What is CSAT?
CSAT, or customer satisfaction score, measures how satisfied someone was with one specific thing. A support conversation, an onboarding step, a release. It is deliberately narrow, which is what makes it useful.
You ask one question, usually “how satisfied were you with X”, on a scale of one to five. The score is the percentage of people who answered four or five.
The CSAT formula
CSAT = (number of 4 and 5 responses Ă· total responses) Ă— 100
So 80 satisfied responses out of 100 gives a CSAT of 80%. Some teams use a seven-point scale and count the top two, or a three-point scale and count only the top. What matters is that the definition does not change between measurements, because the number is meaningless in isolation and only useful as a trend.
What is a good CSAT score?
Commonly cited bands put 70% to 85% in the good range and above 85% as strong, but the benchmark that matters is your own last quarter. Scores vary enormously by industry, by channel and by when you ask.
Two things distort CSAT more than most people expect. Asking immediately after a resolved support ticket produces higher scores than asking a week later. And people who abandon a task rarely stay to rate it, so the sample skews toward those who succeeded.
CSAT vs NPS vs CES
Three metrics that get compared constantly and measure different things.
- CSAT asks how satisfied you were with one interaction. Narrow, immediate, good for spotting a broken step.
- NPS asks how likely you are to recommend the company. Broad, relationship-level, slow to move.
- CES asks how much effort something took. The best predictor of whether someone will stay, because effort drives churn more reliably than delight drives loyalty.
They are not interchangeable and running all three on the same audience is how you get survey fatigue. Pick the one that matches the decision you are trying to make.
When to send a CSAT survey
Immediately after the thing you are measuring, and only about that thing. A CSAT survey sent at a random moment measures general mood, which is what NPS is for.
Good moments: after a support ticket closes, after onboarding completes, after a feature is used for the first time, after a release the customer was waiting for.
What CSAT does not tell you
The score tells you something is wrong and nothing about what. A 62% CSAT on onboarding is a signal, not a finding. The comment box underneath is where the useful part is, and most teams collect it and never read it, because reading a thousand free-text responses by hand is not a job anyone volunteers for.
That is the practical case for grouping responses by topic automatically. A score you cannot explain does not change any decision.
Measuring CSAT with Usersnap
CSAT runs as a micro-survey inside the product, triggered at the point you want to measure rather than in a separate email. Responses arrive alongside bug reports and feature requests rather than in a separate tool, and AI-powered automations group the free-text comments by topic and sentiment so the reason behind a score is visible without reading every response.
For building the survey itself, see our guide to customer satisfaction surveys.
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 fours and fives.
What is a good CSAT score?
Commonly quoted bands treat 70% to 85% as good and above 85% as strong, but industry and timing shift the number so much that your own trend is the more useful comparison.
What is the difference between CSAT and NPS?
CSAT measures satisfaction with one interaction and moves quickly. NPS measures the overall relationship and moves slowly. A support team improving response times will see CSAT move first.
When should you send a CSAT survey?
Immediately after the interaction you want to measure, and only about that interaction.
Run CSAT where the interaction happens. Start free with Usersnap, first 20 feedback items included, no card needed.
.svg)
