CSAT, NPS or CES: which should you use?
Use CSAT to measure how a specific interaction went, CES to measure how easy it was to get something done, and NPS to measure how the overall relationship is going. Most consumer brands need one relationship metric (usually NPS) and one moment metric (CSAT or CES) at the touchpoints that drive revenue. The metric matters less than what you do with it: a score only earns its place when it's linked to behaviour like repeat purchase and spend per visit, and reviewed where trading decisions get made.
Customer experience teams often argue about which metric is "best". It's the wrong question. CSAT, NPS and CES answer three different questions, at three different points in the customer journey. Pick the one that answers the question you actually need answered, at the moment you need the answer.
CSAT: customer satisfaction score
What it measures: how satisfied a customer was with one specific interaction, such as a purchase, a delivery, a store visit or a call with your service team.
The question: "How satisfied were you with [the interaction]?" on a 1–5 scale, from very dissatisfied to very satisfied.
How to calculate it: divide the number of satisfied responses (4 or 5) by the total number of responses, then multiply by 100. If 170 of 200 customers answer 4 or 5, your CSAT is 85%.
Use it when: you want fast, specific feedback on a touchpoint, straight after it happens. It's easy for customers to answer and easy for teams to act on.
Watch out for: CSAT is short-term. A customer can be satisfied with today's delivery and still quietly drift to a competitor. On its own, it won't tell you whether they'll come back.
NPS: Net Promoter Score
What it measures: loyalty and likely word of mouth, across the whole relationship rather than one moment.
The question: "How likely are you to recommend us to a friend or colleague?" on a 0–10 scale.
How to calculate it: group responses into promoters (9–10), passives (7–8) and detractors (0–6). NPS is the percentage of promoters minus the percentage of detractors, giving a score between −100 and +100. If 50% are promoters and 20% are detractors, your NPS is +30. It is never an average of the raw scores.
Use it when: you want a periodic read on the health of the relationship, for example quarterly, and a way to compare segments, channels or markets over time.
Watch out for: NPS is a lagging, high-level number. It tells you something has changed, not why. Response styles also vary by market and channel, so compare like with like, and always ask a follow-up question.
CES: customer effort score
What it measures: how easy it was for a customer to get something done.
The question: agreement with "[Company] made it easy for me to handle my issue" on a 1–7 scale, from strongly disagree to strongly agree.
How to calculate it: report the average score, or the percentage of customers who agree (5–7). Pick one method and stick to it, so the trend means something.
Use it when: the experience is a job to be done rather than a pleasure: returns, delivery problems, account changes, finding the right product, checkout. Effort is often what turns a fixable problem into a lost customer.
Watch out for: CES says little about delight or brand. A low-effort experience can still be forgettable.
Which metric fits which moment
- After a purchase, delivery or store visit: CSAT. You want to know how that moment went.
- After a return, complaint or service request: CES. You want to know how hard you made it.
- Every quarter, across your customer base: NPS. You want to know whether loyalty is rising or falling.
- Across all three: one open question, "What's the main reason for your score?" The comments explain the number, and that's where the fixes are.
Start with the moments of truth that drive the most revenue, and measure those first. Measuring everything at once usually means acting on nothing.
Five mistakes that make these metrics useless
- Chasing the score, not the cause. Teams learn to ask for a 10. The number rises; the experience doesn't.
- Surveying too often. Survey fatigue lowers response rates and skews results towards your most and least happy customers.
- Averaging NPS. NPS is promoters minus detractors. An average of 0–10 scores is a different, less useful number.
- Never closing the loop. If detractors and low-effort scores don't trigger a follow-up and a fix, you're collecting complaints, not acting on them.
- Keeping the metric in the CX team. If the score isn't reviewed where commercial decisions are made, it won't change any.
How to tie customer experience metrics to revenue
The point of measuring customer experience is to grow the business. So link each metric to a commercial behaviour: do customers with high CSAT at delivery come back more often? Do customers who found a return easy spend more next time? Does a rise in NPS in a region show up in footfall and retention?
Then put the numbers in front of the people who trade. Customer metrics belong in the same weekly meeting as sales, margin and stock, so experience is traded off deliberately against cost rather than discussed separately and ignored.
Not sure which capability is holding your customer experience back? Our free CX Maturity Assessment scores you across strategy, insight, journey design, culture, technology and measurement in about three minutes. Or see how our customer experience consultancy turns experience data into commercial decisions.