Good customer interview questions ask people to describe something specific they did recently. They don’t ask people to predict what they would do or to rate your idea. “Tell me about the last time you…” almost always beats “Would you use…?”, because a story about real past behavior is much harder to bend than an opinion. Below are 38 customer interview questions grouped by goal: discovery and problem validation, buying and switching, onboarding, product usage, churn risk and B2B buying. Each one comes with what it reveals and a follow-up probe.
Ask customers to walk you through one specific, recent time they did the thing you’re researching, then follow up with “What happened next?” and “Why then?”. Pick questions by goal, avoid hypotheticals and leading wording, and run live calls for open discovery or async recorded answers for focused questions at scale.
- A story about real past behavior (“Tell me about the last time you…”) is better evidence than a prediction (“Would you use…?”).
- The first answer is usually the tidy version, and the follow-up probe is where the real insight is.
- Live calls suit exploratory discovery and complex or sensitive accounts, while async recorded answers suit focused questions, breadth and customers across time zones.
- A systematic review found saturation within 9 to 17 interviews for fairly similar groups and narrow objectives, so count interviews per segment, not in total.
- Interviews pay off only when answers become a decision: get them into text, code themes, count by segment and write the decision down.
The guide also covers how to run the interviews (live call or async recorded answers), how many to run, how to avoid leading questions and how to turn the answers into decisions. If you’re still choosing a method, our user research methods guide shows where interviews fit next to surveys, usability tests and analytics.
Five rules that make customer interview questions work
- Ask about one specific, recent time. Nielsen Norman Group recommends asking about specific incidents rather than how people “usually” work, because the “usual” answer tends to describe an idealized process that skips the real shortcuts (NN/g, “Interviewing Users”).
- Your research question is not your interview question. Teresa Torres makes the point that people are bad at answering direct questions out of context. Instead of asking “What do you want from a reporting tool?”, collect the story of the last report they built and listen for the answer inside it (Product Talk).
- Don’t pitch. The moment you describe your solution, people start being polite about it. Rob Fitzpatrick’s The Mom Test is built around this problem: customers will tell you your idea is good even when they’d never pay for it.
- One question at a time. “How did you find us and what did you think of onboarding?” gets half an answer. Split it.
- The follow-up is where the insight is. The first answer is usually the tidy version. “What happened next?” and “Why then?” get you the real one.
Discovery and problem validation questions
Use these before you build, or when you’re testing whether a problem is real, frequent and painful enough that people will change what they do.
Replace [the task] with the job you’re researching, such as “collecting customer feedback” or “onboarding a new client.”
“Tell me about the last time you had to [do the task]. Walk me through what happened, step by step.”
RevealsThe real workflow, the tools and people involved, and where it breaks.
Follow-up“What happened right before you started?”
“What was the hardest part of that?”
RevealsWhich pain is sharp and which is mild. If they have to think hard to name one, the problem may not be worth solving.
Follow-up“What did that cost you, in time, money or something else?”
“How are you handling this today?”
RevealsYour real competition, which is often a spreadsheet, a shared inbox, an intern or doing nothing.
Follow-up“What works about doing it that way?”
“How often does this come up?”
RevealsFrequency. A painful problem that happens once a year rarely justifies a new tool or habit.
Follow-up“When was the time before last?”
“Have you tried to fix this? What did you try?”
RevealsWhether they care enough to act. People who have searched, built a workaround or paid for something have a real problem.
Follow-up“Why didn’t that stick?”
“Who else is affected when this goes wrong?”
RevealsOther stakeholders, who owns the budget and how far the problem spreads.
Follow-up“What do they do about it?”
“If this were fully solved, what would be different for you?”
RevealsThe outcome they want, in their own words. That’s useful for messaging, but treat it as language, not as proof of demand.
Follow-up“How would you know it was solved?”
I'm preparing customer interviews about my product. Here are 7 questions in the "Discovery and problem validation questions" group: 1. “Tell me about the last time you had to [do the task]. Walk me through what happened, step by step.” 2. “What was the hardest part of that?” 3. “How are you handling this today?” 4. “How often does this come up?” 5. “Have you tried to fix this? What did you try?” 6. “Who else is affected when this goes wrong?” 7. “If this were fully solved, what would be different for you?” Adapt them to my product and the customers I want to interview: [describe them]. Keep each question open and neutral, and add one follow-up question for each.
Buying journey and switching questions
Ask these of customers who signed up in the last few months, while the decision is still fresh. They’re based on the Jobs to Be Done “switch” interview, which reconstructs the timeline of a real purchase to find four forces: the push of the current situation, the pull of the new option, the anxiety about switching and the habit of the old way (Jobs-to-be-Done Radio, Bob Moesta and Chris Spiek).
“Take me back to when you first started looking for something like this. What was going on?”
RevealsThe push, the trigger event that made the status quo unacceptable.
Follow-up“Why then, and not six months earlier?”
“What else did you look at, including keeping things as they were?”
RevealsThe real consideration set, which is often different from the competitors you track.
Follow-up“What ruled each one out?”
“What almost stopped you from signing up?”
RevealsAnxiety: price, setup effort, fear of picking the wrong tool, a past bad experience.
Follow-up“What made you go ahead anyway?”
“Where did you first hear about us, and what did you do next?”
RevealsThe research path, including which pages, reviews or people they trusted.
Follow-up“What did you read or watch before you decided?”
“What were you using before, and what was it like to stop?”
RevealsHabit and switching cost, including data they had to move and people they had to retrain.
Follow-up“Is any part of the old way still running?”
“When you signed up, what did you expect it to do for you?”
RevealsThe pull, the promise they bought. Compare it with what your marketing actually says.
Follow-up“Has it done that so far?”
“Who else had a say in the decision?”
RevealsHidden influencers, even in small teams.
Follow-up“Did anyone push back? What was their concern?”
I'm preparing customer interviews about my product. Here are 7 questions in the "Buying journey and switching questions" group: 1. “Take me back to when you first started looking for something like this. What was going on?” 2. “What else did you look at, including keeping things as they were?” 3. “What almost stopped you from signing up?” 4. “Where did you first hear about us, and what did you do next?” 5. “What were you using before, and what was it like to stop?” 6. “When you signed up, what did you expect it to do for you?” 7. “Who else had a say in the decision?” Adapt them to my product and the customers I want to interview: [describe them]. Keep each question open and neutral, and add one follow-up question for each.
Onboarding and first value questions
Ask these in the first two to six weeks after signup. Wait longer and people forget the friction they worked around.
“Walk me through your first day with the product. What did you do first?”
RevealsThe path people actually take, which is rarely the one you designed.
Follow-up“Where did you have to stop and look something up?”
“When did you first feel it was working for you? What had just happened?”
RevealsThe first value moment, which is what onboarding should get people to faster.
Follow-up“How long after signing up was that?”
“Was there a point where you thought about giving up?”
RevealsThe drop-off points your analytics show as numbers but can’t explain.
Follow-up“What kept you going?”
“What did you need to set up, import or ask someone for before it was useful?”
RevealsHidden dependencies such as IT approval, data exports or a colleague’s admin access.
Follow-up“How long did that take?”
“Was anything different from what you expected after signing up?”
RevealsGaps between what you promise and what people find.
Follow-up“Where did that expectation come from?”
“If a colleague started using it tomorrow, what would you tell them?”
RevealsTheir mental model of the product and the words they use for it.
Follow-up“What would you warn them about?”
I'm preparing customer interviews about my product. Here are 6 questions in the "Onboarding and first value questions" group: 1. “Walk me through your first day with the product. What did you do first?” 2. “When did you first feel it was working for you? What had just happened?” 3. “Was there a point where you thought about giving up?” 4. “What did you need to set up, import or ask someone for before it was useful?” 5. “Was anything different from what you expected after signing up?” 6. “If a colleague started using it tomorrow, what would you tell them?” Adapt them to my product and the customers I want to interview: [describe them]. Keep each question open and neutral, and add one follow-up question for each.
Product usage and gap questions
Use these with active customers to find what to build, fix or remove.
Feature requests will come up. Treat each one as a clue to an underlying problem, not as a spec.
“Tell me about the last time you used the product. What were you trying to get done?”
RevealsThe job behind the usage, which is often bigger than the feature they opened.
Follow-up“What did you do right after?”
“Which parts do you use every week, and which have you never opened?”
RevealsYour real core product, and features that may be invisible or unneeded.
Follow-up“Why haven’t you needed that one?”
“What do you do outside the product to finish the job?”
RevealsWorkarounds, exports, copy-pasting and integration gaps.
Follow-up“Can you show me?” In a recorded interview, a screen recording answer works well here.
“Tell me about a time the product didn’t do what you needed.”
RevealsGaps with their context attached, which is far more useful than a bare feature request.
Follow-up“What did you do instead?”
“If you could change one thing, what would it be, and what would that let you do?”
RevealsThe second half of the question is the valuable part: the outcome behind the request.
Follow-up“When did you last run into that?”
“Who else on your team uses it, and how is their use different from yours?”
RevealsRoles, expansion paths and where a second user gets stuck.
Follow-up“Is there anyone who stopped using it?”
“What would you do if the product disappeared tomorrow?”
RevealsHow dependent they are and what they see as the alternative.
Follow-up“How big a problem would that be for your week?”
I'm preparing customer interviews about my product. Here are 7 questions in the "Product usage and gap questions" group: 1. “Tell me about the last time you used the product. What were you trying to get done?” 2. “Which parts do you use every week, and which have you never opened?” 3. “What do you do outside the product to finish the job?” 4. “Tell me about a time the product didn't do what you needed.” 5. “If you could change one thing, what would it be, and what would that let you do?” 6. “Who else on your team uses it, and how is their use different from yours?” 7. “What would you do if the product disappeared tomorrow?” Adapt them to my product and the customers I want to interview: [describe them]. Keep each question open and neutral, and add one follow-up question for each.
Churn risk and renewal questions
Ask the first four of active accounts ahead of renewal, and the last two of customers who have recently canceled.
An exit survey checkbox tells you the stated reason. A conversation tells you the sequence of events.
“How has the way you use the product changed over the last few months?”
RevealsDeclining or narrowing usage before it shows up as a cancellation.
Follow-up“What caused that change?”
“Has anything changed in your team or priorities since you started?”
RevealsOrganizational risk: the champion left, the budget moved, a new tool was mandated.
Follow-up“Who owns this decision now?”
“What results can you point to from using it?”
RevealsWhether they can justify the spend internally. If they can’t name anything, renewal is at risk.
Follow-up“Who have you shared that with?”
“What would make you reconsider renewing?”
RevealsThe conditions under which they’d leave, stated while there’s still time to act.
Follow-up“Has any of that started to happen?”
For churned customers: “Take me back to when you first thought about canceling. What was happening?”
RevealsThe real trigger, which usually came weeks before the cancellation itself.
Follow-up“What did you switch to, if anything?”
For churned customers: “Looking back, was there a point where it could still have worked out?”
RevealsYour save window and what would have been needed in it.
Follow-up“What would have had to happen at that point?”
I'm preparing customer interviews about my product. Here are 6 questions in the "Churn risk and renewal questions" group: 1. “How has the way you use the product changed over the last few months?” 2. “Has anything changed in your team or priorities since you started?” 3. “What results can you point to from using it?” 4. “What would make you reconsider renewing?” 5. For churned customers: “Take me back to when you first thought about canceling. What was happening?” 6. For churned customers: “Looking back, was there a point where it could still have worked out?” Adapt them to my product and the customers I want to interview: [describe them]. Keep each question open and neutral, and add one follow-up question for each.
B2B buyer interview questions
In B2B, the person using the product often isn’t the person who approves it. These questions map the buying process. Ask them of recent buyers, lost deals and prospects in your target segment.
“Walk me through the last time your team bought a tool like this.”
RevealsThe real process, approvals and timeline, not the official one.
Follow-up“Where did it slow down?”
“Who has to sign off, and what does each of them care about?”
RevealsThe economic buyer plus security, legal, IT or procurement concerns.
Follow-up“What did they ask that was hard to answer?”
“How did you justify the cost internally?”
RevealsThe business case in the buyer’s own words, which is the language your sales materials should use.
Follow-up“What number or result did they ask for?”
“What did you need from vendors during the evaluation that was hard to get?”
RevealsFriction in your sales process: missing documentation, slow answers, no trial data.
Follow-up“How did you get it in the end?”
“What made this a risky decision for you personally?”
RevealsThe champion’s personal stakes, which drive more B2B decisions than feature lists.
Follow-up“What reduced that risk for you?”
I'm preparing customer interviews about my product. Here are 5 questions in the "B2B buyer interview questions" group: 1. “Walk me through the last time your team bought a tool like this.” 2. “Who has to sign off, and what does each of them care about?” 3. “How did you justify the cost internally?” 4. “What did you need from vendors during the evaluation that was hard to get?” 5. “What made this a risky decision for you personally?” Adapt them to my product and the customers I want to interview: [describe them]. Keep each question open and neutral, and add one follow-up question for each.
A sample six-question async customer interview
If you're sending questions for people to answer on their own time, keep it short and build the follow-up into the question text, since nobody is there to ask it. This set works for a general voice-of-customer round with active customers:
| # | Question | Suggested answer length |
|---|---|---|
| 1 | Tell us about the last time you used the product. What were you trying to get done, and how did it go? | 2 to 3 minutes |
| 2 | What do you do outside the product to finish that job? Show us if you can. | 2 minutes |
| 3 | Tell us about a time it didn’t do what you needed. What did you do instead? | 2 minutes |
| 4 | Has the way you use it changed over the last few months? What caused that? | 90 seconds |
| 5 | If you could change one thing, what would it be, and what would that let you do? | 90 seconds |
| 6 | Is there anything we didn’t ask that we should know? | 1 minute, optional |
For more question formats built for recorded answers, see our video survey questions.
Live or async: how to run customer interviews
A live interview is a scheduled call where you ask questions and probe in real time. An async interview sends the same questions as a link, and each person records their answers when it suits them. Both are real interviews. They’re good at different things.
| Live call | Async recorded answers | |
|---|---|---|
| Best for | Exploratory discovery when you don’t yet know what to ask | Focused questions once you know what you’re testing |
| Follow-ups | Unlimited, in the moment | Written into the question, or sent afterward |
| Scheduling | One calendar slot per person | None; people answer when they can |
| Time zones | Someone takes an early or late call | No overlap needed |
| Scale | A handful per week, per interviewer | Dozens in the same week |
| Interviewer effect | Varies by who asks and how | Everyone gets the same wording |
When async works better
- Your customers are spread across time zones and a 45-minute slot that works for both sides is hard to find.
- You need breadth. Hearing from 30 customers about onboarding is a week of calls, or one link sent on Monday.
- B2B buyers are busy. A director who won’t book a call will often record three short answers between meetings.
- You want answers close to the moment, such as right after onboarding, right after a support case or right after cancellation.
- You want the same questions in the same words for everyone, so differences in answers come from customers, not from the interviewer.
When a live call is better
- Deep exploratory discovery, where the most valuable question is one you’ll only think of halfway through the conversation.
- Complex B2B accounts, where you need to map several stakeholders and follow each thread.
- Sensitive conversations with a key account that’s unhappy or at risk.
You can also combine the two: an async round to hear from many customers quickly, then live calls with the five or six people whose answers raised the most interesting questions.
In RecRam, you send your questions as a video form. Respondents answer by video, voice or text from one link in their browser, with no app or account needed. Every video and voice answer is transcribed automatically and summarized, so you can read the batch instead of watching every recording.
You can add conditional logic to branch, for example showing churn questions only to customers who canceled. If you want to follow up with a specific person afterward, Room keeps a one-to-one async video thread going. For a broader look at the format, see our guide to video surveys.
How many customer interviews to run
The research on qualitative sample sizes is more reassuring than most teams expect. A systematic review of empirical studies found that saturation, the point where new interviews stop surfacing new themes, was reached within 9 to 17 interviews, particularly for fairly similar groups of people and narrowly defined objectives (Hennink and Kaiser, Social Science & Medicine, 2022). Broader questions and more varied groups needed more.
interviews reached saturation in empirical studies, particularly for fairly similar groups and narrowly defined objectives
Source: Hennink and Kaiser, 2022
In practice, that gives you a working rule:
- Count per segment, not in total. Ten interviews split across agencies, enterprise teams and solo founders is about three per segment, which is too few for any of them.
- Keep the objective narrow. “Why do new trial users stall in week one?” saturates faster than “What do customers think of us?”
- Stop when the last few interviews add nothing new, not when you hit a number. Keep a running list of themes and note when each first appeared.
- Keep a steady cadence. Torres recommends weekly touchpoints with customers as a core product discovery habit (Product Talk). A few interviews every week beats a big study once a year.
How to avoid leading questions
A leading question puts the answer you want into the question. NN/g warns that people can form an opinion about almost anything when asked, which is why answers to loaded questions are poor evidence (NN/g).
Some common rewrites:
| Leading | Neutral |
|---|---|
| “How much time does the new dashboard save you?” | “Tell me about the last time you used the dashboard.” |
| “Would you use a feature that exports to Excel?” | “What did you do the last time you needed this data somewhere else?” |
| “Isn’t manual reporting frustrating?” | “How do you put your reports together today?” |
| “What problems did you have during onboarding?” | “Walk me through your first week with the product.” |
| “Would you pay $50 a month for this?” | “What do you spend on this problem today, in money or time?” |
| “Why do you love using it?” | “What do you use it for most?” |
A few more habits help:
- Don’t name a feature they haven’t mentioned. Let them bring it up.
- Get comfortable with silence in live calls. People fill a pause with the detail you need.
- Mirror their words in follow-ups (“You said it felt slow. Slow how?”) instead of reframing in yours.
- Pilot async questions on two colleagues before sending. Fixed wording can’t be rescued mid-interview, so a confusing question stays confusing for every respondent.
How to analyze customer interview answers
Interviews only pay off if the answers turn into a decision. A simple process that works for small teams:
- Get every answer into text. Transcripts are searchable, quotable and quick to skim. Notes from memory are not.
- Read a full batch before you tag anything. Reading interview by interview pulls you toward whatever the first customer said.
- Code the answers. Tag each relevant passage with a short label (trigger, workaround, blocker, outcome wanted), then group labels into themes. This is a light version of thematic analysis, a widely used qualitative method (Braun and Clarke, 2006).
- Count by segment. Note how many customers raised each theme and which segment they belong to. “Seven of nine agency customers export to spreadsheets” is a finding. “Some people mentioned exports” is not.
- Separate stories from opinions. Weight what people describe doing above what they say they would do.
- Pull quotes with context. Two or three verbatim quotes per theme make the case to your team better than a summary.
- Write the decision down. One page: themes, counts, quotes and what you’ll do next.
Our guide to customer feedback analysis goes deeper on coding and prioritizing, and we’re publishing a dedicated guide on how to analyze video survey responses soon.
If you collect answers in RecRam, each video and voice response arrives with a time-stamped transcript (spoken language detected automatically, with transcription in 110+ languages), an AI summary, a subject line, 3 to 5 keywords and a sentiment label. That gets you through steps 1 and 2 faster. AI rules can also tag or flag answers that mention something specific, such as a competitor, using AI credits. You can push responses to Google Sheets, Slack, HubSpot or Zendesk through the integrations.
Treat summaries as a starting point. Read the full transcript for anything you’re going to act on.
When RecRam is not the right fit
- Open-ended exploratory discovery. If you don’t yet know what to ask, you need to follow the conversation in real time. Book a live call on the video tool you already use and record it with the participant’s permission.
- Measuring at scale. If you need a score across hundreds of customers, use a standard survey. Our comparison of NPS, CSAT and CES helps you pick the metric.
- Watching people use a prototype. Task-based usability testing, moderated or unmoderated, is the better method when you need to see where people click and stall.
- Feedback that must be anonymous. Video and voice answers aren’t anonymous. If people need anonymity to be honest, use a text-only anonymous survey.
- A few interviews a quarter. If you only talk to three customers every few months, a calendar link and a call is all you need.
Templates to start from
These templates give you a ready-made starting point that you can edit with the questions above:
- Product Discovery: Problem Validation tests whether a problem exists and how severe it is with potential users. Pair it with questions 1 to 7.
- User Research: Voice of Customer Insights collects open feedback about the overall product experience. Pair it with the usage and churn questions.
- Niche B2B Market Pain Point Discovery looks for underserved pain points in a specific B2B segment. Pair it with questions 34 to 38.
You can browse more in the template library, or start with the Problem Validation template on a 28-day free trial, no credit card required. For running feedback collection as an ongoing program, see how to collect customer feedback at scale.
Frequently asked questions
What are the best customer interview questions?
The best questions ask about a specific recent experience: “Tell me about the last time you…”, “Walk me through what happened”, “What did you do instead?” They avoid hypotheticals (“Would you use…?”) and questions that suggest an answer. Pick questions that match your goal, whether that’s problem validation, switching, onboarding, usage, churn or B2B buying.
How many questions should a customer interview have?
For a live 30-minute interview, plan five to eight core questions and leave most of the time for follow-ups. For an async recorded interview, four to six questions is plenty, with each answer running one to three minutes.
How long should a customer interview be?
Live story-based interviews commonly run 20 to 30 minutes. Teresa Torres’s story-based interview course includes real interviews of that length. With about six questions at one to three minutes each, an async interview takes a respondent roughly 10 to 15 minutes.
What’s the difference between customer discovery and customer feedback interviews?
Customer discovery interviews explore a problem before or outside your product, to decide what to build. Customer feedback interviews ask existing customers about their experience with your product, to decide what to fix, improve or keep. The same rules apply to both: ask about specific past behavior and don’t lead.
Can customer interviews be done asynchronously?
Yes. Sending the questions as a link that people answer by video, voice or text works well for focused questions, busy B2B buyers, customers in other time zones and studies where you need to hear from many people quickly. For deep exploratory discovery, a live conversation is still the better choice.
Ready to try it?
Stop reading. Start listening.
RecRam turns feedback into video conversations your team actually watches. Free forever, no credit card required.
Start Free, No Credit Card
Written by
Nebi KaraFounder & Marketing Specialist
Nebi Kara specializes in marketing, software and video technology, with a focus on practical tools for recording, securing and verifying order evidence.
