30 questions to ask customers about your product (and how to count the answers)
30 customer feedback questions grouped by goal, what a good answer sounds like, what not to ask, and how to tag answers so you can count them.

There is no shortage of lists of customer feedback questions. Most of them are fine. They all stop at the same place: the moment you finish asking.
The value isn’t in one good answer. It’s in knowing, after twenty calls, how many customers said the same thing, which accounts they are, and what they’re worth. So this list is built to be counted. Each question comes with what a good answer sounds like, and the last section shows how to tag answers so they add up.
The question behind the questions
Before you pick questions, decide what you’ll do with the answers. The goal of the call decides what you ask.
- 🗺️ A roadmap call is about jobs and pain: what customers are trying to get done and where it hurts.
- 💵 A pricing call is about value: what they’d lose without you, and who signs off on paying.
- 🚪 A churn call is about switching: what almost made them leave, and what a competitor would need to offer.
Pick one goal per call. A call that tries to cover everything gets shallow answers to all of it.
30 questions to ask customers about your product
🧭 Understanding the job (6)
- “Walk me through the last time you did [task].” A good answer is a specific story with steps, tools and people, not a general description.
- “What happened right before you opened the product?” Listen for the trigger: a meeting, a deadline, a request from a boss.
- “Who else was involved?” Tells you who the hidden stakeholders and approvers are.
- “What did you do once you had the answer?” Shows what the work feeds into and what “done” means to them.
- “How often does this come up?” Daily problems and quarterly ones are different products.
- “What does a good week look like for this part of your job?” Gives you the outcome they care about, in their own words.
🩹 Finding the pain (6)
- “What’s the most annoying part of [task] today?” A good answer names a step, not a feeling.
- “What have you tried instead?” If they’ve tried nothing, the pain may not be real. If they’ve built a spreadsheet, it is.
- “How much time did that take last time?” Turns a complaint into a cost.
- “What happens if it doesn’t get done?” Separates must-have from nice-to-have.
- “When did this last go wrong?” A real, recent failure is worth more than a hypothetical one.
- “If you could fix one thing about how you do this, what would it be?” Their top priority, ranked by them.
💰 Value and willingness to pay (6)
- “What would you lose if we disappeared tomorrow?” A strong answer is specific and slightly panicked.
- “What did you use before us?” Tells you the real alternative you’re priced against.
- “Who else signs off on paying for this?” Maps the buying process.
- “How do you explain the value of this to your boss?” Gives you their business case, ready to reuse.
- “What would make this worth twice what you pay?” Points to expansion, without asking them to name a price.
- “What else is in the budget this sits in?” Shows what you compete with for money, not just attention.
🛠️ Feature requests, the right way (6)
- “What were you trying to do when you wished it did that?” The job behind the request matters more than the request.
- “How do you work around it now?” A painful workaround means a real need. No workaround means a mild one.
- “How often does that come up?” Frequency is half of priority.
- “What would you do with it the first day you had it?” Separates “would use” from “would mention once”.
- “Who else on your team runs into this?” One user’s preference or a team-wide gap.
- “What would ‘done’ look like for you?” Gives you acceptance criteria in the customer’s words.
🔄 Churn and switching (6)
- “What almost made you leave?” Ask happy customers too. The near misses are your early warnings.
- “What would a competitor need to offer for you to switch?” Names your real weak spots.
- “When did you last look at alternatives, and why?” A date and a trigger beat a yes or no.
- “What did you expect that we haven’t delivered?” The gap between the promise and the product.
- “What would you tell a peer considering us?” Their honest pitch, and their honest warning.
- “If you were running our company, what would you change first?” Permission to be blunt, which usually gets you the real answer.
Questions not to ask, and what to ask instead
Some questions feel useful and produce answers you can’t trust. Rob Fitzpatrick’s The Mom Test makes the case well: people are kind, so they’ll tell you your idea is great. Ask about their past, not your future.
- 🎣 Leading: “Don’t you think it’d be easier if…?” puts the answer in their mouth. Nielsen Norman Group has a good guide to spotting these (NN/g).
- 🔮 Hypothetical: “Would you use a feature that…?” gets a polite yes. Ask what they did last time instead.
- 🔫 Double-barreled: “How do you feel about search and filters?” gets one answer to two questions. Split it.
- 🤞 Future promises: “Would you pay for…?” is cheap to say yes to. Ask what they pay for today.
Good feedback questions vs great ones
Great questions are open, past-tense, specific and about behaviour. Teresa Torres calls this story-based interviewing: collect specific stories, not opinions (Product Talk).
| 👍 Good | ⭐ Great |
|---|---|
| 💬 “Do you like the dashboard?" | "When did you last open the dashboard, and what were you looking for?” |
| 💬 “Would you use an export feature?" | "How did you get this data out last time?” |
| 💬 “Is onboarding easy?" | "Walk me through your first week with the product.” |
| 💬 “What features do you want?" | "What were you trying to do when you got stuck last?” |
| 💬 “How likely are you to recommend us?" | "Who did you last tell about us, and what did you say?” |
| 💬 “Is the price fair?" | "What did you compare us against when you bought?” |
The part every list skips: counting the answers

Asking is the easy half. After twenty calls, try answering these:
- 🔢 How many customers raised onboarding friction?
- 🏢 Which accounts are they, and what are they worth?
- 📈 Is it more common than it was last quarter?
If you can’t answer without rereading your notes, the interviews haven’t paid off yet. Three habits fix it.
- 🏷️ Tag by need, not by wording. “Export is a pain”, “I copy it into Sheets” and “my boss wants a CSV” are the same need. Give it one tag, written as the customer’s goal (“get data out for reporting”), so the count adds up.
- 👥 Count customers, not quotes. One customer who mentions a need five times is one customer. Count distinct accounts.
- 🔗 Keep the link to the source. Every tag should point back to the exact moment in the call. When someone doubts the number, you can show them.
Done by hand, this works for a few dozen calls and then breaks. Every new call means re-tagging, and the same need drifts into five slightly different tags. That’s the point where teams move the counting into a tool that processes every conversation, groups the needs and keeps the counts current. Our guide to counting how many customers asked for a feature walks through it.
FAQ
What are good questions to ask customers about a product? Open, past-tense questions about what they actually did: “Walk me through the last time you did X”, “What have you tried instead?”, “What would you lose if we disappeared tomorrow?” They get stories you can trust, not opinions.
How many customers should you interview? Enough to see the same needs repeat. For one focused question, patterns often start to show within five to ten conversations per segment. Keep counting after that, because the counts change as you add calls.
What is the difference between good and great feedback questions? Good questions ask what customers think. Great questions ask what they did, recently and specifically. “Would you use this?” is good. “How did you solve this last time?” is great.
Ask the questions, then count the answers. Connect Evermuse to your calls and it tags and counts every answer, with the quote behind each one. Sign up free.