Customer effort score (CES) is a customer service metric that measures how easy it was for a customer to resolve an issue or complete a task with a company. It comes from a one-question survey sent after the interaction, and it shows where customers have to work too hard.
Effort covers everything the customer had to do: wait, repeat themselves, switch channels, fill in a form, or call back. CES puts a number on that, so a team can find the contact reasons and processes that cause the most friction.
Where customer effort score comes from
The metric grew out of research at CEB published in Harvard Business Review in 2010, "Stop Trying to Delight Your Customers" by Matthew Dixon, Karen Freeman, and Nick Toman. The authors argued that reducing effort predicts loyalty better than delighting customers does.
The book that followed, The Effortless Experience, reports that 96% of customers with a high-effort service interaction became more disloyal, against 9% after a low-effort one (as cited by Qualtrics).
How to calculate customer effort score
Teams calculate CES in one of three ways (Zendesk describes the first and third).
Average score. CES = sum of all ratings ÷ number of responses.
Share of easy answers. CES (%) = responses in the "easy" range ÷ total responses × 100.
Net score. CES = % easy answers − % difficult answers, with neutral answers left out.
Worked example. A team asks "How easy was it to resolve your issue today?" on a 1 to 7 scale, where 7 means very easy. It collects 300 responses: 60 answer 7, 90 answer 6, 60 answer 5, 30 answer 4, 25 answer 3, 20 answer 2, and 15 answer 1.
- The ratings add up to 1,510, so the average CES is 1,510 ÷ 300 = 5.03.
- Easy answers (5 to 7) are 210, so 70% of customers found it easy.
- Difficult answers (1 to 3) are 60, or 20%. The net score is 70 − 20 = +50.
Whichever method you pick, it helps to keep it fixed and to say which way the scale runs. On some surveys 1 means easiest, so a lower score is better.
CES survey questions and scales
A common wording asks, on a scale from "very easy" to "very difficult", how easy it was to deal with the company (Qualtrics). Other versions ask the customer to agree or disagree with a statement that the company made it easy to handle their issue.
Scales of 1 to 5 and 1 to 7 are common. The survey works best right after a single task, such as a return, a password reset, or a support conversation, so the customer rates one experience. It also keeps the result useful: a high-effort score on returns points straight at the returns process.
CES, CSAT, and NPS
CES measures how easy the experience was. It points to process friction.
CSAT measures satisfaction with an interaction. It reflects tone and outcome as well as effort.
Net Promoter Score measures willingness to recommend. It describes the relationship over time.
A customer can be satisfied with a polite agent and still report high effort, because it took three contacts to get there.
How AI agents change customer effort
Most effort in customer service comes from a few sources: waiting in a queue, repeating details after a transfer, being sent to an article that does not answer the question, and coming back a second time.
An AI agent can remove several of them at once. It answers without a queue, completes simple tasks inside the conversation, and passes the full context to a person when it hands over (see intelligent handoff). CES also catches a weak spot of deflection rate: a customer who read an article and then had to email anyway counts as deflected, and will usually report high effort.
Effort also shows up in behavior data, without a survey: repeat contacts within a few days, transfers per conversation, and channel switches for the same issue. Tracking them next to CES helps confirm what the survey says.
Frequently asked questions
What is a good customer effort score?
It depends on the scale, the method, and the contact reason, so there is no universal target. A useful starting point is your own baseline per contact reason, then the trend after each process change.
Is a high or a low CES better?
It depends on how the scale runs. On most surveys a higher number means an easier experience, but some use 1 for "very easy". It is worth checking the scale before comparing two reports.
What is the difference between CES and CSAT?
CES asks how easy it was to get something done. CSAT asks how satisfied the customer was. The two often move together, and when they diverge, the difference usually points to a process problem.
When should you send a CES survey?
Right after a specific task or support interaction, while the customer still remembers the steps. Surveys sent days later tend to mix the experience with other contacts.
How we approach this at Zowie
Zowie builds AI agents for customer service on voice, chat, and email. The language model speaks. A separate, deterministic Decision Engine returns every amount, rate, disclosure, and escalation from your rules, so the agent can finish the task in the conversation and hand over with context when a person is needed. You are welcome to book a demo.
Explore: CSAT, Net Promoter Score, First contact resolution, Intelligent handoff, Deflection rate, Customer self-service, Customer experience