Most support teams track whether customers were satisfied. Far fewer track how hard those customers had to work.
That gap matters, because effort predicts churn better than satisfaction does. A customer can rate an interaction positively and still quietly decide never to deal with you again. Customer Effort Score is the metric that catches what CSAT misses.
This guide covers what CES measures, how to run it properly, what a good score looks like on the standard scale, and the structural changes that move it. Written for ANZ mid-market CX teams.
TL;DR
- Customer Effort Score measures how hard a customer worked to get an issue resolved. It predicts retention better than NPS or CSAT because it tracks friction rather than feeling.
- Use the CES 2.0 format: the statement "The company made it easy for me to handle my issue" on a 1 to 7 agree scale. Older phrasings still circulate and they break comparability with published benchmarks.
- On the 7-point scale, 5.0 and above is solid and 5.5 and above is strong. The most useful benchmark is your own trend, not an industry average.
- CES moves through operating model change: first contact resolution, agent authority, fewer channel switches, better self-service. Training and scripts do not shift it.
- Track CES alongside CSAT and NPS. It is the most actionable of the three for service operations.
What is Customer Effort Score?
Customer Effort Score is a single-question metric that measures how much effort a customer had to expend to get an issue resolved. It focuses on operational friction rather than emotional response, which is what makes it more actionable than NPS or CSAT alone.
CEB, now part of Gartner, introduced it in 2010 through the Harvard Business Review article "Stop Trying to Delight Your Customers". The research finding was counterintuitive at the time. Reducing effort predicted loyalty far more reliably than exceeding expectations did. Customers do not necessarily become loyal when you impress them. They almost certainly become disloyal when you make things hard.
Why does effort predict churn better than satisfaction?
Because effort is what customers remember and act on.
The Gartner research behind the metric is blunt on the size of the gap. 96 percent of customers who have a high-effort experience go on to become disloyal, against just 9 percent of those who have a low-effort one. CEB also found that 81 percent of customers who had high-effort experiences intended to spread negative word of mouth.
This is why teams get confused by a dashboard showing healthy CSAT alongside rising churn. Satisfaction asks whether the interaction was pleasant. Effort asks what it cost the customer to get there. A customer can answer the first question generously and still be done with you.
Customer expectations have shifted here too. Customers no longer compare your service to others in your industry. They compare it to the easiest experience they had that week, usually a consumer app. CES measures that gap directly.
How do you measure CES?
Deploy a single-question survey immediately after each service interaction. Use the CES 2.0 format, which is the CEB and Gartner standard:
The company made it easy for me to handle my issue.
Customers respond on a 1 to 7 scale, from strongly disagree to strongly agree. This phrasing matters more than it looks. Most published benchmark data uses this format, so older variants like "how easy was it to resolve your issue" leave you unable to compare your score to anything outside your own organisation.
There are two accepted calculations:
- Average score. Sum all responses and divide by the number of responses. Better for tracking trends.
- Percentage positive. Divide the count of 6 and 7 responses by total responses, then multiply by 100. Better for executive reporting.
Pick one and stay with it. Switching scales or methods mid-programme makes your trend data unusable, which is the whole point of measuring.
Track weekly. Segment by channel, contact reason and team. Capture verbatim comments alongside the score, because the comments tell you where the friction sits and the score only tells you whether it is improving.
What is a good CES score?
Commonly cited ranges on the 7-point scale look like this:
| CES Range | Interpretation | Likely business impact |
|---|---|---|
| 6.0 to 7.0 | Low-effort experience | Strong retention, organic growth, NPS lift |
| 5.5 to 6.0 | Strong | Healthy retention, supports growth |
| 5.0 to 5.5 | Solid, around the cross-industry average | Acceptable, with churn risk in dissatisfied segments |
| 4.5 to 5.0 | Functional, real improvement opportunity | Friction showing up in escalation volume |
| Below 4.5 | Material friction | Churn risk, NPS suppression |
Treat these as orientation, not targets. Benchmarks vary by industry, touchpoint and scale, and most published figures come from vendor research rather than independent studies. Your own quarter-on-quarter trend is the number that should drive decisions.
How do you improve CES?
Through structural change to how service runs. Five drivers do most of the work.
Increase first contact resolution. Every escalation is friction the customer absorbs. This is the single largest lever in most operations.
Reduce channel switching. Customers who repeat their story when moving between email, chat and phone report markedly lower effort scores. Unified customer context removes that penalty. Our guide to omnichannel versus multichannel support covers what genuine context-sharing requires, and why most setups labelled omnichannel are not.
Expand agent authority. Agents who must escalate routine decisions create friction by design. Document your top contact reasons and check how many Level 1 can resolve without approval. The gap is usually larger than leadership expects.
Improve self-service usability. Test the portal with real users who are not in the support team and watch where they stall. Rename items to match what people actually search for. Structure categories around customer mental models rather than internal org charts.
Pre-empt the contact entirely. A strong knowledge base and AI-powered article suggestions remove the interaction. The lowest-effort contact is the one that never happened.
One caution on AI. Contained AI conversations that fully resolve tend to score well on effort, because they remove queueing and hand-off delay. AI conversations that fail and escalate produce some of the worst effort scores in any channel, because the customer spends the effort twice and starts again with a human. Monitor containment and CES together, or you will mistake deflection for ease.
What effort reduction looks like in practice
National Pharmacies ran support across 65-plus stores through shared email inboxes, spreadsheets and phone follow-ups. Every one of those is an effort driver. Requests had no obvious destination, status was invisible to the person who raised them, and chasing was the default behaviour on both sides.
Working with KlickFlow, the operation moved to Freshdesk with automatic routing to the right department, one visible source of truth for status and ownership, and reporting on recurring issues across the network. Phase 1 went live in six weeks with no additional headcount.
We did not run a CES programme on that engagement, so there is no before-and-after effort score to quote. What the work does illustrate is the shape of the problem. The changes that reduce effort are structural: where a request goes, who can resolve it, and whether the customer can see what is happening without asking. The full story is in our National Pharmacies case study.
CES vs NPS vs CSAT
Track all three. They answer different questions.
| Metric | What it asks | Best used for |
|---|---|---|
| NPS | Would you recommend us? | Relationship health, lagging indicator |
| CSAT | Were you satisfied with this interaction? | Transactional quality |
| CES | How hard did you have to work? | Operational friction, retention prediction |
CES is the most actionable of the three for service operations, because every point of friction it surfaces maps to something you can change. Our guide to improving CX metrics covers how the three fit together in a reporting cadence.
Frequently Asked Questions
When should we ask the CES question?
Immediately after the interaction concludes. Within five to ten minutes for digital channels, and at end of call for phone. Response rates and accuracy both fall away once you delay beyond a day, because the customer is recalling the experience rather than reporting it.
Should we use a 5-point or 7-point scale?
The 7-point scale is the CEB and Gartner standard and carries the most benchmark data behind it. The 5-point version is simpler and works for in-app or high-volume pulse checks. Whichever you choose, do not switch later. Mixing scales across time makes trend analysis unreliable and costs you the value of everything you measured before the change.
Why is our CSAT healthy while customers still leave?
This is the pattern CES was designed to explain. Satisfaction surveys ask whether the interaction was handled well, and customers answer generously about individual agents who were polite and competent. Effort asks what the whole journey cost them: how many contacts, how many repeats, how long spent waiting. A customer can rate every agent highly and still churn because getting there was exhausting.
How long does it take to improve CES?
Meaningful movement takes months rather than weeks, because the changes that shift it are structural. Expanding agent authority, redesigning a self-service portal and unifying customer context across channels are operating model projects. Training-led programmes rarely produce sustained gains, because they ask agents to work around friction rather than removing it.
Where to start
Three things worth taking from this. Effort predicts retention in a way satisfaction does not. The CES 2.0 question on a 7-point scale is the format to use if you want your numbers to mean anything outside your own dashboard. And the score only moves when the operating model changes.
If your CSAT looks healthy but customers still leave, that gap is usually measurable and usually fixable.
Book a diagnostic call. We will look at where effort is accumulating in your service operation and what to address first.
For teams ready to act on what CES surfaces, our CX platform optimisation service covers the structural work: routing, agent authority, self-service and knowledge.