NPS vs. CSAT vs. CES at a glance
Use this comparison to choose a starting point. The right metric is the one that makes the next team decision clearer, not the one with the most familiar benchmark.
| Metric | What it helps answer | Typical question and scale | Best time to ask |
|---|---|---|---|
| NPS | How does the customer view the broader relationship? | Likelihood to recommend, often 0 to 10 | A periodic relationship check-in |
| CSAT | How did a specific interaction feel? | Satisfaction with a recent moment, often 1 to 5 | Immediately after the interaction |
| CES | How easy was it to complete a defined task? | Ease or effort for one task, often 1 to 7 | Immediately after the task |
Start with the decision you need to make
Choose a metric after deciding what the team needs to learn. A relationship question, a just-completed experience, and the effort required to complete a task are different situations. Matching the metric to the decision keeps the result easier to explain and act on.
NPS: relationship and recommendation
Net Promoter Score asks how likely someone is to recommend a company or product, typically on a 0-to-10 scale. It is most useful for a broader relationship question rather than a single service interaction. Follow the rating with an open question that asks why the respondent chose it.
CSAT: the experience just completed
Customer Satisfaction is a direct rating of a recent interaction such as checkout, support, delivery, or setup. Teams often use a 1-to-5 scale and report the share of satisfied responses. Keep the wording tied to the event so the result has a clear operational meaning.
CES: effort in a specific task
Customer Effort Score focuses on how easy or difficult it was to complete a task. It can help teams examine support resolution, onboarding, account changes, or another defined action. Use the same direction and scale each time, so a change in the score does not reflect a changed question.
Do not use one metric as a substitute for another
A positive NPS result does not prove a recent checkout was easy, and a good CSAT score after support does not describe the overall relationship. Use a single metric when it fits the question, or use more than one only when each score has an owner and a defined next action.
Use the score with the story
A metric is a signal, not a diagnosis. Pair the rating with a short comment, the touchpoint, and relevant context such as location or channel when that information is appropriate and expected. The follow-up detail is what helps a team distinguish a recurring issue from an isolated event.
Compare consistently
If you compare results over time, keep the wording, scale, audience, and timing consistent. Document how you calculate each score, especially the definitions of a satisfied response or an effort scale. Avoid importing industry benchmarks without checking how they were collected and whether the populations are comparable.