How to Find Out What Your Customers Really Think

Bizer · 2026-10-01

Watch them use what you sell, and ask about what they did, not what they would do. Customers are polite. Ask "would you use this?" and most say yes. Ask "when did you last have this problem, and what did you do about it?" and you get the truth. The old post on this URL made the same point a different way: work side by side with customers to solve a real problem, and you learn what no form will tell you.

Why are surveys so weak?

A survey asks people to predict their own behavior and rate things on a scale. Both are unreliable. People overstate how much they would pay, forget what annoyed them, and answer the question as asked rather than the one that matters.

Surveys have a use: counting things you already understand. "How did you hear about us?" is a good survey question. "What should we build next?" is a bad one.

Our position: for a small business, the most useful research is watching and conversation, and a survey comes last if at all.

How do you work side by side with a customer?

Find a reason to be there when they use your product or service.

• A software or app business: sit with a customer, in person or on a shared screen, and watch them do a real task. Say nothing. Note every pause, every wrong click, every workaround.
• A contractor: stay for the walkthrough after the job and ask the homeowner to show you how they will use the space. You will hear what they did not like before they think to complain.
• A restaurant or shop: work the counter or the floor during the rush, not just the office. Watch where people hesitate at the menu board or stall in the aisle.
• A B2B service: ask a client if you can spend half a day in their office while they use your work. Offer to fix something small while you are there.

The rule is to help with a real problem. Customers relax when you are solving something for them, and they show you how they really work rather than how they think they should.

What should you ask?

Questions about the past, in specifics:

• "Tell me about the last time you needed this. What happened?"
• "What did you try before you found us?"
• "What almost stopped you from buying?"
• "What do you do right after using this?"
• "What is the most annoying part?"

And avoid:

• "Would you buy this?" Everyone is polite.
• "Do you like it?" Same.
• "How much would you pay?" People guess low or high to please you.

When someone says something interesting, ask "can you show me?" or "what happened next?" The second answer is usually the useful one.

How many customers do you need to talk to?

Fewer than you think. When five or six conversations in a row turn up the same complaint, you have found something. If every conversation says something different, you may be talking to customers who are not alike, which is a finding in itself.

Write notes right after each one. Memory flattens details within a day, and the details are the point.

What about reviews and complaints?

They are free research, but skewed toward people who were thrilled or furious. Read them for specifics, such as a delivery that came late or a staff member named, and ignore star averages.

Do not try to manufacture the opposite. The FTC's rule on consumer reviews, in effect since 21 October 2024, bars offering incentives conditioned on positive reviews and bans fake ones. Asking for honest reviews is fine, and so is asking unhappy customers what went wrong. Their answer is usually the most valuable research you will get all year.

What does this cost?

Time, and comfort. A half day beside a customer is a half day not billing. Hearing what people really think of your work is uncomfortable, and the urge to explain or defend is strong. Do not. Every time you defend, they stop talking.

There is also the cost of acting on it. Learning that customers hate your booking process means fixing your booking process. Research nobody acts on is worse than none, because customers who gave you their time notice when nothing changes. Tell them what you changed because of them.

What is uncertain

Small samples can mislead. Six conversations with your most loyal customers tell you about loyal customers, not about the people who left. Try to include at least one or two who stopped buying. They are harder to reach and more honest. And what customers do this year may not predict next year, so this is a habit to repeat, not a project to finish. If what you hear points toward changing what you offer, how to choose a niche for a small business is a useful next step.