Customer Satisfaction Score
Definition
Customer Satisfaction Score
Customer satisfaction score is the rating customers give after a specific interaction, usually on a five-point scale reported as a percentage. It is the fastest feedback loop a service team has, and it is measured per contact rather than once per quarter.
The speed is the value. A score arriving an hour after the call can be coached on the same shift, which no quarterly programme can match.
Speed also creates the weakness. Response rates are low and skewed toward the delighted and the furious, so small samples swing wildly.
Key takeaways
- Customer satisfaction score divides satisfied responses by total responses received.
- The scale and the cut-off both change the headline number, so both must be published.
- Response rates of 10–20% are normal, which makes small samples unreliable.
- Verbatim comments explain the score; the number alone rarely does.
How it works
Customer satisfaction score is calculated by dividing the number of satisfied responses by the total number of responses, then multiplying by 100. What counts as “satisfied” is a design decision that changes the result substantially.
The formula is: (satisfied responses ÷ total responses) × 100.
Two design choices dominate. The scale sets how many options a customer gets, and the cut-off sets which of those options count as satisfied.
| Scale | Common satisfied cut-off | Effect on the score |
|---|---|---|
| 2-point (yes/no) | Yes | Highest, least sensitive |
| 3-point | Positive only | High, blunt |
| 5-point | Top two of five | Industry default |
| 5-point top-box | Top one of five only | Lowest, most demanding |
Top-box scoring is the strictest reading and the most useful for spotting drift. Switching from top-two to top-box can drop a reported score by 20 points without service changing at all.
Timing changes the answer too. A survey sent immediately after resolution scores higher than one sent three days later, so send timing has to stay fixed.
Response bias is the standing risk. Neutral customers rarely reply, which pushes the score toward the extremes and makes month-to-month movements look larger than they are.
Read it beside outcomes rather than alone. Scores track closely with first contact resolution, and a satisfaction dip usually has a resolution cause.
The mechanics are covered in more depth under customer satisfaction rating, which is the same metric under its common acronym.
Collection method shapes the sample. A post-call survey offered in the same channel gets far better response than an emailed link sent the next day.
Customer evidence should drive what you measure. The American Society for Quality describes the voice of the customer as engagement with customers and stakeholders to understand their needs, which should sit central to strategy and key performance indicators — see ASQ.
National benchmarks give context for the level. The American Customer Satisfaction Index reports satisfaction quarterly and published its latest reading for Quarter 2, 2026 — see ACSI.
Coach on verbatims, not on the number. The comment field tells an agent what to do differently; the percentage tells them only that something went wrong.
Examples
Satisfaction scoring differs by channel, by when the survey is sent, and by how demanding the satisfied cut-off is. Five cases show how teams design the measure and what they do with the result.
Contact centres survey immediately after the call. Same-channel prompts collect enough responses to make agent-level reporting defensible rather than anecdotal.
Retailers survey after delivery, not after purchase. Because the parcel is the experience, asking at checkout measures the wrong moment entirely.
Software companies survey after support tickets and separately in-product. The two scores routinely diverge, and the gap is usually the most useful finding.
Public services survey on ease as well as satisfaction. A citizen can be satisfied with an outcome and still describe the process as painful — which is exactly the signal worth acting on.
Outsourced providers report agent-level and queue-level scores. Contracts increasingly specify the scale and the cut-off, because switching either one changes the reported result.
Related terms
Customer satisfaction score is one of several customer-side measures, each answering a different question. The terms below cover the acronym form, the alternatives, and the outcome metrics it tracks with.
- Customer Satisfaction Rating (CSAT): the same metric under its widely used acronym.
- Net Promoter Score (NPS): the advocacy measure asked about the brand rather than one interaction.
- Customer Effort Score: the ease measure that often predicts loyalty better than satisfaction.
- Caller Satisfaction: the voice-channel expression of the same idea.
- Post-Call Survey: the collection method that produces most contact-centre responses.
- First Contact Resolution: the outcome metric satisfaction tracks most closely.
- Quality Assurance: the internal review that validates whether scores reflect real performance.
FAQ
How is customer satisfaction score calculated?
Divide satisfied responses by total responses and multiply by 100. Publish the scale and the satisfied cut-off alongside it.
What is a good customer satisfaction score?
Consumer service queues commonly sit in the 75–85% range on a top-two cut-off, with top-box scoring running well below that.
What response rate should we expect?
Ten to 20% is typical for post-interaction surveys, and higher if the prompt stays in the same channel.
Why did our score drop without service changing?
Check the scale, the cut-off, the send timing, and the sample size before assuming performance moved.
Is it the same as CSAT?
Yes, CSAT is simply the common acronym for the same measure.
Should agents be ranked on it?
Only at reasonable sample sizes, and always alongside quality review and resolution data.
Source partners building satisfaction measurement into delivery can compare models across Outsource Accelerator hubs.
Independent
