Asking the Right Questions and Conducting Customer Interviews

Discovery happens in interviews, and interviews are only as good as the questions you ask. This tutorial covers the whole arc: how to ask questions that reveal real problems, how to run a 45-minute conversation worth having, and how to turn what you hear into sharp, quantified pain points.

Table of Contents

• Asking the Right Questions
• Open vs. Closed Questions
• The FDE Question Toolkit
• Conducting Customer Interviews
• Preparing and Structuring the Interview
• Active Listening and Probing
• Identifying Pain Points
• Quantifying and Prioritizing Pains

Asking the Right Questions

Asking the right question means choosing a question that makes the customer describe what actually happens, rather than their polished opinion of themselves, their company, or you. People are terrible at predicting their own behavior and excellent at describing past events — great questions exploit that gap.

Rule of thumb: people can't fabricate concrete memories as easily as they can fabricate opinions. Ask about specific past moments, and you'll get truthful, usable answers.

Open vs. Closed Questions

• Closed questions — yes/no or a one-word answer. Fast, good for facts and confirmation.
• Open questions — invite a story. Slow, good for context, reasons, and feelings.

During discovery you want to lean heavily open:

Why the open ones work: they force the customer to reconstruct an actual event. That event — the steps, the tools, the emotions, the time lost — is the raw material of discovery. The closed ones only get you their self-reported verdict, which is worth little.

The FDE Question Toolkit

Most good discovery questions fall into a few repeatable categories. Mix them in any interview:

The "last time" question deserves special attention. It's the strongest probe in the toolkit: "Walk me through the last time you had to assemble that report." Nobody can smoothly fake a detailed account of a recent event, so their response is almost always grounded in reality — warts and workarounds included.

Common Mistake: asking "would you use / like / buy X?" People say yes to be agreeable, then never use it. It's the question that separates junior interviewers from experienced ones — drop it entirely.

Best Practice: write your question guide in advance, but treat it as a safety net, not a script. Before you enter, set one clear goal ("I want to learn how refunds actually flow"). Within the interview, follow what's interesting. The best questions are follow-ups you couldn't have planned.

Conducting Customer Interviews

An interview is a short, structured conversation with one goal: test your assumptions and collect evidence about the customer's problem. It has a shape — prepare, open, explore, close, follow up — and each phase has a job.

Keep it to one person at a time. Group interviews sound efficient but the loudest personality hijacks them, and people won't admit real pain in front of colleagues. A single good interview beats a room of mediocre ones.

Preparing and Structuring the Interview

Preparation should take about fifteen minutes, not three hours:

1. Pick five or six questions from the toolkit, aligned to your goal.
2. Learn the person's role and what they own.
3. State your goal as one sentence ("learn how they process refunds").

Then hold to a rough structure:

Always ask permission to record, and take notes even if they say yes — recordings fail, and hand notes force you to decide what matters on the spot.

Active Listening and Probing

Listening is a skill, and in discovery it's the whole job. Three habits:

1. Follow the last concrete example. When the customer drops a story, go deeper into it before changing topic.
2. Tolerate silence. People fill empty space with their real answers. Count to five before you rescue them with a new question.
3. Probe, don't move on. A typical answer deserves a follow-up before it's done.

Useful probing phrases:

• "What happened right after that?"
• "How long did that take, roughly?"
• "Why is that a problem for you specifically?"
• "Who else feels this?"

Common Mistake: interrupting with your solution. The minute you say "oh, we could fix that with a script," the story stops and the customer turns into an evaluator of your idea instead of a source of truth. Write the idea down on the pad, then go back to listening.

Best Practice: capture verbatim quotes in your notes, marked with quotation marks. "We reconcile these two spreadsheets by hand every Friday" is the sentence your problem statement will later hang from.

Identifying Pain Points

A pain point is a specific, repeated, costly difficulty in the customer's work — the problem your solution will target. Interviews give you the raw material; you extract the pains by listening for signals.

Once the meeting is over, scan your notes for three kinds of evidence:

Anything with a frequency or cost word attached is a candidate. Anything a workaround surrounds (a homemade spreadsheet, a sticky note, a shadow process) is almost certainly a real pain — people don't build workarounds for problems that don't hurt.

Note: a pain is only real once you can attach a cost or frequency. "It's kind of annoying" is a feeling. "It eats two hours every Monday and we've shipped late twice because of it" is a pain point. The second is what you can build against.

Quantifying and Prioritizing Pains

You'll surface more pains than you can solve. Rank them with three factors:

Small example from a discovery session:

The daily, high-severity, multi-team pain wins. It's the one with the clearest business case and the broadest buy-in.

Best Practice: before you leave, play your pains back to the customer. "Let me make sure I've got this right — you're spending two hours every Monday on invoices, and it's usually you who does it?" Confirmation catches misunderstandings at the cheapest possible moment and signals to the customer that you were actually listening.