Customer effort score
In one sentence
Customer effort score measures how much work a customer had to do to get their issue resolved, usually with a single question asked right after an interaction, and lower effort predicts loyalty better than higher satisfaction does.
Not to be confused with CSAT.
Definition
Customer effort score measures how much work a customer had to do to get their issue resolved. Lower effort predicts loyalty better than higher satisfaction does.
For a website, the effort is hunting through your menus for an answer that exists somewhere. CES is the metric that captures exactly that.
CES emerged from research published in Harvard Business Review in 2010 under the title Stop Trying to Delight Your Customers, which argued that reducing effort does more for loyalty than exceeding expectations.
The core finding
- Customers who have to work hard to resolve an issue become disloyal.
- Customers who are delighted beyond expectation do not become proportionally more loyal.
- So removing friction returns more than adding delight.
How it is measured
- Typically a single question about how easy the company made it to handle their issue, on an agreement scale.
- Asked immediately after an interaction.
- Simpler to interpret than NPS, because it maps directly to an actionable thing: what was hard.
Why it is the most relevant of these metrics to a website voice agent
- The problem an on-site voice agent addresses is precisely effort. The alternative is hunting through menus, using a search box, reading a policy page, or giving up.
- Effort is the mechanism, and reducing it is the value proposition.
- CES measures the thing the product is for, which is unusual among these metrics.
Sources of effort worth measuring separately
- The number of interactions required.
- Channel switching, being moved from one route to another.
- Repeating information already given, which is the cold transfer problem.
- Time spent waiting.
- Difficulty finding the right place to start.
Common misconception
That effort and satisfaction measure the same thing. A customer can be satisfied with an outcome and exhausted by the process, and the exhaustion is what predicts whether they come back.
Why it matters commercially
For an on-site voice agent, effort reduction is the product, and CES is the metric that captures it. It aligns with the strategic framing that the enemy is the on-site hunt: the visitor with a question facing a navigation bar, a search box and a policy page.
In voice specifically
Speech reduces effort in a way a page of links cannot, because the visitor states the question rather than translating it into where the answer might be filed. Voice navigation, showing the visitor where the answer lives, is an effort reduction as much as a convenience, and CES is the metric that reads it.
Where AsqVox fits
The Orb exists to reduce the effort of getting an answer from a website. Voice navigation, which takes the visitor to where the answer lives instead of leaving them to find it, is an effort reduction mechanism, and effort is the thing CES measures.
Visual
Why effort costs more than it used to
Visitors increasingly arrive having already done the work elsewhere, which raises the cost of making them work again on your site.
Statistics
Every figure carries its source and year. Vendor numbers are labelled as vendor numbers, and where no reliable figure exists this page says so rather than borrowing one.
Customer effort score emerged from research published in Harvard Business Review in 2010, arguing that reducing customer effort predicts loyalty more effectively than exceeding expectations.
2010independentHarvard Business Review, Stop Trying to Delight Your Customers, 2010 - This is the provenance the metric rests on. The finding is qualitative, so cite it as a research argument about loyalty, not as a benchmark number.
Zero-click search reached 68.01 percent of US Google searches in early 2026, up from 60.45 percent in 2024, so visitors increasingly arrive having already done work elsewhere.
68.01%independentSparkToro with Similarweb data, published June 2026, 2026 - Independent third-party measurement, US Google only, at a point in time. Quote it with the geography and date attached. It matters here because arriving pre-worked raises the cost of making a visitor work again.
Exit-intent overlays convert at roughly 2.87 to 3.94 percent, the point at which a well-timed on-page intervention meets a departing visitor.
2.87 to 3.94%vendor claimWisepops, Popupsmart and OptiMonk, 2026 - Vendor-published aggregates drawn from their own customer bases across 2025 and 2026. Read as directional evidence that a timed intervention works, not as a benchmark anyone can be held to.
Cart-abandonment specific overlays are reported converting at around 17.12 percent.
17.12%vendor claimWisepops, Popupsmart and OptiMonk, 2026 - Same caveat. It is one narrow, high-intent moment, so it does not generalize to on-page intervention as a whole.
There is no reliable cross-industry CES benchmark, and the scale wording varies between implementations, so two published CES figures are rarely measuring the same thing.
-no reliable figureBecause the question wording moves the score, a CES number is only comparable against your own history under a fixed question, not against another company.
There is no published research measuring effort reduction attributable to on-site voice agents specifically.
-no reliable figureThe honest substitute is your own before-and-after on the five effort sources: interactions per resolution, channel switches, repeated information, wait time and time-to-start.
Examples
In practice
An organization measuring only CSAT sees stable scores while retention declines. Adding an effort question reveals that resolution frequently requires three interactions across two channels. Customers are satisfied with the eventual outcome and worn out by reaching it. Effort becomes the improvement target, and retention responds where satisfaction alone never flagged the problem.
The everyday version
Customer effort score asks how hard your customer had to work to get what they needed. The research behind it found that making things easy matters more than making them delightful. For a website, the effort is hunting through your menus for an answer that exists somewhere.
Usage
Who says it
- CX practitioners and service designers, who use it to find what was hard.
- Increasingly favored over NPS for transactional measurement, because it maps to an action.
Where it turns up
- In an RFP, next to CSAT, resolution rate, channel switching and escalation quality.
Common misuse
- Treating it as interchangeable with satisfaction, when the two measure different things.
- Measuring effort without identifying its sources, which makes the number undiagnostic.
- Ignoring it in favor of NPS because NPS is more familiar to executives.
Questions people ask
What is the difference between customer effort score and CSAT?
CSAT measures satisfaction with an outcome; CES measures how hard the customer had to work to get it. A customer can be satisfied with the result and exhausted by the process, and the exhaustion is what predicts whether they come back. Treating the two as the same thing is the most common misuse of both.
Why is customer effort score better than NPS for a voice agent?
CES measures the thing an on-site voice agent is for: effort. NPS measures relationship-level advocacy, which a single conversation rarely moves and cannot be attributed to a deployment. CES asks immediately after the interaction and maps directly to what was hard, which is actionable.
How do you measure customer effort score?
Typically a single question about how easy the company made it to handle the issue, on an agreement scale, asked right after the interaction. To make it diagnostic rather than just a number, break effort into its sources: interactions required, channel switching, repeating information, waiting, and difficulty finding where to start.
Is there a benchmark customer effort score?
No. There is no reliable cross-industry CES benchmark, and the scale wording varies between implementations, so two published figures rarely measure the same thing. Compare a CES number against your own history under a fixed question, not against another company.
Last reviewed 4 August 2026. Written and reviewed by Dhruv Dholakia, founder of AsqVox.