A good customer satisfaction survey is short, specific and serves a single purpose. Customers should be able to answer the questions quickly, and the organization should be able to understand from each response which unit should improve what. This guide includes sample questions by touchpoint and the principles to follow when building a survey.
Basic Structure
For most touchpoints, these three questions are enough:
- Main metric question: A CSAT, NPS or CES question.
- Root cause question: Asks, with answer options, which topic most affected the score.
- Open-ended question: Captures anything the customer wants to add in their own words.
Showing the root cause question only to customers who give a low score (1 and 2 on a 1-5 scale) through conditional logic shortens the survey for satisfied customers and collects more detailed information about problematic experiences. Those who give a high score can be shown a separate "What did you like?" question, so that positive and negative reasons do not get mixed up.
In the examples below, satisfaction questions are kept on a 1-5 scale. The recommendation question (NPS) is asked on its own standard 0-10 scale; the two are reported as separate metrics.
Store and Branch Visits
- How satisfied were you with your store experience today? (1-5)
- (For those who gave a low score) What affected your experience most negatively? (Staff attentiveness, waiting time, product availability, store layout, checkout process)
- Is there anything else you would like to add?
If the location of the QR code (checkout, fitting room, table) is automatically added to the response, the area to improve becomes clear without asking. For details, see the QR code customer satisfaction survey guide.
Restaurants and Cafes
- Overall, how satisfied were you with your visit? (1-5)
- (For those who gave a low score) In which area did we fall short of your expectations? (Taste, speed of service, staff attentiveness, cleanliness, price and value)
- How likely are you to recommend us to a friend or family member? (0-10)
E-commerce Orders
- Overall, how satisfied are you with your order? (1-5)
- Did you experience any problem with your order? (Late delivery, damaged product, product did not match the description, missing item, no problems)
- To what extent did the product meet your expectations? (Better than expected, as expected, worse than expected)
Service and Maintenance
- How satisfied were you with your service experience? (1-5)
- Was your vehicle returned at the promised time? (Yes / No)
- How would you rate the information provided by your service advisor? (1-5)
- How likely are you to recommend our service to a friend or family member? (0-10)
Call Center and Support
- How much do you agree with the following statement: "The company made it easy for me to handle my issue." (1-7; 1 strongly disagree, 7 strongly agree)
- Was your issue resolved during this conversation? (Yes, partially, no)
- Was there anything that made the conversation difficult? (Long wait, transferred more than once, having to repeat information, no solution offered)
Post-Resolution Follow-Up
Sending the customer a short resolution satisfaction question after a complaint has been resolved is the final step of the closed loop. Resolution satisfaction is the share of customers who confirm that their issue has been resolved:
- Has the issue you reported been resolved? (Yes, partially, no)
- (Optional) Is there anything else you would like to add?
Writing Principles
- Ask about one topic at a time. Questions that combine two topics, such as "Are you satisfied with the staff and the waiting time?", cannot be interpreted.
- Do not lead. Instead of "Were you satisfied with our friendly staff?", write "How would you rate the attentiveness of the staff?".
- Keep the scale consistent. Within the same survey, a high score should mean the same thing in every question; questions with a standard scale, such as NPS, are a deliberate exception.
- Ask in the customer's language. Opening the survey in the language of the customer's device and allowing them to switch languages often increases the response rate.
- Ask for personal data only when necessary. Contact information questions should be optional and presented together with the privacy notice.
For details on the metrics, read the NPS, CSAT and CES guides.