What Is CSAT?
CSAT stands for Customer Satisfaction Score — a metric that measures how satisfied a customer was with a specific interaction, transaction, or experience. It is collected through a short post-interaction survey, typically a single question rated on a scale of 1–5 or presented as a simple thumbs up/thumbs down. The score is expressed as a percentage of positive responses out of total responses received.
How CSAT Is Calculated
The CSAT formula is straightforward:
CSAT = (Number of positive ratings ÷ Total ratings received) × 100
What counts as "positive" depends on the scale used:
| Scale | Positive responses |
|---|---|
| 1–5 stars | 4 and 5 |
| 1–10 scale | 8, 9, and 10 |
| Thumbs up / thumbs down | Thumbs up |
| Emoji scale (😞😐😊) | 😊 (satisfied or very satisfied) |
Example: A support team receives 200 CSAT survey responses in a week. 172 are rated 4 or 5 stars. CSAT = (172 ÷ 200) × 100 = 86%.
What Is a Good CSAT Score?
CSAT benchmarks vary by industry, but general guidelines apply across customer support contexts:
| CSAT score | Rating | What it indicates |
|---|---|---|
| 90%+ | Excellent | Consistently strong resolution quality and experience |
| 80–89% | Good | Above average; room for improvement on edge cases |
| 70–79% | Average | Acceptable but below industry leaders |
| 60–69% | Below average | Systemic issues with resolution quality or speed |
| Below 60% | Poor | Urgent action required |
Industry averages (customer support):
- E-commerce: 78–82%
- SaaS / software: 80–85%
- Telecoms: 70–75%
- Financial services: 72–78%
Your target should be at or above your industry average, with a trajectory of improvement over time.
CSAT vs NPS vs CES: What Is the Difference?
These three metrics are often confused. Each measures something different:
| Metric | Question asked | What it measures |
|---|---|---|
| CSAT | "How satisfied were you with this interaction?" | Satisfaction with a specific touchpoint |
| NPS | "How likely are you to recommend us?" | Overall brand loyalty and advocacy |
| CES | "How easy was it to resolve your issue?" | Friction in the support process |
CSAT is the most granular — it measures individual interactions. NPS is the broadest — it captures overall brand sentiment. CES sits in between, measuring process friction at the interaction level.
For customer support teams, CSAT and CES together provide the most complete picture of interaction quality. NPS gives the strategic view of how support performance is affecting brand loyalty over time.
When to Send CSAT Surveys
Timing significantly affects response rates and data quality.
Best practice: Send the survey immediately after ticket resolution — within 5 minutes of closure. Response rates drop sharply after 2 hours. By 24 hours, most customers have moved on and either cannot recall the interaction clearly or are too disengaged to respond.
Channels:
- Live chat: trigger the survey automatically when the chat ends
- Email tickets: include the survey in the resolution confirmation email
- Phone: send a follow-up SMS or email immediately after the call
Response rate benchmarks:
- Live chat surveys: 15–25% response rate
- Email surveys: 10–15% response rate
- In-app surveys: 20–30% response rate
Low response rates are a common problem. The simplest fix is reducing the survey to a single click — "Was your issue resolved? 👍 / 👎" — rather than asking customers to navigate to a separate page or answer multiple questions.
Why CSAT Scores Drop: The Most Common Causes
Slow response times Customers who wait hours for a reply are primed to rate the interaction negatively even if the eventual resolution is correct. Speed is not a separate variable from quality — in the customer's mind, they are the same thing.
First-contact resolution failure Customers who need to contact support multiple times to resolve one issue consistently give lower CSAT scores than customers whose issue is resolved in a single interaction. The FCR-CSAT correlation is well-documented across industries.
Inaccurate or incomplete answers An answer that partially addresses the question, or contains incorrect information, generates a follow-up contact and a low rating. Knowledge base accuracy directly drives CSAT.
Agent tone and empathy Technically correct but cold or dismissive replies generate lower CSAT than warm replies — even when both resolve the same issue. CSAT measures the experience of the interaction, not just the outcome.
Long hold or queue times Customers who wait in a queue before reaching an agent or receiving an AI reply start the interaction in a negative state. Reducing wait times is one of the fastest CSAT improvement levers.
How to Improve CSAT: 8 Proven Tactics
1. Reduce first response time Set and publish response time targets. Use AI to provide immediate acknowledgements and draft replies. Every minute saved on FRT improves the baseline sentiment customers bring to the interaction.
2. Increase First Contact Resolution rate Invest in agent training and knowledge base quality. Agents who have clear, accurate answers at their fingertips resolve issues correctly first time. FCR improvement is the single highest-leverage CSAT driver.
3. Enable AI draft mode AI-generated draft replies help agents respond faster and more accurately. Agents focus on reviewing and approving rather than writing from scratch — which reduces handle time and errors simultaneously.
4. Fix your top 5 reopen topics Every reopened ticket is a failed resolution. Identify the five most common topics that generate reopened tickets and address the root cause — usually a missing knowledge article, an unclear policy, or a process that requires escalation the agent cannot perform.
5. Add empathy to resolution scripts Brief acknowledgement of the customer's frustration before diving into the resolution measurably improves CSAT. "I completely understand how frustrating that must be — let me fix this for you now" outperforms "Here is the solution:" in post-interaction ratings.
6. Proactively notify before customers ask Proactive communication about order status, delays, and resolution progress prevents the frustration that leads to low CSAT. Customers who receive proactive updates before needing to ask give higher ratings than those who have to chase.
7. Personalise where possible Using the customer's name, referencing their specific order or account, and avoiding generic templated language improves perceived care and attention — which flows into CSAT scores.
8. Close the feedback loop For low CSAT scores, have a process to follow up personally. A customer who gave 2 stars and receives a personal follow-up from a team lead often becomes a loyal advocate. The follow-up does not need to offer compensation — acknowledgement and genuine concern are usually sufficient.
Key Takeaways
- CSAT = (positive ratings ÷ total ratings) × 100 — measures satisfaction with a specific interaction
- Industry benchmarks for customer support: 78–85% depending on sector
- Send surveys immediately after resolution — response rates drop sharply after 2 hours
- The biggest drivers of low CSAT are slow response times, first-contact resolution failure, and inaccurate answers
- FCR improvement is the single highest-leverage CSAT driver
- AI draft mode improves CSAT indirectly by reducing handle time, improving accuracy, and freeing agents to focus on empathy
Frequently Asked Questions
What does CSAT stand for? CSAT stands for Customer Satisfaction Score. It is a metric used to measure how satisfied a customer was with a specific interaction — such as a support conversation, a purchase experience, or a product delivery — typically collected immediately after the interaction via a short survey.
What is a good CSAT score for customer support? A CSAT score above 85% is considered good for customer support teams. Scores above 90% are excellent. Industry averages range from 70% in telecoms to 85% in software and SaaS. The most important factor is the trend — a score improving from 76% to 84% over six months indicates healthy progress regardless of the starting point.
How often should I send CSAT surveys? Send a CSAT survey after every resolved interaction. Do not limit surveys to a sample — full coverage gives you the most accurate picture and catches outliers that sampling would miss. Use a single-click format to maximise response rates.
Why is my CSAT score low? The most common causes are slow first response time, unresolved or partially resolved issues requiring follow-up, inaccurate information provided by agents, and cold or unhelpful tone. Audit a sample of low-rated interactions to identify the pattern — most low CSAT scores cluster around a small number of root causes.
Is CSAT the same as NPS? No. CSAT measures satisfaction with a specific interaction. NPS (Net Promoter Score) measures overall brand loyalty and how likely a customer is to recommend your business. They are complementary metrics — CSAT gives interaction-level feedback, NPS gives brand-level feedback. Both are worth tracking for a support team.