TL;DR
Customer Satisfaction Score (CSAT) measures satisfaction with a specific interaction or feature on a 1–5 or 1–7 scale, calculated as % of respondents selecting the top-2 box. Best-in-class support CSAT: 90%+; product-level CSAT: 75–90%. Best for interaction-level measurement, not overall loyalty (use NPS for loyalty).
What is CSAT?
Customer Satisfaction Score (CSAT) is a transactional satisfaction metric that asks customers to rate a specific interaction, feature, or product experience — typically on a 1–5 or 1–7 scale. The score is calculated as the percentage of respondents selecting the top-2 box (4–5 of 5, or 6–7 of 7).
CSAT is most often used post-support-ticket ("How satisfied were you with this support interaction?"), post-onboarding ("How satisfied were you with onboarding?"), or post-feature-launch ("How satisfied are you with the new dashboard?"). It is the most-deployed customer-feedback metric in SaaS, ahead of NPS and CES, because the question is concrete and the response window is short.
Unlike NPS (relationship loyalty, measured 1–2× per year) and CES (effort-based, measured per task), CSAT is interaction-focused and high-frequency — typically triggered after every customer touchpoint that matters. It is the right metric for measuring whether a specific moment in the customer journey is working.
Why CSAT matters
For support and customer success teams, CSAT is the operational metric. A support team with 92% CSAT is delivering quality; a team at 78% has a process or staffing problem. Because CSAT is collected immediately after the interaction, it provides a fast feedback loop that NPS cannot — issues surface in days, not quarters.
For product teams, feature-level CSAT post-launch identifies which features need iteration. A new dashboard that ships with 65% CSAT is a signal to talk to detractors before the next sprint; a feature at 88% CSAT is a signal to invest in adjacent surfaces. Without CSAT, product teams rely on usage metrics that confirm adoption but not satisfaction.
For RevOps and CS leaders, CSAT is one input into customer health score — alongside product usage, billing, and engagement signals. It is not, however, sufficient on its own as a retention predictor: high CSAT correlates with retention, but it doesn't cause it (the customer can be satisfied with support and still churn because of product gaps or budget cuts).
CSAT formula
CSAT (%) = (Top-2 box responses / Total responses) × 100 Example: post-support ticket survey - 200 respondents - 5-point scale: very dissatisfied → very satisfied - 175 responded 4 or 5 (satisfied or very satisfied) - 15 responded 3 (neutral) - 10 responded 1 or 2 (dissatisfied) CSAT = 175 / 200 = 87.5% Variants: - 5-point scale (most common): top-2 box = 4–5 - 7-point scale: top-2 box = 6–7 - "How satisfied?" question is standard; avoid leading variants
Benchmarks
| Use case | Best-in-class | Median | Below average |
|---|---|---|---|
| Support interaction | 92–97% | 82–90% | <80% |
| Product-level CSAT | 85–92% | 70–82% | <65% |
| Onboarding CSAT | 80–90% | 65–75% | <60% |
| Feature launch CSAT | 75–88% | 60–72% | <55% |
| B2B SaaS overall | 85–92% | 75–85% | <70% |
| B2C SaaS overall | 80–90% | 65–78% | <60% |
Benchmarks compiled from Zendesk CX Trends 2025, Salesforce State of Service 2025, and HubSpot Service Hub Benchmarks 2025.
Common mistakes
- Confusing CSAT with NPS. CSAT measures interaction satisfaction; NPS measures relationship loyalty. A customer can have 95% CSAT on support and still churn because of product gaps. Use both for different decisions, never substitute one for the other.
- Survey fatigue. CSAT triggered on every customer touch destroys response rates within 6 months. Triage to the moments that matter (support resolution, onboarding completion, key feature launches) and skip the rest.
- Reading the score without the verbatims. The number tells you direction; the open-ended follow-up explains why. CSAT without verbatims is a vanity metric.
- Mixing 5-point and 7-point scales. Top-2 box on a 7-point scale is harder to hit than on a 5-point scale — they're not comparable. Pick one and use it consistently across surveys.
- Leading questions. "How happy are you?" yields higher scores than "How satisfied are you?". Use the standard "How satisfied?" phrasing to enable cross-time and cross-team comparison.
Related metrics
CSAT sits inside the customer-loyalty stack alongside NPS (relationship), CES (effort), and customer health score (composite). Retention outcomes: churn rate, gross retention, NRR.
At a glance
- Category
- Operations / Cash
- Related
- 3 terms
Frequently asked questions
What is a good CSAT score?
For support interactions, 90%+ is best-in-class, 82–90% is median, below 80% indicates a process or staffing problem. Product-level CSAT typically runs 10–15 points lower. Always benchmark within your specific use case — support CSAT is not comparable to product CSAT or onboarding CSAT.
What's the difference between CSAT and NPS?
CSAT measures satisfaction with a specific interaction (post-support, post-onboarding). NPS measures overall relationship loyalty via a recommendation question. CSAT is high-frequency and operational; NPS is low-frequency and strategic. Use both — they answer different questions.
How often should you measure CSAT?
Immediately after the events that matter — support ticket resolution, onboarding completion, key feature launches. Avoid triggering on every customer touch (survey fatigue destroys response rates). Best practice: 2–4 CSAT triggers per customer per year, on moments that drive churn or expansion.
Is a 5-point or 7-point CSAT scale better?
5-point is more common and easier for respondents; 7-point gives more granular signal but reduces response rates. Pick one, document it, and stick with it. 5-point with clear labels (very dissatisfied → very satisfied) is the most defensible default for SaaS.
Can CSAT predict customer churn?
Low CSAT correlates with churn but doesn't cause it — high-satisfaction customers can still churn for budget or product-gap reasons. Pair CSAT with NPS, product usage, and billing signals in a composite customer health score for actual churn prediction. Standalone CSAT-based churn models miss 40–60% of actual churn.
Sources
- Zendesk. CX Trends 2025, 2025. zendesk.com
- Salesforce. State of Service 2025, 2025. salesforce.com
- HubSpot. Service Hub Benchmarks 2025, 2025. hubspot.com
Fairview integrates CSAT with product usage and NPS into one customer health composite — see the operating intelligence overview for the broader category.
Definitions and benchmarks reviewed by Siddharth Gangal, Founder, Fairview.
See it in Fairview
Track Customer Satisfaction Score (CSAT) automatically.
25-minute live demo. Tailored to your stack. See your operating view live.