The moment an organization decides to measure customer experience, it runs into three acronyms repeated in every presentation: NPS, CSAT, and CES. The most common mistake is treating them as different names for the same thing, when in reality each answers a fundamentally different question: would the customer recommend us? Did they like this particular experience? And was it easy to get what they needed?
Choosing the wrong metric for the question produces decisions built on a distorted reading. Measuring overall loyalty after every small transaction annoys customers and corrupts the number; measuring transaction satisfaction once a year misses hundreds of daily signals. This article explains all three metrics with definitions, formulas, strengths, weaknesses, and benchmarks, then shows how to combine them into a single coherent program.
Net Promoter Score (NPS): the loyalty metric
NPS rests on one question: how likely are you to recommend us to a friend or colleague? on a 0-to-10 scale. Answers fall into three groups: promoters (9-10), your loyal advocates; passives (7-8), satisfied but easily poached; and detractors (0-6), at risk of leaving and of damaging your reputation on the way out.
Formula: NPS = % promoters − % detractors, producing a score between −100 and +100. Worked example: out of 200 responses you get 96 promoters (48%), 64 passives (32%), and 40 detractors (20%), so NPS = 48 − 20 = +28.
Strengths: a single simple question, comparability against industry peers, and a documented correlation with future growth in multiple studies, because it measures intent to recommend rather than momentary mood.
Weaknesses: it never tells you why a detractor is unhappy, it is culturally sensitive (in some markets customers rarely give a 10 no matter how pleased they are), it needs a fairly large sample to stabilize, and it is the wrong tool for judging a single transaction.
Benchmarks: above 0 is acceptable, above +30 good, above +50 excellent, and above +70 exceptional, with wide variation across industries; telecoms score far lower on average than luxury retail.
Customer Satisfaction Score (CSAT): the transaction metric
The CSAT question is direct: how satisfied were you with this visit, service, or product? usually on a 1-to-5 scale, the same scale used by the five-button rating devices common at branches and service points.
Formula: CSAT = (number of 4 and 5 responses ÷ total responses) × 100. Example: 500 ratings of which 410 are a 4 or 5 gives a CSAT of 82%.
Strengths: tied to a specific experience that can be fixed immediately, intuitive for both customers and staff, easy to break down by branch, employee, and hour, and a perfect fit for point-of-experience measurement.
Weaknesses: it captures momentary mood and does not predict long-term loyalty; a customer can be satisfied with every visit and still defect to a closer or cheaper competitor without hesitation.
Benchmarks: most service industries target 75 to 85%, and anything below 70% is an alarm that calls for intervention.
Customer Effort Score (CES): the ease metric
The CES question: how easy was it to get your request handled or your problem solved? typically on a 1-to-7 scale, from strongly disagree to strongly agree with a statement like: the company made it easy for me to resolve my issue.
Formula: most commonly the average score, or alternatively the percentage of respondents answering 5 or higher.
Strengths: Gartner's well-known research found that reducing effort predicts loyalty better than delighting customers: 96% of customers who went through a high-effort experience became less loyal, versus only 9% of those with a low-effort one. That makes CES the strongest metric for support channels and procedural services.
Weaknesses: its scope is narrow. It is designed for task-completion and problem-resolution contexts, is meaningless for enjoyment-driven experiences like a restaurant or resort, and says nothing about the overall brand relationship.
The three metrics side by side
| Aspect | NPS | CSAT | CES |
|---|---|---|---|
| Core question | Would you recommend us? | Did you like this experience? | Was it easy? |
| Scale | 0 – 10 | 1 – 5 | 1 – 7 |
| Formula | % promoters − % detractors | % positive responses | Average score |
| Level | Whole relationship | Single transaction | Single service interaction |
| Best timing | Quarterly pulse | Immediately after each experience | After support or a procedure |
| Healthy range | +30 and above | 75% – 85% | 5.5+ out of 7 |
Combining all three into one program
The three metrics are not alternatives; they are complementary layers. The proven practical stack looks like this:
- CSAT continuously at every service point: via instant rating devices or QR codes, serving as your daily pulse and the trigger for same-day service recovery. Satisfaction measurement systems such as RateHex let you track these ratings live down to each branch and employee.
- CES after every support interaction or procedure: a call-center contact, a maintenance request, a registration process, to expose friction points in the customer journey.
- NPS quarterly to a sample of your customer base: detached from any transaction, to measure the overall health and direction of the relationship.
Arranged this way, each metric answers its own question: CSAT shows where today's problem is, CES shows which processes need simplifying, and NPS tells you whether the overall picture is improving or deteriorating. We cover how to build the full system, channels, timing, and bias included, in How to Measure Customer Satisfaction Properly.
A worked example: a clinic chain
Consider a chain of eight clinics running the three-layer stack. Front-desk CSAT averaged 84% overall, but one branch scored 68% specifically in evening hours; investigation revealed understaffed reception in that shift, fixed by rescheduling. CES for phone-based appointment booking scored 4.1 out of 7; analysis showed bookings required multiple call transfers, and after launching WhatsApp booking the score climbed to 5.9 within two quarters. Quarterly NPS rose from +21 to +38 over the year, giving management proof that the operational fixes translated into a stronger overall relationship. Note the sequence: the two operational metrics located the problems, and the strategic metric confirmed the impact of solving them.
Conclusion
There is no single best metric, only the right metric for a specific question: CSAT for the in-the-moment experience and today's fixes, CES for exposing friction in processes and support, and NPS for tracking relationship health over the long run. Start with the metric that serves your nearest decision, layer in the others gradually, and remember that no score accomplishes anything by itself; all of the value lives in the action that follows it.
