Customer effort score (CES): a leading signal of loyalty

Customer effort score reveals loyalty risk that satisfaction dashboards often miss. Your dashboards may look stable even as experience budgets still grow every cycle. Churn and repeat-contact rates remain stubborn because customers remember long holds and repeated explanations. CES turns the ease of each resolution path into a loyalty signal that CX leaders can track and improve across channels. Most effort programs assume that a human handles every call. Automated resolutions now create a growing measurement blind spot. If a customer resolved an issue this morning without speaking to anyone, would your program register the time, repetition, and friction involved? Coverage across human and automated resolution paths becomes essential as service automation grows.

What effort actually predicts about repeat business

Customers are more likely to return when resolving an issue takes little time or repetition. Customer effort score (CES) captures that with a single post-interaction question. The original wording: "On a scale of 1 to 7, how strongly do you agree: [Company] made it easy for me to resolve my issue."

Dixon, Freeman, and Toman developed the question for research that Harvard Business Review published in 2010, drawing on more than 75,000 customer interactions. Exceeding customer expectations barely moves loyalty. Ease of resolution is what tracks with repeat business. The reason is unglamorous: a customer who called for a resolution remembers the time it cost to get the problem fixed more than a surprise upgrade or goodwill credit. In service recovery, a faster resolution gives the customer time back.

The original research measured what customers said they intended to do next. Customers who got an easy resolution said they would buy again and spend more with the company; those who had to work for a resolution were most likely to tell other people about it. Repurchase, spending, and negative word of mouth are the behaviors a CX leader manages. The original study captured repurchase and spending as stated intentions, so the effort score belongs next to operational data such as repeat-contact rates and churn.

A 2025 peer-reviewed study of 359 CES responses (opens in a new tab) found that lower effort improved satisfaction, which then strengthened repurchase and recommendation intentions.

Most teams calculate CES as the share of respondents who agree that the interaction was easy. A consistent CES survey design standardizes the question, scale, trigger, and calculation method, giving CX leaders a defensible trend to manage.

Where CES breaks down as a standalone metric

A CX leader who reports CES without knowing its limits will misread the score. The metric has known blind spots; design around each one before the number reaches a board slide.

• No expectation context: The standard effort question does not account for cases where more effort is not necessarily worse, or for how the effort compared with what the customer expected.

• No published benchmark: No universal industry CES standard exists, so a score is most meaningful against your own history and trend rather than as a direct comparison with others.

• A 2010 research base: The foundational dataset dates to 2010, and a leader presenting the metric at board level should be able to explain the scope and limits of that research.

Each limit narrows the claim CES can support, so read it as a trended, interaction-level signal.

Customer satisfaction score (CSAT) and Net Promoter Score (NPS) cover the broader outcomes of momentary satisfaction and brand advocacy, respectively. A program that focuses on improving CSAT still needs the effort signal underneath it because the loyalty evidence links ease with repeat business.

The American Customer Satisfaction Index (ACSI) national score reached 76.9 (opens in a new tab) in the fourth quarter of 2025, down 0.5% year over year, and it has not materially increased since 2017. A flat national index gives an operator very little to act on. It does not identify which interaction paths consumed the customer's time or which resolutions required a second contact. An effort signal measures how hard the customer had to work, a factor the satisfaction stack cannot see.

Measuring effort when AI agents answer the call

Many post-call survey programs assume that a human handles the call, and automated service is making that assumption obsolete. Today 91% of service leaders (opens in a new tab) report pressure to implement AI in 2026. Every deployment that follows creates interactions standard post-call surveys do not score.

A single survey may capture only the last contact. The customer's effort can span several earlier contacts. In August 2024, Gartner reported that self-service fully resolved only 14% of customer service issues (opens in a new tab), so customers whose self-service attempts fail may continue through additional contacts. A survey that fires after the final human call scores the last leg and misses the failed self-service attempt that generated most of the effort.

Quality frameworks for human agents do not describe failures inside an automated resolution. Script adherence and coaching notes say nothing about an AI agent that answered confidently and wrongly or looped a caller through the same clarifying question three times. The quality team must rewrite the review criteria for calls AI resolves before the volume arrives.

Many service organizations put AI at the center of CX workflows before assigning responsibility for survey integrity in an AI governance policy. The AI governance policy should place effort-survey integrity under the CX leader who reports the number. The team building the automation can then support the measurement process without controlling it.

Four measurement rules make the score defensible across automated and human service paths.

1. Separate resolution paths

Survey interactions AI agents resolve separately from those human agents resolve. Separate scores reveal whether the AI agent lowered effort or shifted it downstream.

2. Survey after resolution

Trigger the survey when the service operation resolves the issue. A customer who abandoned the automated conversation and called back the next day experienced substantial effort, regardless of what the containment dashboard reports.

3. Assign handoff effort

Attribute effort across an escalation handoff to the leg that generated it. When a caller repeats their issue after a transfer, the handoff design generated that effort.

4. Audit survey suppression

A program that surveys only successful automated resolutions flatters the deployment and leaves the hardest service paths unmeasured. Record the interactions the program excluded and who authorized each exclusion.

Separate path scoring, resolution-based survey triggers, handoff attribution, and suppression audits connect the survey result to the interaction that produced it. Your score matters only if the service team changes the interactions causing repeated calls, transfers, and long waits. Teams use the score to redesign high-effort interactions. On the phone, effort appears in the depth of the menu before a caller reaches anything useful, account details recited a second time after a transfer, the issue explained again to whoever received the escalation, and the hold music in between. Treat the CES survey as one instrument in a wider voice of customer program, read against the operational signals that show the source of the effort.

Reducing effort in the phone channel

The voice channel is where an enterprise contact center can cut effort fastest because a phone call concentrates effort at five predictable points: wait time, menu navigation, authentication, repeated explanations, and transfers.

• Wait time: An AI agent can answer immediately and reduce peak-volume waits, so callers receive help faster.

• Menu navigation: Callers state their need in plain language and skip the menu tree.

• Authentication: Identity verification runs once inside the conversation and holds through the rest of it.

• Repeated explanations: Accurate intent recognition means the caller describes the problem a single time.

• Transfers: Escalation logic carries context forward and turns a handoff into a continuation of the same conversation.

Conversation design can reduce each driver, and the customer does not have to do anything differently for the reduction to work.

Programs often lose ground quietly on repeated explanations and transfers. A transferred caller who has to read out an account number, answer a security question, and explain the reason for calling all over again is proving something the system already knew. Carrying context forward means something specific in practice: the verified identity and diagnosed intent travel with the caller, so the human agent opens with the issue context already available.

Teams can quantify wait time most easily. In the orderbird customer story, orderbird cut customer wait time by 60%, from 98 to 39 seconds.

The Swiss Life customer story reports that Swiss Life became 60% faster at addressing customer concerns. In addition, 73% of customers rated the AI agent 4 or 5 out of 5, and its AI agent routes callers with 96% accuracy. The cited materials do not label the 73% rating as CES. Routing at that level keeps callers out of a second queue and prevents repeated explanations to a new person. The 73% customer rating belongs next to an effort score. An operational metric says the call moved faster. The customer rating shows a favorable evaluation. It does not measure ease. Pair it with CES to determine whether the interaction felt easier. Teams can use customer loyalty analytics to compare lower effort with later retention and spending behavior.

Use customer effort score to expose loyalty risk

Your loyalty risk lives in interaction effort, and a survey program covering only calls that human agents handle measures a shrinking share of it. CES earns its place in a board deck when the instrument covers every resolution path, including those no human touches. Parloa's AI Agent Management Platform follows the Design, Test, Scale, and Optimize lifecycle, testing AI agents against real conversation complexity before launch and monitoring resolution, drop-off, and frustration signals afterward across 140+ languages. Book a demo to see how AI agents lower customer effort in your phone channel. Customers do not remember being surveyed; they remember whether getting help was easy.

FAQs about customer effort score

These answers cover CES calculation, interpretation, and use across human and automated service paths.

What is a good customer effort score?

There is no truly universal, authoritative industry standard that applies across all organizations. Compare the score with your own prior quarters, segment it by resolution path, and manage the direction of travel.

How do teams calculate CES?

The most common convention is the sum of positive responses divided by total responses, multiplied by 100. Calculation conventions vary across survey vendors, so pick one method and hold it constant.

Is CES better than NPS or CSAT?

Each answers a different question. CES is the interaction-level ease signal, CSAT reflects satisfaction with a moment, and NPS reflects brand advocacy. A CX program needs the effort signal because ease of resolution is the variable that tracks with repeat business.

Should AI agent interactions get their own CES survey?

Yes. Score interactions AI agents resolve separately from those human agents resolve, fire the survey on resolution, and assign effort across escalation handoffs to the leg that generated it.

Does low effort actually increase loyalty?

The original research behind the metric found that customers who resolved issues easily were more likely to repurchase and spend more, with a lower likelihood of spreading negative word of mouth. A 2025 peer-reviewed study supported the same mechanism, with lower effort flowing through satisfaction into repurchase and recommendation intent, giving CX leaders a current reason to manage effort as a loyalty input.