What is the Customer Effort Score (CES)?

Why measure effort?

The Customer Effort Score comes out of research by the Corporate Executive Board (CEB, now part of Gartner). In 2010, Matthew Dixon, Karen Freeman and Nick Toman wrote it up in the Harvard Business Review under a provocative title: Stop Trying to Delight Your Customers. Their argument: in customer service, you win loyalty less by surprising customers with extras than by making things easy for them. The idea that you must "delight" customers had, in the authors' view, become so entrenched that hardly anyone questioned it.

Anyone who has had to give their customer number three times, or ended a web chat only to have to phone anyway, knows the feeling. The outcome can be fine and the contact can still leave a bad taste. A satisfaction question sometimes misses that; a question about effort catches it exactly.

The CES question: two versions

The first version asked customers how much effort they personally had to put in to get their issue handled, on a scale where a low number meant little effort. That runs against intuition: on almost every other question, high is good.

In The Effortless Experience (2013) the researchers therefore set out a second version: a statement customers agree or disagree with, on a scale from 1 to 7. A widely used wording:

[Company] made it easy for me to handle my issue.

Adapt the end of the statement to the task you're measuring: to return my order, to book an appointment, to set up my account. The more specific it is, the more useful the result.

How to calculate CES

Again there are two methods. You can report the average score, or the percentage of customers who agree (or strongly agree) with the statement, that is, who gave a 6 or a 7. The percentage is easier to read and to compare:

A worked example

An online shop asks customers whether it was easy once they've completed a return. 120 customers respond:

• CES as a percentage: 30 + 36 = 66 of 120 customers found it easy, so a CES of 55%.
• Average score: 628 ÷ 120 = 5.23 on a scale of 1 to 7.

The bottom end is worth a look here too: 20 customers (17%) gave a 3 or lower. What went wrong for them? That's where the improvements are.

When to use CES

CES suits tasks: moments when a customer wants to get something done. Think of:

• resolving a question or complaint with customer service;
• a return, a change or a cancellation;
• onboarding: creating an account, installing something, logging in for the first time;
• a form or application process on your website.

Ask straight after the task, and measure each task separately. An average CES across everything won't tell you which process is causing friction. For the overall relationship with a customer, NPS is the better fit; for satisfaction with the outcome, CSAT. NPS vs CSAT vs CES explains how they combine into a complete picture.

What is a good CES?

Because CES is measured per task, and different scales and methods are in use, there's no generally useful norm. A cancellation will rarely feel as smooth as a purchase. So compare:

• the same task over time, for example before and after you change your returns process;
• different channels for the same task, such as chat, phone and the online form;
• different steps in the customer journey, to see where most of the effort sits.

Pitfalls

Mixing versions. A CES on the old scale (low is good) and one on the new scale (high is good) can't be compared, and a report that puts them side by side will mislead its readers. Choose the statement version and stick with it.

A vague statement. "[Company] makes things easy for me" without a task measures a general feeling, not a process. Name the task.

Only measuring those who succeeded. Customers who gave up halfway had the most effort of all. Try to reach them too, for instance with a question after an abandoned application.

Measuring effort without asking what was hard. A low CES tells you something is grinding, not what. Was it the wait, being transferred, the form, having to repeat details?

After the score: ask what was hard

With CES the follow-up question matters even more than with other scores, because effort is so concrete. Customers can usually point to exactly where it went wrong: "I had to give my order number twice", "the form wouldn't accept my postcode". So ask: what made it difficult? or what would have made it easier?