NPS vs CSAT vs CES: which customer metric should you use?
NPS measures long-term loyalty ('How likely are you to recommend us?', scored −100 to +100), CSAT measures satisfaction with a specific interaction ('How satisfied were you?', scored 0–100%), and CES measures how easy it was to get something done ('How easy was it to resolve your issue?'). They answer different questions, so most teams use CSAT or CES for transactional moments and NPS for relationship health — not one instead of the others.
What does each metric actually measure?
All three are single-question survey metrics, but they point at different things. NPS (Net Promoter Score) asks about the future — would this person recommend you — which makes it a proxy for overall loyalty. CSAT (Customer Satisfaction Score) asks about the recent past — were you satisfied with this specific purchase, ticket, or visit. CES (Customer Effort Score) asks about friction — how hard did the customer have to work to get what they needed.
How do NPS, CSAT, and CES compare?
| Dimension | NPS | CSAT | CES |
|---|---|---|---|
| Question | How likely are you to recommend us? | How satisfied were you with [interaction]? | How easy was it to [complete task]? |
| Scale | 0–10 | Usually 1–5 (sometimes 1–7 or 1–10) | Usually 1–5 or 1–7 (very difficult → very easy) |
| Formula | %promoters (9–10) − %detractors (0–6) | Satisfied responses (top two boxes) ÷ total × 100 | Average score, or % choosing 'easy' ratings |
| Result range | −100 to +100 | 0% to 100% | Depends on scale (e.g., 1.0–7.0 average) |
| Measures | Loyalty and word-of-mouth potential | Satisfaction with a specific moment | Friction in a specific task |
| Best timing | Quarterly or after major milestones | Right after a purchase, ticket, or visit | Right after support contact or task completion |
| Horizon | Relationship (long term) | Transaction (short term) | Transaction (short term) |
When should you use NPS?
Use NPS when you want a recurring health check on the overall relationship. It works best on a fixed cadence — quarterly, biannually, or after major lifecycle moments like onboarding completion or renewal. Because it asks about recommendation, it captures sentiment about the whole experience: product, price, support, and brand combined. That breadth is also its weakness — when NPS drops, the score alone never tells you why, so always pair it with an open-ended follow-up.
When should you use CSAT?
Use CSAT immediately after a specific interaction you want to monitor: a support ticket closing, a delivery arriving, a training session ending. Because it names the interaction, the feedback is precise and easy to act on. CSAT = number of satisfied responses (typically 4s and 5s on a 5-point scale) ÷ total responses × 100. For example, 150 satisfied responses out of 200 total = 75% CSAT. It is the most flexible of the three — you can run CSAT on almost any touchpoint.
When should you use CES?
Use CES when friction is the thing you are trying to reduce — most commonly after support interactions or self-service flows. The underlying insight is that customers churn less when things are easy, not necessarily when they are delighted. A typical CES question is 'The company made it easy for me to handle my issue' with an agree/disagree scale, or 'How easy was it to resolve your issue today?' from very difficult to very easy. Low CES scores point directly at broken processes: confusing docs, repeated contact, long waits.
Should you use more than one?
Yes — they are complementary, not competing. A common setup for a small team:
- CSAT after every closed support ticket (continuous, automated)
- CES after self-service or checkout flows you are actively redesigning
- NPS quarterly to the whole customer base, with a 'why' follow-up question
- One shared dashboard so you can see whether transactional wins move the relationship score
Never combine two metrics in one survey invitation. A one-question CSAT gets answered because it takes five seconds; adding NPS and CES to the same email turns it into 'a survey' and response rates drop. Keep transactional surveys to one scored question plus one optional comment box.
Which metric is best for a small team?
If you can only run one, start with CSAT. It is the easiest to interpret, the fastest to act on, and it does not require a large sample to be meaningful. Add NPS once you have enough customers for a quarterly cadence to produce a stable score. Tools-wise, you do not need a dedicated platform for any of these — Formkii's free survey maker has rating and NPS question types with real-time charts, which covers all three metrics without a paid plan.
Frequently asked questions
What is the main difference between NPS and CSAT?
NPS measures long-term loyalty by asking how likely someone is to recommend you, scored from −100 to +100. CSAT measures satisfaction with one specific interaction, scored as a percentage from 0 to 100. NPS is a relationship metric; CSAT is a transactional metric.
What is a good CES score?
There is no universal CES benchmark because companies use different scales and wordings. As a practical rule, you want the large majority of respondents choosing the 'easy' end of the scale, and you should track your own average over time rather than comparing to published numbers.
Can you convert CSAT to NPS?
No. They use different questions, different scales, and different formulas, so there is no valid conversion between them. A customer can be satisfied with a single interaction (high CSAT) and still unwilling to recommend you overall (low NPS).
How often should you send NPS vs CSAT surveys?
CSAT and CES are event-triggered — send them immediately after the interaction they measure. NPS is cadence-based — quarterly is the most common rhythm, with each customer surveyed at most a few times per year to avoid fatigue.