Customer satisfaction score (CSAT) is a survey measure of how happy customers are with a specific purchase, visit or interaction. It usually comes from a single question such as “How satisfied were you with your delivery?” answered on a scale of 1 to 5, and is reported as the percentage of people who chose 4 or 5.
How CSAT works
The survey is sent straight after a defined moment: an order arriving, a support query being closed, an appointment ending. If 160 customers answer and 128 of them pick 4 or 5, the CSAT is 128 ÷ 160 = 80%.
Some businesses report the average score instead (4.2 out of 5, say), and some use a 1 to 10 scale or a row of faces. None of these is wrong, but they cannot be compared with each other, so pick one method and keep it. A short open question underneath (“What could we have done better?”) is where most of the useful information comes from.
CSAT is often confused with Net Promoter Score. CSAT measures how one interaction went. NPS asks how likely someone is to recommend you overall, so it reflects the whole relationship. A customer can rate a delivery 5 out of 5 and still be unlikely to recommend you because your prices are high. A third measure, customer effort score, asks how easy it was to get something done, and suits support and booking processes.
Why it matters
CSAT is an early warning. A drop in satisfaction after deliveries usually appears weeks before it shows in repeat orders or in churn rate, which gives you time to fix a courier problem or a confusing booking step. It also tells marketing where the experience does not match the promise in the ads, and that gap is expensive, because the money spent winning a customer is wasted if they never come back.
Unhappy customers also leave public reviews, and reviews affect both local search visibility and whether people enquire. There is a UK legal point here. Using CSAT results to decide who gets asked for a public review, so only happy customers are invited, is known as review gating. Google’s review policies prohibit selectively asking for positive reviews, and UK rules on fake and misleading reviews under the DMCC Act 2024 make it a real risk too. Ask every customer, or none.
Common mistakes
- Surveying only customers you expect to be happy. The score becomes meaningless.
- Asking at the wrong time. A survey about a sofa sent before it is delivered measures the checkout, not the purchase.
- Long surveys. Every extra question lowers the response rate and leaves you hearing mostly from the very pleased and the very angry.
- Tracking the number and ignoring the comments.
- Rewarding staff for high scores Which encourages them to ask friendly customers only.
How to act on it
Choose two or three moments that matter most to customers and trigger one short survey after each automatically. Keep it to one rating and one open comment; a post-purchase survey can carry a “how did you hear about us” question too. Read the comments every week, contact anyone who scores 1 or 2 within a day or two, and fix recurring causes. Track the monthly score against repeat purchases and churn so you can see whether it predicts them. Connecting customer feedback to how marketing is planned and judged is part of my digital marketing strategy work.