closing debrief. The learning happens in the gap between round one and round two.

What is the most common mistake students make in customer interviews?

They pitch instead of listen. Within about ninety seconds, most students stop asking and start explaining their idea, then interpret politeness as validation. The second most common is asking about the future, as in would you use this, rather than about the past, as in what did you do last time this happened. Future answers are free to give and predict nothing.

Do students need a real business idea for a customer discovery exercise?

No, and it works better without one. Use a universal problem the whole room has experienced, such as finding parking on campus, getting a group assignment scheduled, or choosing what to eat. Removing the idea removes the defensiveness, so students can see their interviewing behaviour instead of protecting their concept.

How long should a classroom customer discovery exercise be?

Seventy-five minutes is comfortable and fits most class slots. It compresses to fifty by shortening both interview rounds to ten minutes and the final debrief to five. Below fifty minutes you have to cut the second round, which removes the part where students apply the correction, and the exercise loses most of its value.

Customer Discovery in One Class Session: A Classroom Exercise

John Richards · November 2nd, 2026

The most useful hour in any entrepreneurship course I have taught is the one where students discover, on themselves and in front of other people, that they cannot interview a customer.

Not that they do not know how. That they actively do the wrong thing, within about ninety seconds, while believing they are doing the right one.

You cannot teach this by explaining it. Every student nods along to “ask about the past, not the future” and then, given a live human, immediately asks “so would you use something like this?” The gap between knowing and doing is the whole lesson, and the only way to close it is to let them fall into it while someone is watching.

Here is the exercise. It needs no preparation, no materials beyond paper, and no student business ideas.

Setup

Time: 75 minutes Group size: any, works from 8 to 200 Materials: paper and pen, no laptops Prerequisite: none, and it is better early in the term

Pick a problem the entire room has personally experienced. Good options: finding parking on campus, scheduling a group project, deciding where to eat with friends who will not decide, doing laundry in a shared building.

Deliberately not their own startup ideas. The idea creates defensiveness, and defensive students cannot observe their own behaviour. Remove the idea and they can see the technique.

Minutes 0 to 10: framing

Tell them this:

“You are going to interview a partner about how they handle [problem]. Your job is to understand what they actually do today. You have seven minutes. Go.”

That is the entire briefing. Do not teach technique yet. Resist every instinct to help them, because the failure is the content.

Minutes 10 to 25: round one, unguided

Pairs. Seven minutes each direction, with a hard switch. Interviewer takes notes, interviewee just answers.

Walk the room. Do not correct anyone. Listen for the moment each pair turns, because it will happen in nearly every pair, and note a few examples verbatim for the debrief.

What you will hear, reliably:

The interviewer explaining their own experience of the problem within two minutes
“Would you use an app that…” before minute three
Questions that contain the answer: “It is annoying when the lot is full, right?”
Long stretches where the interviewer talks more than the interviewee
Almost no silence

Minutes 25 to 40: the debrief that does the work

This is the heart of it. Ask these in order, and let silence sit after each one.

“Who talked more, you or your partner?”

Hands go up slowly. In a typical room, half the interviewers talked more than their interviewee. Let that land before saying anything.

“How many of you described your own experience of the problem?”

Most hands.

“How many asked whether they would use something?”

Most hands again. Then: “What did they say?” The answer is always some version of yes, maybe, sure. Then ask: “What did that tell you?” and wait. Someone will eventually say “nothing,” and that is the sentence you want to come from a student rather than from you.

“Read me one thing your partner said that surprised you.”

Usually a long pause. In round one, most students have nothing, because they were not listening for surprise, they were listening for agreement.

“Read me one specific thing they do today.”

Some have this. Point out, without ceremony, that the pairs who have a specific answer here are the pairs who asked about the past.

Minutes 40 to 50: the technique, now that they want it

Now teach it, and it takes ten minutes because they are listening differently.

Four rules:

Ask about the past, not the future. “What did you do the last time this happened?” not “would you use.” Past behaviour is a fact. Future intention is a courtesy.
Ask what it costs them. Time, money, embarrassment, a workaround. A problem with no cost is not a problem.
Never pitch. The moment you describe a solution, you have converted an interviewee into a critic and lost the information.
Be quiet. Count to five after they stop talking. The second thing people say is more honest than the first.

That is enough. Resist adding more.

Minutes 50 to 65: round two

New partners, same problem, seven minutes each direction.

The difference is audible from the front of the room. It is quieter, and the interviewees talk more. Students notice this themselves without being told, which is why the second round cannot be cut.

Minutes 65 to 75: closing debrief

Three questions:

“What was different?” They will say it was harder, slower, and more uncomfortable. Agree with all three.

“What did you learn about your partner’s actual behaviour that you did not get in round one?” Now the specifics come out: the workaround, the real cost, the thing they already pay for.

“Who heard something that would make a product idea worse?” A few hands. Close on those: that is the whole job. The purpose of customer discovery is not to find out you are right. It is to find out you are wrong while it is still cheap.

Why it works

Students arrive believing that interviewing is a soft skill they already have, because they talk to people all day. The exercise disproves it in fifteen minutes using their own notes as evidence, and the correction lands because they asked for it.

I have run this with undergraduates, MBAs, and founders in their forties who had already raised money. The failure rate in round one is roughly the same for all three groups, which is usually worth mentioning afterward.

Variations

Large lectures: run the same structure, debrief by asking for hands rather than individual answers, and take three verbatim examples from your walk of the room.

Online: breakout rooms work, but shorten to five minutes each direction because the switch costs more. Keep both rounds.

Second-year students: use the same exercise but have them interview someone outside the class between sessions, then debrief the following week. The discomfort of a stranger is the next thing to build.

Try the ToolSuite free with your class Customer discovery and business model canvas tools built for coursework. Free to start, with academic pricing for programs. Check out our software free →

Related: a customer discovery rubric that survives AI for grading the assignment this exercise prepares students for, and an intro to entrepreneurship syllabus showing where it fits in a term.

The customer discovery tools in Startup Ignition ToolSuite give students somewhere to log interviews with dates and sources once they move from classmates to real customers, and give you visibility into the work between class sessions. Academic pricing is available for accredited programs.

Back to All News