
You have customers, and you could not say for certain why any of them picked you over the next business on the list. Is it the price, the reviews, how fast you answered, or something you have never thought to mention? Your website and your ads are guessing at the answer, and a guess reads like every other business in your field.
Customer surveys can replace that guess with the customer's own words. You do not need software, a dashboard or a consultant to start. You need two questions, five recent customers and a place to write down what comes back.
Key Takeaways
"What was happening when you decided to look for a business like ours?" and "What nearly stopped you from hiring us?" give you the reason and the obstacle in the customer's own words.
Open with the customer's experience, then choice, obstacles, satisfaction and what would bring them back, and send one reminder at most.
A satisfaction or NPS score tracks a trend; a follow-up question asking for the main reason is what explains it.
Sort by the phrases customers use, not by the categories you prefer, then turn the repeats into page copy, FAQ answers and ad lines.
A customer's words used in advertising are an endorsement, and a reward offered for comments must be disclosed.
Two questions that show why customers picked you
Start with these two, each sent as its own question and worded the same way for every customer.
- "What was happening when you decided to look for a business like ours?"
- "What nearly stopped you from hiring us?"
The first tells you what the customer was doing when they began looking, which is the moment your website needs to meet them in. The second names the doubt you had to beat. Keep both open, with no list of answers to pick from, because a list can steer the reply. Pew Research Center notes that wording, question order, response options and open versus closed formats all change the answers people give.
If a customer finished a purchase, add a third: "What was the main reason you chose us?" Send the "nearly stopped you" question to people who bought. If you also have recent leads who did not move forward, send them their own version, "What was the main reason you decided not to move forward with us?", and keep their answers in a separate pile. Mixing buyers and non-buyers blurs both.

A short survey in the order people can answer easily
Once the first two questions work, you can build a slightly longer form. The order matters. Pew's guidance supports starting with the customer's experience and then moving on to the choice they made, anything that got in the way, how it went, and what they would do next. The AAPOR best practices add that you should let people skip a sensitive question and protect any personal details they give you.
| Step | Question to ask | What it tells you |
|---|---|---|
| Experience | What was happening when you decided to look for a business like ours? | The moment and the need behind the search |
| Choice | What other options did you consider? What was the main reason you chose us? | Your real competitors and your real draw |
| Obstacles | What nearly stopped you from hiring us? What did you expect before buying? | The doubt your site has to answer |
| Satisfaction | What worked better or worse than expected? | Where the job beat or missed the promise |
| Coming back | What would make you choose us again? | What earns the second job |
You will not need every line. Pick three to six questions for the survey, test it on one person you know, and cut anything that repeats another question. For a customer who did not return, two more fit: "If you decided not to return, what was the main reason?" and "What could we have done differently?"
Send it while the job is fresh, then stop at one reminder
Send the survey soon after the customer has seen the result: after delivery, installation, project completion, a support issue being resolved, or a cancellation. The memory is fresh, and you are asking about something they just lived through. SurveyMonkey's post-purchase guidance points the same way.
Use email when your relationship with the customer runs by email. Use text only when the customer already texts with you and has agreed to business messages. Aim for a form that takes under three minutes, send one reminder, and then let it go.
Do not plan around a response rate you read somewhere. SurveyMonkey's own response rate benchmark guide says expectations differ by audience, frequency, timing, format and length. Instead, write down how many people you invited, how many finished, and which group they came from. AAPOR's response rate calculator shows how that count is figured.

A score tells you the trend, a reason tells you what to change
You may already collect a satisfaction rating or an NPS score. Keep it, because a score can show whether things are getting better or worse over time. It does not say why. Someone who gives a 6 could be unhappy about price, timing, a late callback or a single bad moment, and the number looks the same each way.
Add one question right after the score: "What is the main reason for your score?" The reason is the part you can act on.
Also remember who answered. A high response rate does not by itself mean the replies represent everyone, and a low one does not make the answers useless; it means the people who replied may differ from the people who did not. AAPOR's standard definitions explain how response is counted. One comment is a lead worth following up, not proof of what every customer thinks.
Group replies by the words customers repeat, then use them
Read every answer and copy the customer's phrases exactly, without tidying them into marketing language. Then sort them by what the customer said, not by the categories you would choose. Four piles are enough:
- Reasons for choosing you: what pulled them in
- Obstacles before buying: what nearly stopped them
- What they experienced: what worked or fell short
- Reasons for leaving: why a customer did not return
When a phrase shows up in several replies, it earns a place on your site. Five answers cannot tell you what every customer thinks, so treat a repeat as a thing to test, and label how many people you asked.
| What customers repeat | Where it can go |
|---|---|
| The reason they chose you | Homepage or service-page benefit statement |
| A question they asked before booking | A FAQ entry using their wording |
| The doubt that nearly stopped them | A short passage that answers it near the button |
| A problem they described in their own words | An ad headline, once you have confirmed it is accurate |
| A theme across several jobs | A case-study angle |
Do not use a reply to say something you cannot back up. If a customer says you were the fastest in town, that is their opinion, not a claim you can print as fact. The two posts on what to put on a service page and the swap-the-name test show where this language helps most, because a phrase a customer used is a detail a competitor's page cannot copy.

Quote a customer without creating a problem
A survey answer becomes an endorsement the moment you use it in advertising as a customer's opinion or experience. The FTC's endorsement guide for businesses says you may ask customers to allow their comments to be quoted, and that endorsements must reflect the endorser's real experience or opinion and cannot make claims that would be deceptive or unsubstantiated if you made them yourself.
Get written permission before you publish a person's words, name, photo, company or any detail that could identify them. A short written release that names the quote and where it will appear is enough to keep a record.
If you offered a benefit for the comments, such as a discount or a gift card, say so wherever the comment appears. The FTC's consumer reviews and testimonials rule, effective October 21, 2024, targets fake, false and deceptive reviews and testimonials.
Public reviews have their own rules. Google's Maps user-generated content policy prohibits fake and misleading content and content created through incentives tied to a positive review. Ask for feedback, never tie a reward to a favorable answer, and do not ask only your happiest customers. The full routine for asking for reviews is in how to get more reviews.

Google Forms or Microsoft Forms, whichever matches your account
Both free tools are adequate for a short customer survey. Use the one that matches the account you already have.
Google Forms lets you build a form, share it by link or email, and see summaries as answers arrive. Under the form's Responses tab, Google lets you view and manage the replies, export them to Google Sheets or download a CSV, and get notifications for new ones.
Microsoft Forms works from forms.office.com or Microsoft 365 and offers text questions, choice questions, ratings, ranking, sections and Net Promoter Score questions, with results visible as they arrive.
If your business lives in Google Workspace, use Forms from Google. If it lives in Microsoft 365, use Microsoft's. Check your privacy needs and where you want the replies stored before you pick. Neither tool will fix weak questions; the wording does that work.
Start alone, and know when help pays for itself
Five recent customers, one neutral question and a written record are enough for a first look, and you can review the wording yourself. A survey will not guarantee more sales. It tells you what to say, what to answer and what to fix, and it is up to you to make the change.
Outside help earns its fee in a few situations: you have several customer types with different reasons for buying, you plan a survey you will repeat, the answers need to connect to a CRM or support system, the data is sensitive, or you want the replies turned into site structure and campaigns. The useful work is better question design, analysis that keeps the customer's wording, and putting the findings into your pages and ads. Whoever you hire should leave you with the raw answers and a say over any public quote.
Do you know how to run a customer survey?
Pick an answer to begin.
1. Which question tells you what nearly kept a customer from hiring you?
2. A customer gives you a 9 out of 10. What explains why?
3. Before you print a customer's survey answer in an ad, what should you have?
Frequently Asked Questions About Customer Surveys
What are good customer survey questions?
Start with "What was happening when you decided to look for a business like ours?" and "What nearly stopped you from hiring us?" Add "What was the main reason you chose us?" and "What other options did you consider?" Keep every question open and neutral.
What questions can I ask to learn what customers need?
Ask what they expected before buying, what worked better or worse than expected, and what would make them choose you again. Those three show the need behind the purchase and how the job measured up against it.
How many questions should a customer survey have, and when should I send it?
Use three to six, ideally under three minutes, and send it soon after delivery, installation, project completion or a resolved support issue. Send one reminder and stop.
Can I use survey answers as testimonials?
Yes, with written permission before you publish the person's words, name, image or company. The FTC says endorsements must be genuine and not misleading.
Is NPS enough to explain why customers leave?
No. A score tracks a trend. Add a question asking for the main reason, which is what explains it.
Can I reward customers for completing a survey?
A reward should not depend on a positive answer or a positive review. If a benefit was offered for comments you later use in advertising, disclose it.
Moving Forward
You now have the order of a short survey, the two questions to send first, a way to sort the replies and the permission rule for quoting anyone. The aim is simple: learn why customers picked you and what nearly stopped them, then say it on your pages in their words.
Those answers keep paying off. Each batch of replies shows whether your homepage, ads and follow-up still match why people hire you, and it gives you a short list of changes worth making. A survey cannot promise sales, but it removes the guesswork about what to say.
If you want help turning the answers into pages and ads, Web Leveling can organize the replies, write the copy and set up the campaigns around it. Our digital marketing and content marketing work starts from what your customers say. We work with small and medium businesses across the country and overseas. Tell us what your customers are telling you, and we will help you put it to use.
Terms
Customer survey words in this post
Tap a term to see what it means.
Customer survey. A short set of questions sent to customers to learn why they chose a business, what nearly stopped them and how the job went.
Open question. A question with no answer choices, so the customer replies in their own words.
NPS. Net Promoter Score, a rating of how likely a customer is to recommend a business; it tracks a trend but does not give the reason.
Response rate. The share of invited people who finish a survey; it varies by audience, timing, channel and length.
Endorsement. A customer's opinion or experience used in advertising, which needs the customer's permission and accurate wording.
Material connection. A reward or benefit given in return for comments, which must be disclosed when the comments are used.




