What CSAT measures
CSAT captures the quality of one specific interaction, not the overall attitude to the brand. The essential part is that it is asked immediately after the event:
- “How satisfied were you with the checkout process?” — right after checkout
- “Rate your delivery” — 24 hours after receipt
- “How would you rate the reply from our support team?” — after the ticket is closed
The standard scale is 1–5 or 1–10. CSAT is calculated as the share of positive ratings (4–5 out of 5, or 9–10 out of 10) among all responses.
CSAT = (ratings of 4 + ratings of 5) / all responses × 100%
Where CSAT works and where it does not
CSAT is good at diagnosing concrete pain points — if the CSAT for delivery has dropped, something changed in delivery specifically. That makes it a valuable instrument for operational teams.
Its limits: CSAT says nothing about long-term loyalty. A customer can be happy with one purchase and never come back. To measure loyalty and the probability of a repeat purchase, use NPS and retention metrics.
| Metric | What it measures | When to collect it |
|---|---|---|
| CSAT | The quality of a specific event | Immediately after the event |
| NPS | Overall loyalty to the brand | Once a quarter |
| CES | The effort the customer spent solving a task | After support or onboarding |
Common mistakes with CSAT
Asking too late. Three to five days after the purchase, accuracy declines and the response rate falls with it.
Not segmenting the results. An average CSAT across all channels hides the differences: delivery in metropolitan areas versus outside them, mobile versus desktop, new versus returning customers can all give fundamentally different ratings.
Not closing the feedback loop. CSAT collected without action on the negative ratings changes nothing. Set up escalation: ratings of 1–2 should land in the CRM automatically for follow-up.