Customer Satisfaction Score (CSAT) measures how satisfied a customer is with a specific interaction or purchase. You ask "How satisfied were you with [the interaction]?" on a 1 to 5 scale, then report the percentage of respondents who answered 4 or 5. A CSAT of 75% to 85% is a good score in most industries.
Key takeaways
- CSAT is a transactional metric: measure it right after a touchpoint, not quarterly.
- The score is the share of positive responses (4s and 5s), expressed as a percentage.
- Benchmarks differ by industry; US cross-industry averages sit in the mid 70s according to ACSI.
- The score tells you where satisfaction drops. The open feedback next to it tells you why.
What is CSAT and how do you measure it?
Customer Satisfaction Score is the most direct customer experience metric there is: you ask customers how satisfied they were, immediately after an experience. Because it is tied to one touchpoint, it localises problems precisely, which makes it the workhorse metric for operational teams. Measurement is a two-step: pick the touchpoints that matter (support contact, delivery, checkout, renewal), then trigger a short survey at each one through email, SMS or in-app via an omnichannel setup. Response rates drop fast with survey length, so keep it to the scored question plus one open question. More on that trade-off in our piece on survey response rates.
When should you send a CSAT survey?
CSAT surveys belong immediately after the interaction they measure, while the memory is specific: within hours of a support contact, within a day of a delivery, straight after checkout for the buying flow. Beyond timing, frequency discipline decides data quality. Cap how often one person can be surveyed (a 30-day quarantine is a workable default), spread measurement across the customer base instead of blasting everyone quarterly, and keep each survey to two questions. Length is not a matter of taste: cross-platform completion research shows surveys of 4 to 8 questions completing at around 65%, dropping to roughly 42% at 15 or more questions, and SurveyMonkey's completion-time research shows abandonment climbing sharply past the 7 to 8 minute mark. A two-question CSAT survey takes under a minute, which is exactly why it can run continuously without burning goodwill.
Which CSAT question should you ask?
The CSAT question names the interaction and asks for a rating of it. Examples that work in practice:
- How satisfied were you with your store visit today?
- How satisfied are you with the handling of your support request?
- How satisfied are you with your delivery?
- How satisfied are you with the ease of ordering?
Always follow with one open question: "What is the main reason for your score?" That single question converts a monitoring tool into an improvement tool, because it captures the cause in the customer's own words. Avoid the classic wording traps: no double-barreled questions ("How satisfied are you with our speed and friendliness?" produces unusable answers), no leading framing ("How much did you enjoy..."), and name the specific interaction rather than asking about "your experience with us" in general, which customers answer with whatever happened most recently anywhere.
How do you calculate your CSAT score?
The CSAT formula is: (number of positive responses ÷ total responses) × 100. Positive means 4 or 5 on a 5-point scale. If 550 of 1,000 respondents answer 4 or 5, your CSAT is 55%. Report it per touchpoint and per period rather than as one blended company number; a strong store score can mask a failing delivery score, and the blended average hides exactly the thing you need to see.
What is a good CSAT score in 2026?
A good CSAT score sits between 75% and 85% for most sectors, meaning at least three of every four customers rate the experience positively. Where you should sit depends on your industry. The American Customer Satisfaction Index tracks satisfaction across 40+ sectors; the pattern in recent data:
| Industry | Typical CSAT range |
|---|---|
| Full-service restaurants, consulting | 82 to 84% |
| E-commerce and retail | 78 to 82% |
| Banks and insurance | 78 to 81% |
| Telecom and utilities | 72 to 76% |
| Social media platforms | around 73% |
Treat these as orientation, and treat your own trend as the real benchmark: a stable 80% that drops to 74% after a process change is a louder alarm than any industry table.
How do you improve your CSAT score?
CSAT scores improve when you fix the causes behind low ratings, so the work starts in the open feedback. At any real volume, reading every comment manually stops scaling; AI analysis like ISAAC groups verbatims into themes (waiting time, staff friendliness, product availability, pricing) and shows which theme drags the score per location or team. From there, suggested actions and coaching on real feedback turn the finding into behaviour change. Closing the loop on low scores compounds the effect: CustomerGauge's close-the-loop research links follow-up within 48 hours to double-digit retention gains among responders. Teams choosing tooling for this can compare options in our 10 best customer feedback software guide.
Why is customer satisfaction stagnating, and why is that your opportunity?
Customer satisfaction at the macro level has been flat for years. The American Customer Satisfaction Index closed 2025 at 76.9, down 0.5% on the year and, by ACSI's own account, without material improvement since 2017. Forrester's 2025 Customer Experience Index tells the same story more sharply: perceived CX quality in the US fell for a fourth consecutive year to an all-time low, with 25% of evaluated brands declining and only 7% improving. For a mid-market company the reading is straightforward: the average is beatable. Most competitors collect satisfaction data and change nothing with it. A team that closes the loop on low scores and fixes recurring themes moves against a flat field, which makes CSAT improvement a competitive weapon rather than housekeeping.
What are the limitations of CSAT?
CSAT has three known blind spots. It is short-term: a customer can be satisfied with every individual interaction and still defect over price or a better competitor offer, which is why it needs NPS beside it for the relationship view. It is mood-sensitive: scores dip after price increases, in peak-season queues, and in bad weather for delivery businesses, so read trends over weeks rather than reacting to single days. And it under-predicts loyalty compared with effort: Gartner's research found customer effort roughly 40% more accurate at predicting loyalty than satisfaction measures, which is why CES belongs at process-heavy touchpoints. None of this argues against CSAT. It argues against CSAT alone: the metric is unbeatable at localising problems, provided something acts on what it finds.
How do you report CSAT to management?
CSAT reporting earns attention when it answers business questions instead of listing scores. A monthly management view needs four elements: the score per journey step with its trend (delivery at 81% and falling beats "company CSAT 79%"), the three themes currently dragging scores with their share of negative feedback, the actions taken on last month's themes and what happened to the score afterwards, and the money link where you can make it, such as repeat-purchase rate of satisfied versus dissatisfied customers from your own data. That last element is the one that changes budget conversations, and it is exactly the connection Forrester's 2025 research found most teams unable to make, with only around half linking CX metrics to business outcomes. Present the four elements on one page, with drill-down available for the questions that follow, and the CSAT review stops being a ritual.
When should you use CSAT instead of NPS or CES?
Use CSAT when you want to evaluate a specific interaction, CES when the question is how effortless a process was, and NPS when you want relationship-level loyalty. They complement each other: NPS tells you the customer is drifting, CSAT and CES tell you which touchpoint is doing the damage. How to combine all three in one programme is covered in our voice of customer survey guide.
FAQ about CSAT
What does CSAT stand for?
CSAT stands for Customer Satisfaction Score. It expresses the percentage of customers who rate a specific interaction positively, usually 4 or 5 on a 5-point scale.
What is the difference between CSAT and NPS?
CSAT measures satisfaction with one interaction; NPS measures loyalty to the company as a whole. CSAT is asked after a touchpoint, NPS periodically across the relationship.
Is 70% a good CSAT score?
A 70% CSAT is slightly below the 75% to 85% range considered good in most industries, but context matters. In telecom or utilities 70% is near average, while in consulting it would signal a problem.
How often should you measure CSAT?
Continuously, triggered by interactions, rather than in periodic campaigns. Event-triggered measurement catches problems within days and spreads survey pressure across the year.
Can you measure CSAT without annoying customers?
Yes, by keeping surveys to two questions, triggering them only after meaningful interactions, and capping how often one person can be surveyed. Feedback quality beats feedback volume.
What is a good response rate for a CSAT survey?
Well-timed two-question transactional surveys typically see 10 to 30%, higher in strong relationships and in-app contexts. Falling response rates usually mean surveys are too frequent or visibly lead to nothing.
Can you use a 7-point or 10-point scale for CSAT?
Yes, as long as you define which answers count as positive and never change scale mid-programme. The 5-point scale remains the default because it reads instantly on mobile and keeps cross-touchpoint comparison simple.
---