Simple Customer Surveys That Change What You Sell

A customer survey earns its keep when it changes a decision: which product you stock, what goes into a service package, or which option you stop promoting. A spreadsheet full of people saying they “love the brand” is pleasant, but it will not tell you whether to offer a smaller pack or charge separately for delivery.

You do not need a research department to get useful answers. You need one clear decision, a short set of questions and a way to check whether stated preferences turn into purchases. For a small business, a three-minute survey followed by a modest sales test is often more useful than a lengthy questionnaire sent once a year.

Choose the decision before writing the questions

Start by finishing this sentence: “We will use these answers to decide whether to…” Keep the ending narrow enough to act on within a month.

• A bakery might decide whether to trial a four-piece pastry box alongside its six-piece box.
• A freelance designer might decide whether to offer a fixed-price brand review before a full identity project.
• A creator might decide whether to sell spreadsheet templates separately from a training course.

“Understand our customers better” is not specific enough. It encourages a sprawling survey, and sprawling surveys tend to produce findings nobody owns. Write down the decision, the cost of getting it wrong and the smallest test you could run.

Set a practical evidence threshold too. For example: “If several recent buyers describe the same portion-size problem, we will test 20 smaller boxes.” That is a trigger for an experiment, not proof that the whole market wants one.

Ask about what happened, not what sounds appealing

People are usually better at describing a recent purchase than predicting their future spending. “Would you buy our new premium package?” invites an easy yes. It costs nothing to be encouraging.

Instead, anchor questions to a particular occasion. Ask what prompted the purchase, what alternatives they considered and what nearly stopped them. These answers expose the circumstances your offer must fit.

Keep the wording neutral

Avoid planting your preferred answer. “How useful would our convenient new subscription be?” contains both the idea and its sales pitch. First ask how customers currently reorder, how often they run out and what they do when that happens.

Give people room to say “none”, “not applicable” or “I cannot remember”. Forcing an answer creates tidy data at the expense of accuracy. Where answer choices have no natural order, shuffle them if your survey tool allows it.

Use a five-question survey you can actually analyse

For recent buyers, the following structure gives you a manageable starting point. Adapt the language to the purchase: a consulting engagement needs different wording from a bag of coffee.

1. Trigger: “What was happening that made you look for this product or service?”
2. Alternative: “What else did you consider, including doing nothing?”
3. Friction: “What, if anything, nearly stopped you buying from us?”
4. Experience: “After using it, what did not work as you expected, if anything?”
5. Priority: “If we could change one thing about this offer, what should it be?”

Use short text fields for unexpected insights, but do not make every answer compulsory. If you already know the main alternatives, question two can be multiple choice with an “other” field. Test the survey on a phone yourself, then ask someone unfamiliar with it to complete it. Only promise “three minutes” if that timing is realistic.

Ask price questions cautiously

A question about previous spending gives context; it does not establish willingness to pay for your offer. Someone who spent £200 on an unsuccessful solution may now have less budget, not more.

If price is the decision, use the survey to understand scope and alternatives. Then test a clearly specified offer at a real price. A paid pilot or transparently refundable deposit provides stronger evidence than hypothetical enthusiasm.

Invite the right people, not just the loudest followers

Your audience determines what the answers mean. Recent buyers can explain purchase friction. Repeat buyers can explain why they return. People who enquired but did not buy can explain why the offer lost out. Mixing these groups without labels hides useful differences.

Start with one group that matches your decision. For a product-use question, invite customers after they have had enough time to use it. For a booking-process question, ask soon after the booking. An Instagram poll can generate ideas, but followers are not automatically customers.

Keep the invitation plain: “We are reviewing our pastry box sizes. Could you answer five questions about your last order? We will use the feedback to plan next month's trial.” Send one polite reminder rather than chasing repeatedly.

If you offer an incentive, make it independent of whether the feedback is positive. Collect only the personal information you need, explain its purpose and make follow-up contact optional.

Read small samples without pretending they are forecasts

Suppose you invite 80 customers and receive 18 responses. Seven mention leftovers. That is a useful clue, but it does not mean 39% of all customers need smaller boxes. People who reply may differ from those who ignore the invitation.

Use a small survey to choose the next test, not to declare what the whole market wants.

Tag each response with one or two themes: portion size, delivery timing, unclear instructions, missing option or price. Count how many respondents mention each theme, rather than counting every mention. Keep a few exact phrases alongside the counts; they often explain the problem better than your labels.

Separate a requested solution from the underlying need. “Make an app” might mean “reordering takes too long”. A saved order or reminder email could solve that problem for far less money.

Finally, check the feedback against transaction records, support questions and returns. When several sources point to the same issue, you have a stronger reason to act.

Turn one finding into a small sales test

Imagine the bakery trials 20 four-piece boxes at £12 alongside its six-piece boxes at £16. Before launch, it sets a two-week test and tracks units sold, contribution after variable costs, average order value and leftovers at closing.

Selling all 20 smaller boxes is encouraging, but not sufficient. If existing customers simply switch from a more profitable box, the change may reduce earnings. The bakery also needs to learn whether the smaller option attracts additional orders or improves repeat purchasing.

For a service business, the equivalent could be five paid brand reviews with a fixed scope. Track delivery time and requests for extra work, not just bookings. An attractive offer that consistently overruns is not yet a viable product.

Record the finding, the change, the success measure and the review date in one place. Tell respondents what you tested and what happened. That closes the loop without promising to implement every suggestion.

Conclusion: let answers earn a sales test

The best simple survey reduces uncertainty around one commercial decision. Ask about recent behaviour, listen for repeated problems and check the answers against what customers actually do. Then make a small, measurable change before committing to a bigger one.