What is customer effort score (CES)?

Customer effort score (CES) is a single-question customer experience metric that measures how easy or difficult it was for a customer to resolve an issue, complete a task, or get help — typically expressed as an average on a 1-to-7 scale where higher means easier. It is collected by asking one question immediately after an interaction and is calculated either as the mean of all responses or as the percentage of respondents who rated the experience easy.

CES was built around a specific research insight: reducing the effort a customer has to exert predicts loyalty better than trying to delight them. That finding, published by researchers at the Corporate Executive Board in 2010 and extended in the 2013 book The Effortless Experience, reframed the goal of customer service from "exceed expectations" to "remove obstacles."

This page covers what CES measures, where the methodology came from, how to calculate it, what scale to use, what a "good" score looks like, how CES compares to CSAT and NPS, where the metric is strong, where it falls short, and a short FAQ.

CES in one sentence

CES is a score that tells you how easy your customers found it to get what they needed.

What CES actually measures

CES is a friction metric. It measures the amount of effort a customer had to invest in a specific task — not whether they were satisfied, not whether they would recommend the brand, but how hard the experience was to get through. The question is always pinned to a concrete, recent interaction: a support ticket, a self-service flow, a return, a purchase, a troubleshooting step.

That scoping is the whole point. CES is built to find the places in the customer journey where effort is getting in the way. If a particular self-service flow scores poorly, that is investigable this week. If post-ticket CES drops on phone interactions but stays flat on chat, that is a signal the phone queue has a specific friction problem.

Customer effort as a concept is intuitive. CES as a metric is narrow and post-interaction. The narrowness is a feature — it is fast to collect, fast to act on, and predictive of the outcome that matters most to contact center operators: whether the customer will come back.

Where CES came from

CES was introduced in a July 2010 Harvard Business Review article titled "Stop Trying to Delight Your Customers" by Matthew Dixon, Nick Toman, and Rick DeLisi, who were then researchers at the Corporate Executive Board (CEB). The article challenged the prevailing assumption that delight — exceeding customer expectations — was the key driver of customer loyalty.

The CEB research found the opposite. Customers who had low-effort interactions were substantially more loyal than customers who had high-effort ones, regardless of whether the interaction had exceeded their expectations. High effort — having to call back, explain the issue again, navigate multiple channels — predicted disloyalty more reliably than dissatisfaction did. The first version of the metric (CES 1.0) asked customers to rate agreement with the statement "the company made it easy for me to handle my issue" on a 1-to-5 scale.

In 2013, Dixon, Toman, and DeLisi published The Effortless Experience: Conquering the New Battleground for Customer Loyalty, which extended the CEB research and introduced the second version of the metric (CES 2.0). The revised survey question — "how easy was it to handle your request?" — shifted the framing from agreement-with-a-statement to direct ease-of-experience, and the scale was extended from 5 to 7 points to produce more behavioral differentiation between response bands. CES 2.0 is the current standard.

CEB was acquired by Gartner in 2017, and subsequent Gartner research has continued to validate and update the core CES findings. The most-cited current figures come from Gartner's updated analysis: 94 percent of customers with low-effort interactions intend to repurchase, compared to 4 percent of those with high-effort interactions. According to Gartner, a low-effort interaction also costs 37 percent less to deliver than a high-effort one.

How CES is measured

A CES survey has three moving parts: the question, the scale, and the formula.

The question

CES is a single-question survey. The current standard wording (CES 2.0) is:

Or the equivalent phrasing: "The company made it easy for me to resolve my issue" (Likert agreement form), "How would you rate the ease of your experience?", or "How easy was it to [complete the specific task]?" — scoped tightly to the interaction that just ended.

The scale

The dominant format for CES 2.0 is a 7-point scale:

• 1 (very difficult) to 7 (very easy). The full-range format. Higher scores are better — the customer found the experience easy.

• 1-to-5 scale. An older format, still in use, with the same directional logic: 5 = very easy. Less granular but more familiar to customers who see CSAT surveys regularly.

• Emoji/emoticon scale. Used when brevity matters — embedded in a chat close, a post-IVR prompt, or an in-app nudge. Assign numeric values (1–3 or 1–5) for calculation.

The 1-to-7 scale is preferred because it produces more spread in the data and reduces the ceiling effect common on 1-to-5 scales.

The formula

Two calculation methods are in common use.

The output is a decimal between 1 and 7. A score of 5.5 or higher is generally considered healthy.

The output is a percentage. A top-box CES of 70 percent or higher is considered strong.

Worked example (average method): 400 responses. The sum of all scores is 2,300. CES = 2,300 ÷ 400 = 5.75 — a healthy score on the 1-to-7 scale.

Worked example (top-box): 400 responses. 300 rated their experience a 5, 6, or 7. CES (top-box) = (300 ÷ 400) × 100 = 75% — above the 70% threshold.

Both methods are defensible. The average-score method is easier to track over time; the top-box method is easier to explain to non-CX stakeholders.

When to ask

CES works best immediately after the interaction. The most common trigger points:

• Right after a support ticket closes

• At the end of a chatbot or self-service session

• After a customer navigates an IVR tree

• After a return or exchange is processed

• After an onboarding step is completed

For depth on how to design effective post-interaction surveys, see Gladly's guide on customer satisfaction surveys.

CES vs. NPS vs. CSAT

CES is one of three single-question metrics that appear in almost every customer experience program. They measure different things, ask different questions, and are useful in different moments.

The shortest way to keep them straight: CES is about the friction, CSAT is about the moment, NPS is about the relationship. Most mature CX programs run all three because each one answers a question the others cannot. For a deeper read on how these metrics work together inside a contact center, see the Gladly guide to contact center metrics for modern CX teams.

A note on overlap: CES and CSAT are sometimes confused because both fire post-interaction. The difference is that CSAT asks how the customer felt about the experience, while CES asks how hard the experience was. A customer can find something easy and still be unsatisfied — a quick ticket close that solved the wrong problem, for example. A customer can also find something hard and still rate the outcome positively — a complicated return that the agent handled with patience and care. Running both produces a more complete picture than running either one alone.

What is a good CES score?

There is no universal "good" CES because the metric is newer than NPS or CSAT and cross-industry benchmark data is less mature. The most useful yardsticks available:

On the 1-to-7 scale:

• 5.5 or higher — generally healthy. Most customers found the interaction easy.

• Below 5.0 — a signal worth investigating. A meaningful share of customers found the experience difficult.

• 6.5 or higher — excellent. Very high ease ratings, typical of well-optimized support or self-service flows.

Top-box percentage (5s, 6s, and 7s):

• 70% or higher — strong. The majority of customers found the interaction easy.

• Below 50% — more customers found the experience difficult than easy. A priority for intervention.

Industry variation is significant. CES tends to run lower in technically complex B2B environments (enterprise software, financial services) and higher in consumer retail and ecommerce. Telecom and utilities historically have some of the lowest CES scores of any industry, because the interactions customers have with those companies tend to involve billing disputes and service outages — high-stakes, multi-step, high-frustration scenarios.

The more useful question than "is my CES good in absolute terms?" is "is my CES improving over time, and is the effort coming down after we made a change?" CES is at its most valuable as a trend signal and an A/B testing instrument, not as a static benchmark.

Strengths of CES

CES has stuck around because it measures something the other metrics do not, and because its predictive power on loyalty is well-supported.

It predicts disloyalty before it shows up as churn. The core CEB insight — that high effort is a stronger predictor of disloyalty than delight is a predictor of loyalty — has held up across subsequent research. Customers do not leave because they were not delighted. They leave because something was too hard and they found an easier option. CES catches that signal early.

It is operationally specific. Because CES fires at the end of a discrete interaction, a low score is investigable. It is not "customers are unhappy with the brand" — it is "customers found this specific flow difficult." That is something a support team, a product team, or a self-service team can act on.

It is fast to collect and easy to answer. One question, immediately after the interaction ends. Response rates are typically high because the experience is fresh and the ask is small.

It is the right metric for self-service. CSAT and NPS were designed for human-to-human interactions. CES is purpose-built for task completion — which makes it the natural fit for self-service flows, IVR trees, chatbot interactions, and any digital experience where the goal is for the customer to get in and get out.

It lowers costs. Gartner's finding that low-effort interactions cost 37 percent less to deliver than high-effort ones gives CES a direct line to operational ROI. Improving CES is not just a loyalty play — it is a cost-reduction play.

Limitations of CES

CES is a powerful signal in its lane, and a weak signal outside of it.

It only measures ease, not sentiment. A customer can find an interaction easy and still be unhappy with the outcome. A refund processed quickly is still a refund — the customer lost the product they wanted. CES will score that interaction well; CSAT might not. Teams that run CES without CSAT risk optimizing for process speed while missing outcome quality.

It does not tell you why effort was high. A CES of 4.2 tells you that customers found a flow difficult. It does not tell you whether the difficulty came from a confusing IVR menu, an agent who lacked information, a policy that required escalation, or a knowledge base that couldn't answer the question. To get from "effort was high" to "here's what to fix," the score has to be combined with conversation data, session recordings, or agent feedback.

It is weak as a relationship metric. CES is a transactional instrument. It captures one touchpoint. A customer who calls five times in a year may have five high-CES interactions, but each one adds to an accumulation of effort that NPS will capture and CES will not. CES measures each interaction in isolation; the relationship lives in the sum.

It is vulnerable to recency bias. The final moments of an interaction carry disproportionate weight in how customers answer. An agent who closes a ticket warmly after a difficult diagnostic process may get a high CES despite the difficulty. This is not unique to CES — all post-interaction surveys share it — but it is worth knowing.

It is a lagging indicator on its own. Like CSAT and NPS, CES tells you what happened after it happened. The more powerful application is pairing it with operational data — handle time, channel switches, repeat contacts — so a team can predict which interactions are likely to produce high-effort scores before customers have to answer a survey.

The practical takeaway: CES is the right instrument for measuring friction at a specific touchpoint, and a poor instrument for measuring the overall customer relationship. It tells you where the walls are. It does not tell you how to take them down.