Free CSAT calculator with customer effort score

How to use the CSAT calculator

1. Pick the metric. Choose CSAT (1 to 5) for a satisfaction survey or Customer effort (1 to 7) for an effort survey.
2. Enter the counts. Type how many customers gave each rating. Most help desks and survey tools export these counts as a table. The score updates as you type.
3. Read the margin of error. Keep Show the 95% margin of error on when you compare two periods. A change smaller than the margin can be noise.

The calculator runs in your browser. Your numbers are never sent to a server.

How to calculate CSAT

Customer satisfaction score (CSAT) comes from one question, usually sent after a ticket is closed: "How satisfied were you with the support you got?" Customers answer on a scale from 1 (very dissatisfied) to 5 (very satisfied).

CSAT = answers of 4 or 5 ÷ all answers × 100. The result is a percentage from 0 to 100: the share of customers who were satisfied.

Why a 1 to 5 scale, and why only 4 and 5?

A 1 to 5 scale is short enough to answer on a phone and is the most common choice in help desk tools, so your numbers stay comparable from one tool to the next. Counting only the top two answers (the "top two box") gives a number that is easy to explain: "75% of customers were satisfied". A 3 is neutral. It counts in the total, so it lowers the score, but it does not count as satisfied.

Some teams report the average rating instead (for example 4.02 of 5). The calculator shows it too. The average hides how answers spread: a mix of 5s and 1s can have the same average as a page of 3s. If you change from one method to the other, your history is no longer comparable, so choose one and keep it.

If your survey uses a different scale, convert it before you enter the counts. On a 1 to 10 scale, a common rule is to count 8 to 10 as satisfied. On a thumbs up or down survey, CSAT is the share of thumbs up.

A worked example

A support team gets 200 answers in one month:

• Satisfied: 90 + 60 = 150.
• CSAT: 150 ÷ 200 × 100 = 75%.
• Average rating: 804 ÷ 200 = 4.02.
• 95% margin of error: about ±6 points, so the true score is likely between 69% and 81%.

If next month's CSAT is 78%, that is inside the margin. It is not yet evidence that service got better. With four times as many answers (800), the margin falls to about ±3 points.

Customer effort score (CES)

Customer effort score asks a different question: "How easy was it to get your issue solved?" The common version uses a 1 to 7 scale, where 1 is very difficult and 7 is very easy. CES = the average answer. The calculator also shows the share of easy answers (5 to 7) and difficult answers (1 to 3), which are easier to explain to a team than an average.

Example: 120 answers, from rating 1 to 7: 4, 6, 10, 14, 22, 34 and 30.

• CES: 626 ÷ 120 = 5.22.
• Easy (5 to 7): 86 of 120 = 71.7%.
• Difficult (1 to 3): 20 of 120 = 16.7%.
• 95% margin of error: about ±0.29 points.

Effort answers often point at the fix more directly than satisfaction does. Low effort scores usually come from the same few causes: the customer had to explain the issue twice, was transferred, had to wait for a reply to a simple question, or could not find an answer that already exists in the help center.

How the margin of error works

A survey is a sample, so the score is an estimate. For CSAT the calculator uses the usual formula for a percentage: 1.96 × √(p × (1 − p) ÷ n), where p is the share of satisfied answers and n is the number of answers. For CES it uses 1.96 × the standard deviation of the answers ÷ √n. Both give a 95% range.

The margin falls with the square root of the number of answers, so small teams should compare months or quarters, not days. The formula is less reliable below about 30 answers and when almost every answer is satisfied. In those cases, read the comments and treat the number as a rough guide.

What CSAT and CES do not tell you

• Who did not answer. Customers who had a very good or a very bad experience answer more often. Track the response rate next to the score.
• Why. The number shows that something changed, not what. The free text comments and the ticket tags behind the low ratings tell you what to fix.
• Loyalty. A customer can be satisfied with one ticket and still leave. For the wider relationship, use the NPS calculator, and track churn with the customer lifetime value calculator.

How to raise CSAT and lower effort

• Group the 1 and 2 ratings by cause each month and fix the largest group first.
• Answer repeat questions on your help pages, and link to them from the places where customers ask.
• Keep the context when a ticket moves between people, so customers do not explain the issue twice.
• Send the survey at the same moment each time, for example one hour after the ticket is closed.

If many tickets are simple questions with answers on your website, a support chatbot that answers only from those pages can take them and pass the rest to your team.