Customer Effort Score (CES): The Complete Guide

CES has its roots in a 2010 Harvard Business Review article, 'Stop Trying to Delight Your Customers,' by Matthew Dixon, Karen Freeman and Nicholas Toman of the Corporate Executive Board. Their research made a counterintuitive argument: going above and beyond to delight customers does surprisingly little for loyalty, while making things hard — repeating information, switching channels, jumping through hoops — reliably drives customers away. The most powerful thing a company can do to keep customers, they found, is simply to reduce the effort it takes to do business with them.

That insight reframed loyalty around a single idea: effort is a liability. Every extra step, every repeated explanation, every dead end quietly erodes the relationship, even when the customer eventually gets what they wanted. CES was created to make that hidden cost visible and measurable, so teams can find the high-effort moments in a journey and remove them before they cause churn.

The link to loyalty is what makes CES so useful. A low-effort experience does not just feel good in the moment — it reduces the reasons a customer has to look elsewhere. For a fuller look at that mechanism, see reducing customer effort.

Effort shows up everywhere in the customer journey, and each high-effort moment is a place where you can lose someone. CES is valuable precisely because you can point it at any of these touchpoints and get a clean, comparable read on how hard that specific task was.

• Support — the original home of CES. Resolving an issue in one contact, without repeating information or being transferred, is the single biggest driver of a low-effort support experience.
• Checkout — every extra field, forced account creation or confusing step adds effort and abandoned carts. CES after purchase surfaces exactly where buyers struggle.
• Delivery — tracking that is hard to find, unclear timing or a complicated handover all raise perceived effort even when the product arrives fine.
• Returns — a returns process is often a customer's most effortful interaction. An easy return builds more loyalty than a smooth purchase.
• Self-service — help centers, chatbots and knowledge bases exist to lower effort; CES tells you whether they actually do, or whether people give up and contact you anyway.
• Mobile — small screens magnify friction. Tasks that are easy on desktop can be high-effort on a phone, so measuring effort by device matters.
• Retention — because effort predicts loyalty so strongly, rising effort scores are an early warning that churn is coming, often before satisfaction or NPS moves.

The practical payoff is prioritization. When you measure effort across touchpoints, you can rank them by how hard they are and fix the worst offenders first — the moments doing the most quiet damage to retention.

CES starts with a single agreement question about how easy a task was. The modern, widely used wording is an ease statement that the customer rates:

To what extent do you agree with the following statement: [Company] made it easy for me to handle my issue. (Strongly disagree → Strongly agree)

The response is captured on a rating scale. Two scales are common: a 5-point scale (1–5) and a 7-point scale (1–7), each running from strong disagreement (high effort) up to strong agreement (low effort). On Surveys.expert, the CES question uses a 1–7 scale labelled from 'Very difficult' to 'Very easy', where higher numbers mean an easier, lower-effort experience.

The most common way to report CES is the simple average of all responses. Add up every rating and divide by the number of responses:

CES = sum of all effort ratings ÷ number of responses

For example, on our 1–7 'easy is high' scale, if ten customers give ratings of 7, 6, 7, 5, 6, 7, 4, 6, 7 and 5, the total is 60 and the average CES is 6.0 — a strong, low-effort result. Interpreting that number depends entirely on the scale direction: because higher means easier here, 6.0 is good, whereas the same 6.0 on a reversed 'high means more effort' scale would signal a problem. Skip the arithmetic with the CES calculator, which returns the average and distribution from your raw ratings, and see customer effort score calculation for worked examples on both scales.

The rating question gives you the number; a short, well-placed follow-up gives you the reason. The strongest CES surveys pair the ease rating with a single open-text question so you learn what created the effort, not just how much there was.

• The core CES question — 'To what extent do you agree: [Company] made it easy to handle my request?' rated on the ease scale. This is the number.
• A task-specific variant — 'How easy was it to complete your purchase today?' or 'How easy was it to return your item?' when you want the question tied to one touchpoint.
• The universal follow-up — 'What made it easy or difficult?' (open text). This one question turns the score into specific, fixable friction and belongs in every CES survey.
• A low-score probe — for anyone who reports high effort, 'What was the most difficult part of this experience?' to pinpoint the exact step that failed.
• An optional resolution check — 'Was your issue fully resolved?' (yes/no) to separate effort from outcome, since a hard experience with a good result is different from a hard one with no resolution.

Keep the question focused on ease, not overall happiness — that is what distinguishes CES from satisfaction. For a full bank of wording by touchpoint, see customer effort score questions.

A great CES survey is short — usually the ease rating plus one follow-up, sent right after the interaction it measures. Here is a complete, ready-to-use structure with light conditional logic, written for a support interaction:

1. 1[Company] made it easy for me to handle my issue. (1–7 scale, Very difficult → Very easy, required) — the core CES question.
2. 2What made your experience easy or difficult? (open text) — shown to everyone; the single most valuable follow-up.
3. 3If high effort (rated 1–3): What was the most difficult part of resolving your issue? (open text)
4. 4If low effort (rated 6–7): What made it so easy? (open text) — captures what to protect and repeat.
5. 5Was your issue fully resolved? (yes/no, optional) — separates effort from outcome.
6. 6May we follow up with you about your feedback? (yes/no, optional)

You do not have to build a CES survey from scratch. A good starting template already includes the correctly-worded ease question on the right scale, the open-text follow-up and sensible defaults, so you can launch in minutes and customize from there.

• Start with the ready-made CES survey template — the ease question plus follow-up, editable end to end.
• Match the survey to the moment with real examples for support, checkout, delivery, returns and more in customer effort score examples.
• Browse the full template library for industry-specific survey variants.

• Asking too late — CES measures a specific interaction, so send it immediately after that interaction while the effort is fresh. A survey days later measures memory, not effort.
• Confusing satisfaction with effort — 'How satisfied were you?' and 'How easy was it?' are different questions. A customer can be satisfied with the outcome yet report high effort getting there; blurring the two hides the friction CES exists to find.
• Asking too many questions — CES works because it is short. Padding the survey with extra ratings lowers completion and dilutes the signal; keep it to the ease question and one or two follow-ups.
• Ignoring the comments — the open-text 'why' is where the fixable friction lives. Reporting only the average number tells you effort is high but nothing about which step to fix.
• Comparing scores across different scales or touchpoints — a checkout CES and a support CES, or a 1–5 and a 1–7 scale, are not the same measurement and should not be compared directly.

The hard part of a CES program has never been the average — it is reading the open-text answers to find out what is actually creating effort, and doing it fast enough to act. That is exactly what AI is good at.

• Sentiment analysis — the emotional tone of every written comment scored automatically, so you can triage frustration at volume instead of reading one comment at a time.
• Theme detection — recurring friction points surfaced across all the open-text answers, turning scattered complaints into a ranked list of what is making things hard.
• Friction summaries — a concise, per-touchpoint summary of where effort concentrates, so you can see at a glance which step is doing the most damage.
• Recommended actions — concrete suggestions for reducing the effort behind the biggest themes, connecting the score directly to what to fix.

Together these turn CES from a number on a dashboard into a working feedback loop: the score tells you the trend, and AI tells you the friction and the fixes. To go deeper, see how to improve customer effort score.

Ready to start? Create your CES survey with AI, calculate a score from existing data with the CES calculator, or read how effort compares to other metrics in customer satisfaction vs CES and NPS vs customer effort score.