CSAT Calculator
Work out your customer satisfaction score from survey ratings, see the full spread of responses, and calculate customer effort score too. Free, no sign-up.
Free. No sign-up. Everything is calculated in your browser, nothing you enter is sent anywhere.
How it's calculated
- CSAT score
- CSAT = (responses rated 4 + responses rated 5) / total responses x 100
- Average rating
- Average = sum of (rating x count) / total responses
- Customer effort score
- CES = sum of (rating x count) / total responses, on a 1 to 7 scale
Worked examples
CSAT and CES calculation examples
Four survey results worked through, including why the headline score is not the whole story.
100 responses to "How satisfied were you with this conversation?"
Ratings 1 to 5: 5, 5, 10, 30, 50.
30 + 50 = 80 satisfied responses out of 100, a CSAT score of 80%. The average rating is 4.15 out of 5.
Your helpdesk only reports the headline counts.
412 satisfied responses out of 500.
412 / 500 x 100 = a CSAT score of 82.4%.
Two teams that both score 80%.
Team A: 0, 10, 10, 80, 0. Team B: 10, 5, 5, 10, 70.
Both are 80% CSAT, but Team B has 10% of customers at "very unsatisfied" and Team A has none. Always look at the spread, not just the headline score.
100 responses on a 1 to 7 scale, where 7 means very easy.
Ratings 1 to 7: 2, 3, 5, 10, 20, 30, 30.
The average is 5.53 out of 7. 80% found it easy (5 to 7) and 10% found it hard (1 to 3).
Benchmarks
How to read your CSAT and CES
General rules of thumb. Your industry, survey timing and channel all shift what a typical score looks like.
80% CSAT and above is strong
Most support teams land somewhere between 70% and 85%. Scoring above 80% consistently is a good sign as a general rule of thumb.
Below 70% deserves a closer look
Read the comments on the 1 and 2 ratings first. Low scores usually cluster around a few specific problems, like slow first replies or one confusing policy.
CES of 5.5 or higher means low effort
On a 1 to 7 scale where 7 is very easy, an average above 5.5 suggests customers rarely struggle. Below 4.5 points to friction worth fixing.
Compare like with like
CSAT varies with survey timing, wording and channel. Track your own score over time and compare teams using the same survey, not against someone else's number.
FAQ
Questions people ask
How do you calculate CSAT?
Add up the responses that rated 4 (satisfied) or 5 (very satisfied) on a 5-point scale, divide by the total number of responses and multiply by 100. The result is a percentage from 0 to 100.
What is a good CSAT score?
As a general rule of thumb, 75% to 85% is solid for most support teams and above 85% is excellent. The most useful comparison is your own score over time.
Why not just use the average rating?
CSAT counts the share of happy customers, which is easier to act on and compare. The calculator shows the average rating too, because two teams with the same CSAT can have very different averages.
What is the difference between CSAT, NPS and CES?
CSAT measures satisfaction with a specific interaction. NPS measures how likely people are to recommend you overall. CES measures how easy it was to get something done. Many teams track CSAT and CES on support and NPS for the whole product.
How is customer effort score calculated?
Ask customers to rate "The company made it easy for me to handle my issue" from 1 to 7. CES is the average of all responses. Higher is better on this scale.
Is anything I enter stored?
No. The calculator runs entirely in your browser. Nothing you type is sent to a server or saved.
More free tools
- Customer Effort Score CalculatorCES as an average of 1 to 7 scores or as percent favorable.
- NPS CalculatorNet Promoter Score from promoters, passives and detractors.
- SLA CalculatorUptime to allowed downtime, and support response due times.
- Churn Rate CalculatorCustomer churn and retention rate for any period.
The easiest way to lift CSAT is a faster, right first answer.
Bund AI answers your customers on web chat and email in seconds, from your own help content, and passes anything it can't handle to your team with the full conversation, so nobody has to repeat themselves.