What is CSAT (customer satisfaction score)?
CSAT stands for customer satisfaction score, a metric that measures how satisfied customers are with a product, service, or specific interaction, such as a support call. It's usually collected through a short survey and reported as the percentage of customers who gave a positive rating.
Measuring customer satisfaction is essential to any customer-facing business. Below, learn what a CSAT score is, how to calculate it, what counts as a good score, and how to measure and improve it.
What is a customer satisfaction score (CSAT)?
A CSAT score is a number that helps your business measure how satisfied your customers are with your product, service, or company as a whole.
Using this customer satisfaction metric, you can measure how happy customers are after a purchase or other specific interaction with your business, like a support phone call. Done right, it helps you understand where the customer journey is working and where to focus on improving your CSAT scores.
CSAT scores are usually generated from a customer satisfaction survey sent right after an interaction. A typical survey asks a single question, such as "How satisfied are you with your experience today?", on a rating scale of 1 to 5 or 1 to 10, where the highest number means "very satisfied."
What is CSAT in customer care?
In customer service and contact centers, CSAT most often measures satisfaction with a specific support interaction rather than the brand overall. Teams send the survey after a call, chat, or email is resolved, then track scores by agent, team, queue, or issue type. That makes CSAT one of the most direct signals of how customers experience your support, alongside operational metrics like first contact resolution and service level.
How to calculate a CSAT score
The standard CSAT formula is:
CSAT (%) = (Number of positive responses ÷ Total number of responses) × 100
On a 1 to 5 scale, "positive" usually means a rating of 4 (satisfied) or 5 (very satisfied). On a 1 to 10 scale, teams often count 8 and above. CSAT scores range from 0 to 100%.
CSAT calculation example
If 2,000 customers completed your CSAT survey and 1,800 of them gave a rating of 4 or 5, your CSAT score would be 90%:
1,800 ÷ 2,000 × 100 = 90%
Average score method
Some teams report CSAT as an average rating instead, such as 4.3 out of 5. This shows more nuance between scores, but it's harder to compare with industry benchmarks, which usually use the percentage method. Whichever method you choose, use it consistently so scores are comparable over time.
What is a good CSAT score?
There's no single number that counts as a good CSAT score. What's realistic depends on your industry, the channel you're measuring, the kind of interactions customers are rating, and how your survey is designed. A support team that mostly handles complaints will often see lower scores than one that mostly answers simple questions, even if both are doing excellent work.
A few guidelines make benchmarking more useful:
Compare against your own trend first. Your score over time, by channel and by team, is the most reliable benchmark you have. A steady rise is a better sign than hitting any external number.
Compare like with like. If you use outside benchmarks, make sure they measure the same thing in the same way. Industry indexes such as the American Customer Satisfaction Index (ACSI) use their own methodology and scale, so they aren't directly comparable to a percentage-based CSAT score.
Look at the distribution, not just the average. A 90% CSAT score can still hide a cluster of very unhappy customers. Reading low-score comments, or the conversations behind them, often tells you more than the headline number.
Watch response rates. A high score from a small, self-selected group of respondents may not reflect how customers feel overall.
The limits of survey-based CSAT
CSAT depends on customers answering the survey, and that's its biggest weakness.
Low response rates: Many customers skip post-interaction surveys, no matter how short they are. Fewer responses mean scores can swing sharply from day to day.
Skewed respondents: Customers with a very good or very bad experience are often the most likely to respond, so results may not represent how most customers feel.
Gaps in coaching data: With only a handful of surveys per agent, managers may not have enough feedback to see who needs support or which issues drive dissatisfaction.
Encouraging more customers to respond helps, but some teams also use AI to fill the gap. Dialpad's AI CSAT, available for Dialpad Support and Dialpad Sell, detects customer satisfaction from call transcripts instead of relying only on the customers who complete a survey. AI CSAT Explanations go a step further, analyzing conversations to show why customers were satisfied or frustrated. Together with survey results, that gives teams a more complete view of customer satisfaction across their calls.
When to measure CSAT
CSAT works best when it's tied to a specific moment, so customers rate something they just experienced. Common moments to survey include:
After a support interaction: Right after a call, chat, or email is resolved. This is the most common use in contact centers, and in Dialpad's AI platform for customer experience, CSAT surveys can play automatically after customer calls.
After purchase or onboarding: Shortly after a customer buys or completes customer onboarding, when first impressions are fresh.
After using the product: Once customers have had time to experience what they bought.
Ahead of renewal: Several months before a renewal date, giving your team time to act on low scores before a customer decides whether to stay.
After self-service: A quick rating at the end of help articles shows whether your knowledge base is answering questions.
Getting the timing right is one of several customer satisfaction best practices that make CSAT results more reliable, along with clear question wording and a consistent rating scale.
CSAT vs. NPS vs. CES
CSAT is one of three common customer satisfaction metrics. Each answers a different question, so many teams track more than one:
Metric | What it measures | Typical question |
|---|---|---|
CSAT (customer satisfaction score) | Satisfaction with a specific interaction or experience | "How satisfied were you with your experience today?" |
NPS (net promoter score) | Overall loyalty and likelihood to recommend | "How likely are you to recommend us to a friend or colleague?" |
CES (customer effort score) | How easy it was to get something done | "How easy was it to resolve your issue today?" |
CSAT is best for individual touchpoints, NPS reflects the overall relationship, and CES highlights friction, so choosing between CSAT vs. NPS vs. CES usually comes down to which decision you need the data to inform.
How to improve your CSAT score
Improving CSAT usually comes down to resolving issues faster, more accurately, and with less effort for the customer. A few places to start:
Train and coach agents continuously. Use CSAT results, alongside call reviews, to spot who needs support and on which topics, then coach to those specific gaps.
Route customers to the right agent. Call routing based on skills or issue type reduces transfers, one of the most common sources of frustration.
Handle hold times honestly. Check in with callers during holds and give accurate wait estimates rather than optimistic ones.
Give fast answers to routine questions. Self-service and AI customer service agents can resolve simple requests immediately and hand off to a human agent with context when needed.
Review results regularly. Small, frequent check-ins catch problems early, before a dip turns into a long decline.
Contact centers have additional levers, such as staffing to service level targets and using real-time coaching, that can help improve CSAT scores in a call center specifically.
Measure and improve CSAT with Dialpad
Dialpad Support for contact centers combines post-call CSAT surveys with AI CSAT and conversation insights, so you can see how customers feel across your calls and act on it.