What is CSAT?
CSAT, or customer satisfaction score, is a metric that measures how satisfied a customer was with one specific interaction, shown as the percentage of people who rated it 4 or 5 on a five-point scale.
The interaction can be a support conversation, an onboarding step or a release. CSAT is deliberately narrow, and that's what makes it useful: a low score points at one place in the experience instead of at the whole relationship.
The same acronym is also used for a counselling certification and for a civil service aptitude exam. This page covers the customer metric only.
How do you calculate CSAT?
Divide the number of satisfied responses (4s and 5s on a five-point scale) by the total number of responses, then multiply by 100.
CSAT = (number of 4 and 5 responses Ă· total responses) Ă— 100
Say 200 people rate a support conversation. 90 choose 5, 64 choose 4, and the other 46 pick 1, 2 or 3. That's 154 satisfied responses out of 200, so the CSAT is 77%.
Counting the top two answers (often called top-two-box) is the most common method. Some teams report the average rating instead, for example 4.1 out of 5. Both work, but they give different numbers, so write down which one you use and keep it between measurements. A CSAT score means little on its own and a lot as a trend.
What scale is used for CSAT?
Most CSAT surveys use a five-point scale from 1 (very dissatisfied) to 5 (very satisfied). Other scales work too, as long as you count the same way every time.
- 1 to 5: count the 4s and 5s.
- 1 to 7: count the 6s and 7s.
- 1 to 3, or three faces: count only the top option.
- Stars or emoji: map them to 1 to 5 first, then use the five-point rule.
A 1 to 10 scale shows up as well, but it makes CSAT easy to mix up with NPS, which runs from 0 to 10. Label both ends of the scale in words, so nobody reads a 1 as the best score.
What is a good CSAT score?
Commonly cited bands treat 70% to 85% as good and anything above 85% as strong. Use them as rough orientation. There is no single official CSAT benchmark, and scores swing widely by industry, channel and timing, so the comparison that matters most is your own score for the same question last quarter.
For a public cross-industry reference, the American Customer Satisfaction Index (ACSI), developed at the University of Michigan and published since 1994, scores industries and companies on a 0 to 100 scale. Its method differs from a top-two-box CSAT, so read it as context and don't compare the two numbers directly.
Two things distort CSAT more than most people expect. Asking straight after a resolved support ticket produces higher scores than asking a week later. People who abandon a task also rarely stay to rate it, so the sample leans toward the ones who succeeded.
What questions does a CSAT survey ask?
A CSAT survey asks one rating question about one interaction, followed by one open question about the reason.
- The rating: "How satisfied were you with [the support you received today]?" on a labelled 1 to 5 scale.
- The reason: "What's the main reason for your score?" as an optional text field.
Keep the bracketed part concrete. "How satisfied are you with our product?" measures overall mood, which NPS already covers. "How satisfied were you with setting up your first report?" points at a step your team can change.
When should you send a CSAT survey?
Send it immediately after the thing you're measuring, and ask only about that thing. A survey that appears at a random moment can't tell you which part of the experience the score belongs to.
Good moments: after a support ticket closes, after onboarding completes, after someone uses a feature for the first time, after a release the customer was waiting for.
What is the difference between CSAT, NPS and CES?
CSAT measures satisfaction with one interaction, NPS measures the overall relationship, and CES measures how much effort a task took.
- CSAT asks how satisfied you were with one interaction. Narrow and immediate, good for spotting a step that isn't working.
- NPS asks how likely you are to recommend the company, on a 0 to 10 scale. Broad and slow to move, it tracks the relationship.
- CES asks how easy something was. Research from CEB (now part of Gartner), published in Harvard Business Review in 2010, found that reducing effort did more for loyalty than exceeding expectations.
They aren't interchangeable, and running all three on the same people is a quick route to survey fatigue. Pick the one that matches the decision you need to make.
What doesn't CSAT tell you?
CSAT tells you that an interaction fell short for some people, and nothing about why. A 62% CSAT on onboarding tells you where to look.
The reason sits in the comment box underneath. Many teams collect those comments and never read them, because going through a thousand free-text answers by hand is a job nobody volunteers for. Grouping comments by topic automatically is the practical answer, since a score you can't explain won't change any decision.
CSAT for enterprise product teams
At enterprise scale, CSAT gets measured in many places at once. Support runs it on tickets, product runs it on onboarding, and each team defines "satisfied" a little differently, so the numbers can't be added up or compared.
Three things make CSAT usable across teams: one written definition (scale, top-two-box or average, which touchpoints), the same question wording everywhere it runs, and comments that stay attached to their scores. When a product manager reports that onboarding CSAT went from 81% to 74%, the first question in the roadmap review is why. The answer should be a short list of themes with a count next to each, every one traceable to the responses behind it.
How Usersnap measures CSAT
Usersnap runs CSAT as an in-app micro-survey: a satisfaction rater shown inside your product at the moment you want to measure. Responses land in the same workspace as the rest of your feedback, with the follow-up comment attached to each score.
- Timing: start from the ready-made CSAT template, show the survey widget on the pages where the step happens, and on plans that include API-event targeting, trigger it when the step is completed.
- Outside the product: support agents can embed the CSAT rater in an email at the end of a conversation.
- Reading the comments: Sentiment Sensor, included on Professional plans and above (an add-on on Starter and Growth), tags each new response with a topic and a sentiment and shows the top three positive and negative topics, each linked to the responses behind it.
- Asking your AI assistant: with the MCP connector, ChatGPT, Claude or Cursor can query your live workspace with your own permissions, for example "what are the five most trending topics in this month's CSAT comments?" The connector sees new data, and Customer Success can index older responses.
- Deciding: link the responses behind a theme to an opportunity on the Opportunities Board, where it gets a value and effort score next to the rest of the roadmap.
CSAT FAQ
How is CSAT calculated?
Divide the number of responses in the top two options by the total number of responses, then multiply by 100. On a five-point scale, that means counting the 4s and 5s.
What is a good CSAT score?
Commonly quoted bands treat 70% to 85% as good and above 85% as strong. Industry and timing shift the number so much that your own trend is the more useful comparison.
What is a good CSAT score out of 5?
If you report CSAT as an average rating, 4 or higher means the typical answer is "satisfied", because 4 carries that label on a five-point scale. Compare averages only with other averages, never with a top-two-box percentage.
Is CSAT a KPI?
Yes. CSAT is one of the most widely used customer experience KPIs, usually tracked next to NPS, CES and churn rate. It works best as a KPI for one touchpoint, such as support or onboarding, with the target set against your own past scores.
Can Usersnap run CSAT surveys inside a product?
Yes. Usersnap shows a CSAT rater as an in-app micro-survey and keeps each score together with its comment. On Professional plans and above, Sentiment Sensor groups new comments by topic and sentiment.
Related terms
See how product teams run CSAT at the moment of the interaction and keep every score tied to its reason.
.svg)
