NPS, CSAT and CES: Three Different Customer Questions

Ask the question the decision needs

NPS vs CSAT vs CES is often presented as a contest between three customer scores. The useful question is which customer experience you need to understand. Recommendation, satisfaction with an interaction and perceived effort are related, but they are not interchangeable. A customer can receive polite support, find the process exhausting and still recommend a product they love. This guide explains the measures, the calculations and the survey choices that make comparisons fair. It also shows why a score needs the words behind it before it becomes a sensible brief for marketing or customer communication. Explore our marketing frameworks and measurement guide for the wider planning context.

NPS measures a recommendation response

The familiar Net Promoter Score question asks how likely someone is to recommend the business on a zero-to-ten scale. Responses of nine or ten are promoters, zero to six are detractors, and seven or eight are passives. NPS equals the percentage of promoters minus the percentage of detractors among valid responses. Passives remain in the denominator. The result ranges from minus 100 to plus 100 and is expressed as a score, not a satisfaction percentage. Bain describes this calculation as part of its Net Promoter System. ¹ Keep the exact question, respondent population and survey timing beside the result.

CSAT describes satisfaction with the specified experience

Customer satisfaction measurement asks about the experience named in the question. A commonly used five-point method calculates the percentage selecting satisfied or very satisfied, typically scores four and five, among valid responses. ² Other scales and calculation choices exist, so a reported CSAT number needs its scale and top-box rule. Satisfaction with a delivery is different from satisfaction with the overall company relationship. If one survey arrives after purchase and another after a difficult complaint, their populations will differ. Do not interpret the difference as a change in service without checking who was asked and when.

CES describes effort under a stated scale

Customer Effort Score asks how easy or difficult the specified task felt, or uses an agreement statement about ease. Implementations vary in scale and direction. State the wording, anchors and whether a higher score means easier or harder. A mean score is the sum of valid response values divided by valid responses; a top-box percentage is a different calculation. Do not mix them in one trend. Effort can be useful after a support interaction, account setup or return. It still describes reported experience, rather than proving that the underlying process improved for every customer. If that is your next question, read our guide to customer retention rate.

Read the comments before writing the campaign

A low score might reflect slow delivery, misleading expectations, a complex form or a product defect. Each needs a different response. Code open-text feedback with a consistent method and inspect examples before counting themes. If automated analysis helps, review its classifications and preserve the distinction between what a person wrote and what a model inferred. Do not label all negative comments as disloyalty. Some customers care enough to describe a fix. Connect recurring themes with operational records, such as resolution time or delivery failure, then ask which communication or process change would address the issue. In any NPS vs CSAT vs CES comparison, the comments explain what the number cannot. Conversion rate optimisation is explained in a separate guide.

Close the loop without pressuring the respondent

Where appropriate and consented, follow up with customers who describe an unresolved problem. The purpose is to understand and resolve it, not persuade them to revise a score. Give service teams the authority and information to respond. At an aggregate level, share what the business learned and what changed. Measure the result of those changes using comparable survey rules and operational outcomes. A score target alone can encourage selective invitations or coached answers. Protect the measurement from those incentives. The customer should be able to answer honestly without believing that a worker will be punished for the response. For the wider picture, see our guide to employer branding metrics.

Keep the invitation process visible

Maintain a survey log showing when eligible customers were invited, through which channel and after which experience. Define how repeat invitations are suppressed and how incomplete responses are handled. A customer who receives a survey only after a successful resolution belongs to a selected population; the resulting score cannot describe every support case. Likewise, an invitation sent by a service representative may affect who answers and what they feel safe saying. Record these conditions and keep them stable for trend comparisons. If you change the survey, run a transition study where feasible or clearly mark the break in the series. Give colleagues a few anonymised comments alongside the score so they can see what customers meant. A stable method makes a modest score more useful than a spectacular number collected through selective invitation.

Choose a small useful scorecard

A support journey might combine task completion, resolution time, CES and complaint themes. A relationship review might use NPS alongside retention and observed behaviour. A delivery review might use CSAT with on-time performance and damaged-order rates. Quiamo can help translate those findings into clearer expectation setting, service messages and customer communication. The brief should name the experience being improved and the evidence that would support the decision. Keep the score and the underlying records accessible to the people responsible for the change. An attractive dashboard should not separate the team from the customer’s actual difficulty. See how our customer communication service can support the communication and reporting work described here.

Questions people ask

Use scores as evidence not verdicts

NPS vs CSAT vs CES becomes a useful comparison when each measure has a specific purpose, stable wording and a visible respondent base. Read scores with customer comments and operational evidence, and avoid turning one survey into a claim about all customers. If your team collects feedback but struggles to turn it into a communication change, bring Quiamo the questions and the recurring themes. We can scope a review that connects the customer experience you are measuring with the information and service choices people need. Contact Quiamo to discuss the measurement and communication scope.

Sources

1. Bain Net Promoter Score calculationPromoter detractor and passive categories and NPS formula
2. Qualtrics CSATFive-point top-two-box CSAT calculation
3. Qualtrics Customer experience benchmarkingDifferent customer experience measures and mean CES approach

Worked examples on this page are fictional and labelled. Search phrases and figures from external sources are cited; nothing here describes a Quiamo client result.