CSAT vs. CES vs. NPS: Which Customer Metric Actually Predicts Revenue Churn?
CSAT vs. CES vs. NPS: Which Customer Metric Actually Predicts Revenue Churn?
Every executive dashboard tracks customer sentiment. But most leadership teams treat customer metrics like an alphabet soup:
"Our NPS is +42, our CSAT is 88%, and our CES is 2.1... but our net revenue retention just dropped to 91%. Why are customers churning when their satisfaction scores are high?"
The problem is that companies often use the wrong metric to measure the wrong customer interaction.
- CSAT (Customer Satisfaction Score): Measures short-term emotional happiness with a specific event.
- NPS (Net Promoter Score): Measures long-term brand advocacy and likelihood to recommend.
- CES (Customer Effort Score): Measures the operational friction required to accomplish a task.
If you want to forecast contract renewals, spot silent enterprise churn, and prioritize product roadmaps, which metric actually correlates with revenue retention?
Here is the empirical breakdown.
1. CSAT (Customer Satisfaction Score)
The Question:
"How satisfied were you with your recent experience with [Event/Support Ticket]?" Scale: 1 to 5 (Very Unsatisfied to Very Satisfied)
How It's Calculated:
CSAT (%) = (Number of 4 & 5 Ratings / Total Responses) * 100
When to Use CSAT:
- Immediately following a customer support interaction.
- After a product onboarding webinar or training session.
- Following a physical delivery or service call.
The Fatal Blind Spot for Churn:
CSAT measures momentary politeness, not product dependency. A customer can love your polite support agent (CSAT: 5/5), but still cancel their subscription next month because your software lacks an essential API integration. Correlation with Churn: Weak to Moderate (r = -0.28).
2. NPS (Net Promoter Score)
The Question:
"How likely are you to recommend [Company/Product] to a colleague or friend?" Scale: 0 to 10 (Not at all likely to Extremely likely)
How It's Calculated:
NPS = % Promoters (9-10) - % Detractors (0-6)
When to Use NPS:
- Quarterly or semi-annual relationship pulse surveys.
- Identifying brand champions for case studies, G2 reviews, and referral programs.
- Benchmarking overall competitive market position.
The Fatal Blind Spot for Churn:
In B2B SaaS, the person paying the bill (VP of Finance) rarely recommends software to their peers on social media. Many critical enterprise utilities (database backup tools, payroll processing) have mediocre NPS scores, but near-zero churn because they are indispensable infrastructure. Correlation with Churn: Moderate (r = -0.42).
3. CES (Customer Effort Score)
The Question:
"To what extent do you agree with the following statement: [Company] made it easy for me to handle my issue / accomplish my goal." Scale: 1 to 7 (Strongly Disagree to Strongly Agree)
How It's Calculated:
Average CES Score or % High Effort (Scores 1-3)
When to Use CES:
- Immediately after a user completes a complex workflow (exporting data, setting up an integration, upgrading a billing plan).
- After resolving an escalated customer service dispute.
Why CES Is the Strongest Churn Predictor:
Landmark research published in the Harvard Business Review (Dixon, Freeman, & Toman) studying 75,000 customers found that:
Customer loyalty is not driven by delighting customers; it is driven by reducing customer effort.
- 96% of customers who report a high-effort experience become disloyal / churn.
- Customers who report low-effort experiences have a 94% repurchase rate.
When software is hard to use, when integrations break, or when getting a simple question answered requires four back-and-forth emails, customers look for alternatives.
Correlation with Churn: Extremely Strong (r = -0.71).
Comparison Matrix: The 3 Metrics at a Glance
| Dimension | CSAT (Satisfaction) | NPS (Advocacy) | CES (Effort) |
|---|---|---|---|
| Primary Focus | Short-term happiness | Long-term brand loyalty | Operational friction |
| Best Survey Timing | Post-support ticket | Quarterly relationship pulse | Post-workflow completion |
| Scale Type | 1 to 5 Stars / Emojis | 0 to 10 Numeric scale | 1 to 7 Agreement scale |
| Predictive Power for Churn | Low to Moderate | Moderate | Highest |
| Actionability for Product Teams | Low | Low (Requires open probe) | Very High |
The Unified Customer Feedback Architecture
High-performing product organizations don't choose just one; they deploy each metric at its rightful milestone:
Customer Journey:
[ Signs Up ] ──► Completes First Onboarding ──► [ CES Survey: How easy was setup? ]
│
▼
[ Uses Product Daily ] ──► Every 90 Days ──► [ NPS Survey: Likelihood to recommend? ]
│
▼
[ Files Support Ticket ] ──► Ticket Resolved ──► [ CSAT Survey: How was the agent? ]
By deploying short, targeted micro-surveys natively inside your application using platforms like InfiForm, you capture the exact pulse of customer health at every stage of their lifecycle.
Related Articles
CSAT vs. CES vs. NPS: Which Customer Metric Actually Predicts Revenue Churn?
Customer satisfaction scores often fail to predict cancellations. Compare CSAT, CES, and NPS, and learn which metric truly forecasts customer churn.
The Art of the Follow-Up Question: Using Conditional Logic to Probe Bad Ratings
A bad rating without an explanation is useless data. Here is how to use conditional branching logic to gently probe negative feedback and trigger immediate alerts.
A Free Delighted Alternative: Running Unlimited NPS Surveys on Your Own Domain
Delighted charges $224/mo for basic multi-channel NPS surveys. Here is how to run beautiful, unlimited Net Promoter Score surveys on your own domain for free.