KPIs and metrics

Customer Effort Score for Phone Support: How to Measure It

What customer effort score (CES) tells you about phone support, how to ask the question, how to read the answers and what to fix first.

Call editorial team · Published · 6 min read

Satisfaction scores tell you whether a customer was happy. Customer effort score, usually shortened to CES, asks something more practical: how hard did the customer have to work to get their problem solved? On the phone, effort is easy to picture. It is the long menu, the second call, the transfer to a team that asks for the same details again, and the hold time in between.

A person on a headset reviewing survey results on a laptop in a support office
Illustrative image

This guide explains what CES measures, how to collect it without annoying callers, how to read the results and which fixes tend to lower effort. If you already track satisfaction, read it alongside CSAT After Calls and Net Promoter Score for Support; the three questions measure different things.

What customer effort score measures

The classic CES question is a single statement, such as "The company made it easy for me to handle my issue", with a scale from strongly disagree to strongly agree. Some teams use a five-point scale, others seven. The wording and the scale matter less than keeping them the same over time, because the value of the number is in the trend.

The idea behind CES is that people do not remember a support call as a list of pleasant moments. They remember friction. A friendly agent cannot fully make up for a caller who had to explain the issue three times. That is why effort often explains loyalty better than satisfaction does, and why it is a good companion metric for a small team.

How CES differs from other survey scores

  • CSATAsks how satisfied the customer was with the interaction. Good for a quick read on mood and on individual conversations.
  • NPSAsks how likely the customer is to recommend you. It looks at the whole relationship, not one call.
  • CESAsks how easy the issue was to resolve. It points at process problems you can actually fix.
  • First-call resolutionCounts whether the issue was solved on the first contact. It is a hard operational number, where CES is the customer's own view.

When and how to ask

The best moment to ask is straight after the interaction, while the memory is fresh. For phone support that usually means one of three routes: a short automated question at the end of the call, a follow-up message a little later, or a question the agent asks aloud at the close. Each has trade-offs.

Ways to collect effort feedback on calls

Automated end-of-callFollow-up messageAgent asks aloud
Response rateUsually the highest, because the caller is still on the lineOften lower, because it is easy to ignoreDepends on the agent and the mood of the call
HonestyFairly honest, no person is listeningHonest and usually more detailedCallers may soften answers to be polite
Effort for the callerOne key pressA link and a few secondsA spoken answer
RiskFeels abrupt if the call ended badlyArrives when the issue is already forgottenInconsistent between agents

Whichever route you pick, keep the survey short. One scored question and one optional free-text box is plenty. Every extra question lowers the response rate and adds effort, which would be ironic for a survey about effort.

Reading the results without fooling yourself

A raw average is a blunt tool. Look at the shape of the answers first: a small group of very low scores matters more than an average that looks fine. Then break the score down so you can act on it.

Treat comments as the main source of insight. Ten short comments such as "I had to press five keys and then repeat my account number" will tell you more than a decimal point of difference in the average. Group the comments into a few themes and count how often each theme appears.

Be careful with small numbers. A team that handles a modest volume may get only a handful of responses a week, so one unhappy caller can move the average. That is a reason to focus on themes and not to chase the score.

What usually drives effort on phone support

When teams read their comments and call data together, the same causes keep appearing. Look for these first.

A typical high-effort call

  1. Long menuThe caller listens to options that do not fit, or cannot find a way to a person.
  2. AuthenticationThey are asked for the same details by the menu, then the agent, then the next agent.
  3. TransferThe first agent is the wrong team, so the call moves and context is lost.
  4. WaitingHold time grows with each handover, and the caller is never told how long it will be.
  5. Second callThe issue is not fully solved, so they call again and start from zero.

Each step has a measurable twin. Menu depth shows in the menu design; transfers show in Transfer Rate; repeat contacts show in Repeat Call Rate; waiting shows in Average Speed of Answer. Lining those numbers up next to the CES comments usually points to one or two dominant causes.

Ways to lower effort

Once you know the main cause, the fixes are often small. Some of the most dependable ones:

Before you launch a CES question

  • Choose one wording and one scale and write them down
  • Decide when the question is asked and for which calls
  • Make sure the question is easy to answer on a phone keypad or in one click
  • Set a place where comments are read every week
  • Agree what a good result looks like for you, using your own trend rather than another company's number
  • Decide who owns the fixes, so the survey leads to action

Finding effort with Call

Call does not replace a survey tool, but it holds much of the operational evidence that explains effort. Call history lists each call with its summary and transcript, so you can open a low-scoring call and see what happened. You can search transcripts for phrases such as "I already told" or "called yesterday" to find repeat-contact patterns, and the Reports section shows intervals, agent scorecards and patterns in arrival times that help you link long waits to specific hours.

If your team keeps a survey result next to a call, add a note to the contact record so the follow-up has context. Our guide First-Call Resolution shows how to treat the operational side of the same question.

Frequently asked questions

What is a good customer effort score?

There is no universal number. Scales and wording differ, so compare yourself with your own past results and look for trends. A steady improvement after a fix is a better sign than a score that happens to look high.

Is CES better than CSAT?

They answer different questions. CSAT tells you how the customer felt; CES tells you how much work the issue took. Many teams use CES for process improvement and CSAT for a general mood check.

How many responses do I need?

Enough that one unhappy caller does not swing the result. For small teams, read every comment and watch themes over several weeks rather than relying on the average.

Can I ask the CES question by voice?

Yes. A short automated question at the end of the call, answered with a keypad press, is a common approach. Keep it to one question and make sure the caller can skip it.

Should agents see their own CES results?

Showing the trend and the comments is useful for coaching. Avoid tying pay or rankings to it, because that encourages agents to dodge difficult calls.

What if the score is low but handle time is short?

A short call can still be a hard one, for example when the caller has to call back. Look at repeat contacts and transfers to see whether quick calls are really solving the issue.

Review your call history and find high-effort calls in Call

In the demo console you can try incoming calls in a simulation.

Related guides