CSAT Survey Template for Customer Support

Respondents rate their satisfaction on a scale — most commonly 1–5, from “Very dissatisfied” to “Very satisfied.” Your CSAT score is the percentage of respondents who chose one of the top two ratings (4 or 5).

CSAT = (number of 4 and 5 ratings ÷ total responses) × 100

You can compute your score from raw counts with our free CSAT calculator. Whichever scale you choose, keep it consistent — changing scales mid-year makes trends meaningless.

This structure keeps the survey under a minute for satisfied customers while collecting enough detail from dissatisfied ones to act.

1. 1How satisfied were you with the support you received? (1–5 scale, required)
2. 2What is the main reason for your rating? (open text, required)
3. 3Was your issue fully resolved? (Yes / Partially / No)
4. 4How easy was it to get the help you needed? (1–5 scale, from “Very difficult” to “Very easy”)
5. 5Is there anything we could have done better? (open text, optional)

Conditional logic lets you ask more only when it is warranted:

• If the rating is 1–2, show a question asking what went wrong and whether they would like someone to follow up.
• If the issue was “Partially” or “No” resolved, ask what remains unresolved.
• If the rating is 4–5, skip straight to the optional final comment — do not make happy customers work.

This keeps median completion time low while ensuring every negative experience leaves you with enough context to respond.

• Send shortly after the interaction ends — the same day, while the experience is fresh.
• For tickets, trigger on resolution, not on every reply.
• Cap frequency: a customer with five tickets this week should not receive five surveys. One survey per customer per reasonable window is plenty.
• For live chat, offer the survey at the end of the conversation rather than emailing later.

The score itself is only the starting point. To improve support, dig into the reasons:

• Theme the open-text reasons: response time, agent knowledge, tone, resolution quality and process friction are the usual suspects. See our guide to analyzing open-text responses.
• Cross-reference satisfaction with resolution status — low scores with resolved issues point to experience problems, not outcome problems.
• Track the trend over time rather than reacting to single-day swings, especially at low response volumes.
• Compare across channels (email, chat, phone) to find where the experience breaks down.

• Asking about the agent when the real problem is policy or product — customers rate the whole experience.
• Making every question required, which suppresses response rates.
• Surveying only resolved tickets, which inflates the score by excluding your worst experiences.
• Using CSAT to punish individual agents rather than fix systemic causes — this incentivizes score-begging.
• Collecting reasons and never reading them.