Reading your Customer Effort Score
CES distils one thing into a single number — how hard your customers had to work to get helped. Here's how it's calculated, what a good score looks like, and how it differs from CSAT and NPS.
What is customer effort score?
Customer effort score (CES) is a one-question survey metric that measures how much work a customer had to do to get their problem solved. It is asked straight after an interaction — a support chat, a return, a signup — while the effort is still fresh.
It exists because satisfaction turned out to be the wrong thing to optimise. The research behind CES found that delighting customers does far less for loyalty than simply not making things hard: the customers most likely to leave were not the unhappy ones, they were the ones who had to repeat themselves, switch channels, or chase an answer. Effort predicts churn earlier and more reliably than satisfaction does, which is why CES sits alongside CSAT and NPS rather than replacing either.
Customer effort score formula: how CES is calculated
After an interaction you ask one question — “[Company] made it easy to handle my issue” — on a 7-point scale (1 strongly disagree → 7 strongly agree). There are two ways to score the answers: take the average rating, or report the % who agreed (the top boxes). The calculator does both.
CES = Σ ratings ÷ responses
Average method
% easy = agree (5–7) ÷ total × 100
% easy (top-box) method
The customer effort score survey question
CES lives or dies on the wording. The standard phrasing, and the one the research validated, is an agreement statement rather than a question:
“[Company] made it easy for me to handle my issue.”
Strongly disagree (1) → strongly agree (7)
Three things to get right when you ship it:
Ask immediately, not weekly
CES is about one interaction. Sent a week later it measures memory, not effort — and response rates collapse.
Keep agree on the high end
Flipping the scale so 1 means easy is a common mistake and it inverts every number you compare against. Agreement high, always.
Add one open follow-up
“What made it hard?” is where the score becomes actionable. The number tells you there is friction; the free text tells you where.
What is a good customer effort score?
On a 7-point average, the scale runs 1 to 7 — and your live result lands on this same axis the moment you calculate. (On the % easy method, the bands are 80%+ excellent, 65–79% good, 50–64% moderate, and below 50% high effort.)
Excellent — Top-tier. Customers find it genuinely easy to get help.
Good — Solid. Most interactions feel low-effort to your customers.
Moderate — Workable, but customers are hitting friction worth removing.
High effort — Customers are working too hard — a clear churn risk.
Customer effort score examples
Same 100 responses, both scoring methods, so you can see how the two numbers relate. Ratings: twelve 7s, thirty-one 6s, twenty-two 5s, sixteen 4s, eleven 3s, five 2s, three 1s.
| Total responses | 100 |
| Sum of all ratings | 504 |
| Average method (504 ÷ 100) | 5.04 — Good |
| Agreed, rated 5–7 | 65 |
| % easy method (65 ÷ 100) | 65% — Good |
The bands differ by scale and method, so a 4.4 is a different verdict on a 5-point survey than on a 7-point one:
5-point average
Excellent — Top-tier. Customers find it genuinely easy to get help.
Good — Solid. Most interactions feel low-effort to your customers.
Moderate — Workable, but customers are hitting friction worth removing.
High effort — Customers are working too hard — a clear churn risk.
Top-box (% easy)
Excellent — Top-tier. Customers find it genuinely easy to get help.
Good — Solid. Most interactions feel low-effort to your customers.
Moderate — Workable, but customers are hitting friction worth removing.
High effort — Customers are working too hard — a clear churn risk.
How to improve your customer effort score
Effort is structural. It comes from the shape of your support, not from how politely your agents write, which is why a low CES rarely improves through coaching alone. These are the four sources that account for most of it.
The single biggest driver. Every time a customer re-explains their problem to a new agent or a new channel, effort goes up and the score goes down. Shared history beats better handoff notes.
Being told to email after starting in chat, or to call after emailing, reads as being passed around. Resolve where the customer started whenever the channel allows it.
Most high-effort interactions should never have been interactions. If the answer exists but is not findable, the customer pays for your search problem with their time.
A two-day resolution with a clear timeline feels lower-effort than a four-hour one in silence. Uncertainty costs more than delay.
The pattern behind all four is the same: the customer is doing work your system should be doing. A searchable help center removes the third one outright, and an AI agent grounded in it removes most of the first and the fourth — instant answers, with full context carried into the handover when a person is needed.
Customer effort score software
CES itself needs almost no tooling: one question, two numbers, a spreadsheet. What tooling buys you is the loop around it — triggering the survey after the right interactions, tying each score back to the ticket and the agent that produced it, and tracking the trend without anyone rebuilding a pivot table each month.
Three broad options. A dedicated survey tool sends and stores responses but knows nothing about your tickets. A help desk with surveys built in ties the score to the conversation, which is where the diagnosis actually happens. A CX platform adds cross-channel analysis at a price that only makes sense at volume. If you are starting out, the calculator above and a monthly export will tell you as much as a subscription will — start measuring, then buy the loop once the number is moving.
CES vs. CSAT vs. NPS
All three are one-question surveys, but they measure different things at different moments. CES is the best leading indicator for support: high effort predicts churn before satisfaction even drops.
Use CES right after a support conversation, CSAT for a specific touchpoint, and NPS for the overall relationship. Most teams run CES with one of the others rather than choosing a single metric — calculate all three with the CSAT and NPS calculators.
