Customer Satisfaction Survey: Find What to Fix
A customer satisfaction survey asks customers to rate a product, service, or experience. Its purpose is to understand satisfaction and identify areas for improvement. Qualtrics Start with a question that names the experience, add an optional explanation, and use the answers to decide what needs attention.
Customer satisfaction survey template
A short starting point is a satisfaction rating followed by a written explanation. SurveyMonkey describes this combination as a typical CSAT survey and recommends a follow-up that customers can answer regardless of their rating (SurveyMonkey).
The wording below is an editable template. Replace the bracketed text with the specific product, service, or interaction being assessed.
Introduction
“Please share your feedback about [experience]. We will use your answers to improve [product, service, or process].”
1. Satisfaction rating
“How satisfied are you with [specific product, service, or interaction]?”
Use the following five-point satisfaction scale, with the highest score assigned to the most positive response (SurveyMonkey).
- 1 — Very dissatisfied
- 2 — Somewhat dissatisfied
- 3 — Neither satisfied nor dissatisfied
- 4 — Somewhat satisfied
- 5 — Very satisfied
2. Reason for the rating
“What led you to choose that rating?”
Response format: optional text box.
Keep this question open to positive, neutral, and negative feedback. Use the same wording for every rating so the comments can be reviewed together or filtered by score.
3. Permission to follow up, when follow-up is available
“May we contact you about this feedback?”
Response options: Yes / No.
Qualtrics recommends a separate permission question when responses will feed into customer follow-up. Add a contact field only when contact details are needed, and state a response time only when it can be met (Qualtrics).
For questions that do not apply to every recipient, provide an unscored option such as “Not applicable” or “I have not used this.” Keep it separate from the neutral satisfaction rating. AAPOR recommends response choices that cover reasonable answers, including inapplicable or unknown responses where appropriate (AAPOR).
Questions to add when more detail is needed
Choose additional questions according to the improvement being investigated. Select the suggested questions that address the information needed.
| Feedback needed | Suggested question | Response format |
|---|---|---|
| Product quality | How satisfied are you with the quality of [product]? | Satisfaction scale |
| Delivery timing | How satisfied are you with the time taken to deliver [order]? | Satisfaction scale |
| Clarity of support | How clear was the information provided during [support contact]? | Scale from very unclear to very clear |
| Issue resolution | Has [reported issue] been resolved? | Yes / Partly / No / Unsure |
| Ease of completing a task | How easy or difficult was it to [task]? | Scale from very difficult to very easy |
| Improvement priorities | What change to [product or service] would matter most to you? | Optional text box |
Each question measures a separate aspect of the experience. Pew Research Center recommends asking about one concept at a time and using clear, concrete language. Keep quality, speed, clarity, and resolution in separate questions when separate answers are needed (Pew Research Center).
Satisfaction, ease, and recommendation also measure different things. CSAT asks about satisfaction; Customer Effort Score concerns the difficulty of a task; Net Promoter Score asks about willingness to recommend. Label effort and recommendation questions separately and calculate CSAT from satisfaction ratings alone (Qualtrics).
When to send a customer satisfaction survey
For feedback on a particular interaction, choose timing according to when it can be assessed. Qualtrics describes surveys after product or service delivery and notes that the appropriate delay depends on the product (Qualtrics).
Use the subject of the question to set the invitation trigger:
| Survey subject | Invitation trigger to use | Experience to name in the question |
|---|---|---|
| Purchase process | Completion of checkout or purchase | The purchase process |
| Delivery | Confirmation that delivery has occurred | The delivery |
| Product setup or use | An opportunity to complete setup or use the product | Setup or product use |
| Support conversation | End of the support interaction | The help received during that contact |
| Final issue resolution | A point when the resolution can be assessed | Whether the reported issue was resolved |
These are suggested trigger rules. Set the delay according to when the experience can be judged, and keep the invitation wording aligned with that trigger. An invitation about product use needs a different trigger from one about purchasing the product.
A relationship survey assesses the broader experience over time. Qualtrics describes relationship surveys as periodic measures and recommends defining the customer group and distribution frequency (Qualtrics).
For this format, specify the period in the main question: “Overall, how satisfied are you with [service] during [time period]?” Define eligibility before distribution, including how much experience is needed to answer. Report relationship ratings separately from individual transaction ratings.
Limit repeated invitations to the same customer. Qualtrics advises selecting useful points in the customer journey rather than repeatedly surveying every touchpoint. Qualtrics Record the most recent invitation, set a repeat-invitation limit, and stop reminders after a response.
Keep the questionnaire clear and easy to answer
Use neutral wording. Remove praise from the question, keep response options in a logical order, and ask about one identifiable experience. Pew explains that wording and answer choices can affect responses; ordered rating categories should follow their underlying scale (Pew Research Center).
Place the main satisfaction question before detailed questions about individual features or service attributes. Qualtrics recommends putting the key metric near the beginning, followed by feedback that helps explain it (Qualtrics).
Use email or a shared survey link according to how eligible customers can be reached. SurveyMonkey documents both approaches for CSAT surveys. SurveyMonkey Test the form on the devices recipients use, including whether scale labels, text fields, and the submit button are visible.
Keep the rating and explanation as the core. Add another question only when its answer will guide a distinct decision. AAPOR recommends short questionnaires and limiting difficult or open-ended questions to reduce respondent burden (AAPOR).
Before launch, pretest the wording with people similar to the intended respondents and pilot the distribution process. AAPOR recommends both questionnaire pretesting and a pilot to check that survey procedures work as intended. AAPOR Check which experience people think they are rating, whether every relevant answer is available, and whether the response is recorded correctly.
How to calculate the CSAT score
For the five-point scale above, a common CSAT calculation counts ratings of 4 and 5 as satisfied responses. Divide that count by the number of valid answers to the satisfaction question, then multiply by 100. SurveyMonkey’s GetFeedback guidance documents this convention (SurveyMonkey).
CSAT (%) = (ratings of 4 or 5 ÷ all valid satisfaction ratings) × 100
The denominator includes ratings of 1, 2, 3, 4, and 5. Neutral ratings remain in the denominator. Invitations, unanswered questions, and unscored inapplicable responses are outside this calculation because they are not satisfaction ratings. Report CSAT as unavailable when there are no valid ratings.
Label the calculation explicitly in the report. The percentage of satisfied respondents and the mean rating are different summaries; SurveyMonkey distinguishes the CSAT percentage from the average rating (SurveyMonkey).
Keep the count for every rating alongside the headline percentage. Review dissatisfied, neutral, and satisfied responses separately, including their comments. Use that detail to choose a follow-up investigation.
Interpret the result with its respondent group
CSAT describes the people who answered the satisfaction question. To assess how far the result can be generalized, examine who was eligible, who received the invitation, and who participated.
A response rate alone cannot establish the accuracy of the result. AAPOR explains that response rates do not reliably distinguish accurate from inaccurate survey estimates and recommends considering other indicators of quality (AAPOR).
Report the score with:
- The exact question and answer scale.
- The experience and period assessed.
- Eligibility rules and how recipients were selected.
- The invitation channel and collection dates.
- The number of valid ratings and the full rating distribution.
- How duplicates, incomplete responses, and exclusions were handled.
This record follows the reporting principles in AAPOR’s disclosure guidance, which covers measurement wording, population, recruitment, collection methods, dates, sample sizes, and data processing (AAPOR).
Also retain invitation and participation counts. Describe the denominator used for any participation percentage; keep it separate from the denominator used for CSAT. AAPOR distinguishes response, cooperation, and completion rates rather than treating them as interchangeable (AAPOR).
For comparisons over time, retain the question wording, response scale, and survey method. Pew advises using identical wording for trend questions and checking the effect of changes in survey mode. Pew Research Center Record changes in eligibility or invitation timing alongside the results.
Set a satisfaction target against a clearly defined baseline for the same experience. Evaluate progress alongside the rating distribution and recurring feedback, using comparable surveys.
Use the answers to decide what to fix
Review comments for recurring themes and connect them to the experience being assessed. Qualtrics recommends analyzing patterns and directing feedback to the people responsible for the relevant product or service (Qualtrics).
Use the following review process:
- Group the feedback. Create categories from the comments, retain the original wording, and allow more than one category for a comment that covers several issues.
- Check the counts. Count distinct responses mentioning each issue and show those counts alongside the relevant ratings. State whether percentages use all rating respondents or only people who wrote comments.
- Verify the problem. Compare the feedback with records about the experience and investigate the reported cause before choosing a fix.
- Assign the change. Record the issue, the evidence, the person responsible, the proposed change, and a review date. Prioritize using the frequency and severity of the reported problem.
- Review subsequent feedback. Use comparable survey questions and invitation rules, and check both the score and whether comments about the issue persist.
Keep missing answers separate from neutral or inapplicable answers when preparing the data. AAPOR’s analysis guidance distinguishes missing responses from substantive response options (AAPOR).
For customers who agreed to contact, follow up about the specific feedback and record the outcome. Qualtrics recommends incorporating permission and follow-up expectations into the survey design (Qualtrics).