Customer satisfaction survey questions
The customer satisfaction survey questions that give the most usable answers are one overall satisfaction rating, one question on the specific part being measured (service, product or support), a likelihood-to-return or recommend question, and one open question asking why. Match the question to the interaction, and keep the wording about one thing.
Start from what the customer did
A satisfaction question has to match an experience the customer actually had. Refiner’s review of customer satisfaction questions makes the point with a sample that assumes what good product use looks like. A customer who checks email twice a day is not dissatisfied if they log in less often than a dashboard expects. Before you write a question, name the interaction it is about: a purchase, a support ticket, an onboarding call, a delivery.
Example customer satisfaction survey questions
| Purpose | Example question | Response format |
|---|---|---|
| Overall | How satisfied are you with your experience overall? | 1 (very dissatisfied) to 5 (very satisfied) |
| Service | How would you rate the quality of our service? | 1 to 5 rating |
| Speed | How satisfied are you with how quickly your issue was resolved? | 1 to 5 rating |
| Expectations | To what extent did our service meet your expectations? | Not at all to completely |
| Return | How likely are you to buy from us again? | Not at all likely to very likely |
| Recommend | How likely are you to recommend us to a colleague? | 0 to 10 |
| Open | What is the main reason for your score? | Free text |
The list above is the same set of ideas that SmallBiz Trends uses in its examples of overall experience questions: likelihood to return, service quality and whether the service met expectations. The wording is ours.
Opinion questions need a comment box
A rating scale is the wrong format for a question that asks for an opinion. Refiner makes this point about a question whose answer is a rating even though the question asks what the customer thinks. If you want a reason, put a text box under the rating, or ask the reason as its own open question. Seagit Forms has both a rating field and a free-text field, so use each for its job.
One thing per question
“How would you rate our product and our support?” gets one answer for two things. A customer who loved the product and was let down by support has nowhere to say so. Ask two questions, one for each. The double barreled questions guide has more examples and the fix for each.
The response order can change the answer
Choose the number of questions by the decision
Write down what you will do with the results before you write the questions. If the satisfaction score will decide whether a support team gets training, the survey needs the service question and the open reason. If it will feed a quarterly report, add the return question so the trend has a second measure. A question that does not feed any decision makes the survey longer for no gain.
A worked example
Suppose a support team wants to know whether faster replies are the main gap. A hypothetical survey sent when each ticket closes asks one overall satisfaction rating, one rating for speed of resolution, and one open question on the reason for the score. Replies that rate speed low and overall satisfaction low, and mention waiting in the open answer, point to response time. Replies that rate speed high and satisfaction low point to the quality of the answer, which speed does not explain.
Two questions answer the question the team cares about because each one isolates a cause. The same survey with only an overall rating would show a dip and no reason for it.
Check before you send
- The question names the interaction, so the customer knows what it is about.
- Each rating asks about one part of the experience.
- The scale labels are the same on every rating in the survey.
- There is an open question for the reason behind the score.
- The survey is sent soon enough after the interaction for the customer to remember it.
Build it in Seagit Forms
Start from the customer satisfaction survey template, then change the questions to the interaction you are measuring. For the recommend question, build the 0 to 10 scale as one radio question with eleven options, and the NPS calculator scores the answers. Set the form to one question per page if you want customers to answer in order.
Sources
- Refiner, “The Best Customer Satisfaction Survey Questions: 50+ Examples” (format of opinion questions, sample assumptions).
- SmallBiz Trends, “7 Free Examples of Customer Satisfaction Survey Questions” (likelihood to return, service quality, expectations met).
- Jotform, “Likert scale survey questions and examples” (response order effects).
Common questions
- What is the best single customer satisfaction question?
- There is no single best question. A satisfaction rating answers how the customer felt, a likelihood-to-return or recommend question answers what they will do, and an open question explains why. Pick one rating and one open question as the core, and add the rest only when you will act on the answers.
- Should a customer satisfaction survey use a 5-point or 10-point scale?
- Either works if you use it the same way every time. Five points is the common choice for satisfaction. A 0 to 10 scale is used for the recommend question, which the NPS calculator scores for you. Do not mix the two scales in one block.
- When should I send a customer satisfaction survey?
- Send it soon after the interaction it asks about, while the customer still remembers it. A survey about a support ticket is usually sent when the ticket closes; a survey about a product makes more sense after the customer has used it for a while.