What Is CSAT vs NPS vs CES?

CSAT, NPS, and CES are the three most common customer experience metrics, each measuring something genuinely different: CSAT (Customer Satisfaction) measures satisfaction with one specific interaction, NPS (Net Promoter Score) measures overall loyalty and likelihood to recommend, and CES (Customer Effort Score) measures how much effort it took to get something done. Using the right one for the right situation — and understanding that they're not interchangeable — matters more than picking a single "best" metric.

Quick facts

• CSAT: "How satisfied were you with [X]?" — measures satisfaction with one specific interaction or transaction.
• NPS: "How likely are you to recommend us?" — measures overall loyalty and brand sentiment. See full NPS guide.
• CES: "How easy was it to [accomplish X]?" — measures effort required, especially common after a support interaction.
• Each is asked at a different moment and answers a different question — none is a universal replacement for the others.

Side-by-side comparison

When to use CSAT

CSAT is best for evaluating a specific, bounded interaction — right after a purchase, a support ticket resolution, or a specific in-app action. Because it's tied to one moment, it's useful for pinpointing exactly which touchpoints in a customer journey are working well and which aren't, in a way a broader, less frequent metric like NPS can't isolate as precisely.

When to use NPS

NPS is best for tracking overall brand or product health over time — a periodic, relationship-level check-in rather than a reaction to one specific moment. It's the metric most commonly reported to leadership as a single, memorable number representing overall customer sentiment, though as covered in the full NPS guide, it works best paired with a follow-up "why" question to be genuinely actionable.

When to use CES

CES is specifically useful for evaluating how much friction or effort a process required — most commonly used right after a support interaction, but also applicable to any specific task, like completing checkout or setting up an account. Research behind CES suggests that reducing customer effort is often a stronger driver of loyalty than trying to delight customers with above-and-beyond service — a low-effort experience tends to build more loyalty than a high-effort one, even if the high-effort one ultimately gets resolved successfully.

A worked example showing all three together

Picture a software company after a customer contacts support about a billing issue:

• CES (asked immediately after the support interaction): "How easy was it to resolve your issue today?" — measures the specific support experience's effort level.
• CSAT (asked shortly after): "How satisfied were you with this support interaction?" — measures satisfaction with that specific interaction overall.
• NPS (asked periodically, unrelated to this specific interaction): "How likely are you to recommend us to a colleague?" — measures the customer's overall relationship sentiment, which this one support interaction is just one input into, among many others over time.

Using all three appropriately gives a fuller picture: a customer might report a difficult support interaction (low CES) that was still ultimately resolved satisfactorily (moderate CSAT), while their overall loyalty (NPS) remains high due to their broader positive experience with the product overall — three genuinely different signals from the same underlying customer relationship.

Common mistakes when using these metrics

• Using NPS to evaluate a specific interaction, when it's designed to measure overall relationship sentiment — this produces a poorly timed, less actionable signal.
• Treating CSAT scores as a measure of overall loyalty, when they only reflect satisfaction with one specific, bounded moment.
• Ignoring CES, assuming CSAT alone covers effort. A customer can report reasonable satisfaction with an outcome while the process to get there was genuinely high-effort — CES captures this distinctly.
• Surveying too frequently with all three metrics, causing survey fatigue that reduces response rates and data quality across the board.

FAQ

Should a company track all three metrics, or pick just one? Most mature customer experience programs track a combination, since each answers a genuinely different question — the right combination depends on your specific business, but relying on just one metric alone typically leaves a real gap in understanding.

Which metric best predicts customer churn? Research suggests CES (effort) is often a strong predictor of dissatisfaction and churn risk following a service interaction, though NPS is more commonly used as a general, ongoing churn-risk indicator at the relationship level — using them together provides a stronger signal than either alone.

Is CSAT the same as star ratings on a review site? They're conceptually similar — both measure satisfaction with something specific — but CSAT is typically collected directly by the company as part of a structured measurement program, while public review ratings are collected by a third-party platform and serve a different, more public purpose.

How often should CES be measured? CES is best measured immediately following a specific interaction it's meant to evaluate (like right after a support ticket closes), rather than as a periodic, standalone survey unrelated to a specific recent interaction — timing close to the actual event produces more accurate, reliable responses.