Customer Interview Questions: 60+ Examples for Discovery, Validation, Pricing, and Churn

The 5 principles behind every good customer interview question

Before the list, the rules. Most customer interview questions fail for the same five reasons, no matter how much time you spent writing them. Get these right and almost any specific question works.

1. Anchor in the past, not the future. "Tell me about the last time…" beats "Would you ever…" Past behavior is real. Future intent is wishful thinking — people overestimate what they will do, especially when being polite to a researcher.
2. Ask about their life, not your idea. "How do you currently handle X?" beats "What do you think of our new feature?" The Mom Test rule. As soon as you mention your product, the participant turns into a polite reviewer.
3. Dig for specifics. When someone says "I always" or "I never," ask for the last specific instance. Generalizations are summaries; the truth lives in the example.
4. Listen for the cost. Time, money, emotional energy. A problem without a cost is not a problem worth solving. Always probe: "How much does this take?" "What is the impact when this happens?"
5. Stay quiet after asking. The most powerful follow-up is silence. Most people will keep talking after their first answer if you let them. The second sentence is usually where the gold is.

Koji's AI interviewer applies these principles automatically based on the methodology framework you pick. If you select Mom Test, it refuses to ask leading future-tense questions. If you select Jobs to be Done, it probes the switching trigger. If you select Customer Discovery, it digs into the problem cost. Either way — human or AI — the principles are the same.

Discovery interview questions (problem exploration)

Use these when you're trying to understand whether a problem exists, how often it happens, and what people currently do about it.

1. Tell me about the last time you had to [task or situation]. Walk me through it.
2. How often does that happen for you?
3. What was going on in your life or your work that day?
4. What did you do about it?
5. How did that work out?
6. What was the hardest part?
7. How much time did that take?
8. Who else was involved?
9. What tools or workarounds do you currently use?
10. What don't you love about how you handle it today?
11. Have you tried other things in the past? What happened?
12. If you had a magic wand, what would be different?

Customer development questions (Steve Blank-style)

Use these when validating a problem and the customer segment that has it.

13. Walk me through how you currently solve [problem].
14. How important is solving this on a scale of "minor annoyance" to "I cannot run my business without fixing it"?
15. What's the cost of leaving this unsolved — in time, money, or frustration?
16. Who else on your team is affected by this?
17. If you could fix only one thing about this, what would it be?
18. Have you ever looked for tools to help with this? What did you find?
19. Why did you stop using the last tool you tried?
20. What's the trigger that makes you start looking for a better way?

Jobs to be Done questions (switching and progress)

Use these to understand the "job" customers hire your product for, and the push-pull dynamics that drive switching.

21. Take me back to when you first started thinking about this.
22. What was the trigger that made you say "I need to do something"?
23. What were you trying to accomplish?
24. What solutions did you consider? Why did you rule them out?
25. What almost stopped you from making a change?
26. What did you have to give up to make this switch?
27. How is your work or life different now?
28. If you switched away tomorrow, what would push you to do that?

Pricing and willingness-to-pay questions

Use these to understand price sensitivity and the value frame customers bring to the decision.

29. How is the budget for this kind of tool currently allocated?
30. Last time you bought something like this, walk me through how the decision happened.
31. At what price would this be so cheap you'd question the quality?
32. At what price would it start to feel expensive but you'd still consider it?
33. At what price would it be too expensive to seriously consider?
34. What do you currently spend on solving this problem — in tools, services, or time?
35. Who has to approve a purchase like this?
36. What would convince you it's worth twice the price?

For a full Van Westendorp pricing study with these questions structured automatically, see the willingness-to-pay template below.

Onboarding and activation questions

Use these with users who recently signed up to understand what helps or blocks their first wins.

37. Tell me about the moment you decided to try [product].
38. What did you expect to happen when you signed up?
39. Walk me through your first 10 minutes after signing up.
40. Where did you get stuck, even briefly?
41. When did you feel like "okay, I get this"?
42. What almost made you give up?
43. If a friend was about to sign up, what would you tell them to do first?

NPS detractor follow-up questions

Use these with customers who gave you a low NPS score. The follow-up is where the actionable insight lives, not the score itself.

44. You gave us a [score]. What's the one thing that drove that the most?
45. What were you hoping the product would do that it didn't?
46. When did you first feel disappointed?
47. Tell me about the last time the product let you down — what happened?
48. What would have to change for you to give us a 9 or 10?
49. Have you started looking at alternatives? What are you looking at?
50. If we fixed [the thing they mentioned], would that be enough?

Koji teams typically set this up as an automated follow-up: when an NPS detractor submits their score, they get a Koji interview link in the follow-up email. The AI handles the 5-10 minute conversation and surfaces themes in real time.

Churn and cancel-flow questions

Use these with customers who churned or are about to. The goal is the regret moment — when did they first feel they'd outgrown the product or stopped trusting it?

51. Tell me about when you first started thinking about leaving.
52. What changed?
53. Was there a specific moment that made you decide?
54. What did the product fail to do for you?
55. What are you using instead now?
56. What's better about the new solution?
57. If we built [the thing you wanted], would you consider coming back?
58. What advice would you give us about the next customer like you?

Concept testing questions (handle with care)

Use these only when you genuinely want feedback on a specific concept — and ground every answer in past behavior to avoid polite-lie validation.

59. Here is the rough idea. Without telling me whether you like it, walk me through how this would fit into your current workflow.
60. Where in your current process would this slot in?
61. What would you have to stop doing or change to use this?
62. Looking at your current alternatives, what would make you choose this instead?
63. What's the first thing that gives you pause?
64. Who else on your team would need to be sold on this?

Notice none of these ask "do you like it?" The Mom Test rule says people will lie politely. Instead, you map the concept to their real workflow and see whether it fits.

How Koji turns these questions into a real interview

Writing 60 questions is one thing. Running them as actual conversations with real participants is where most teams stop. Koji handles the rest:

1. Pick the questions that match your goal — or describe the goal and Koji's AI Consultant suggests the right ones.
2. Mix in structured questions — NPS scale, multiple choice, ranking — to capture quantitative data alongside the open-ended exploration. Koji supports six question types in a single interview.
3. Send the link — async, voice or text, in any language. Participants take it when convenient.
4. Watch the dashboard — themes, quotes, sentiment, and quality scores update in real time as interviews complete.
5. Generate the report in 30 seconds when you're ready to share.

A traditional moderator running these questions would need 30 minutes per call plus 2-3 hours of analysis per interview. With Koji's AI moderator, 20 interviews finish overnight and you're looking at themes Monday morning.

Anti-patterns to avoid

• "Would you use…?" People say yes to be polite.
• "What do you think of…?" Opinions are unreliable. Behavior is reliable.
• "Why?" Too open-ended. Replace with "What happened next?" or "Tell me about that."
• Leading questions. "Don't you think it would be better if…?" guarantees a yes.
• Multiple questions stacked together. One question at a time. Then silence.

A 12-question script that works for almost any discovery study

If you need to start tomorrow and don't have time to read the full list, this script — adapted from the Mom Test — works for most discovery research:

1. Tell me about your role and what you spend most of your time on.
2. Walk me through the last time you had to [task].
3. What was hard about that?
4. How often does that come up?
5. What did you try before to make this easier?
6. Why did that not stick?
7. How are you handling it now?
8. What would have to be true for you to change?
9. Who else feels this problem on your team?
10. What would it be worth to you if this was solved?
11. Is there anything I didn't ask that I should have?
12. Can I follow up if I have more questions?

Drop those 12 into Koji, pick the Mom Test methodology, and you have a study ready to send within minutes.

Related Resources

• Structured Questions Guide — combine open-ended questions with NPS, scale, ranking, and choice questions
• Mom Test User Interviews — methodology behind questions that surface truth
• Jobs to be Done Framework — the switching-trigger interview style
• Probing and Follow-Up Questions — how to dig deeper without leading
• Open-Ended Interview Questions — the qualitative core of every good interview
• AI Interview Question Generator — generate study-specific questions automatically
• Discussion Guide Template — turn questions into a runnable interview script
• Churned Customer Interviews — full churn-interview methodology