Fyncall wrote this page. Definitions and methods link to the sources we used, which were checked on 19 November 2026. The worked examples use made-up numbers to show the arithmetic, not real results.
What is customer effort score?
Customer effort score is a single-question metric for how easy or hard it was to deal with you. Qualtrics describes CES as “a single-item metric that measures how much effort a customer has to exert” to get a task done.
For a Shopify store, the task is usually a support request: finding out where an order is, changing an address, or starting a return. CES asks whether that was easy, not whether the customer liked you.
Where did customer effort score come from?
Qualtrics credits researchers from CEB in 2010 with the finding that reducing customer effort predicted loyalty better than delighting customers. In the same year, Harvard Business Review published “Stop Trying to Delight Your Customers” by Matthew Dixon, Karen Freeman and Nick Toman (July to August 2010 issue), which questions the common belief that companies must delight their customers.
How do you ask a customer effort score question?
Ask one question about how easy the interaction was, and give a scale with clearly labelled ends. Qualtrics gives the typical wording as “on a scale of ‘very easy’ to ‘very difficult’, how easy was it to interact with [company name]”.
Common formats include:
- Numbered: a 1 to 5 scale. Zendesk's example runs from 1 (easiest) to 5 (most difficult), so check which end is “good” in the tool you use.
- Labelled: very difficult, difficult, neutral, easy, very easy.
- Faces: Medallia notes that emoji faces can replace numbers, as long as each end of the scale is clear.
The question can also target one part of the experience. Medallia's examples include “On a scale of 1 to X, how quickly were you able to solve the problem you experienced?” For a store, something like “How easy was it to get help with [order number] today?” keeps the answer tied to one conversation.
How do you calculate customer effort score?
There is no single standard formula, so pick one method and keep it. The two common approaches are an average of all ratings, or the share of customers who found it easy.
| Method | Formula | Source |
|---|---|---|
| Average rating | Sum of all CES ratings ÷ number of responses | Zendesk |
| Share who found it easy | Customers who said it was at least somewhat easy ÷ all responses | Medallia |
| Positive minus negative | Percentage of positive responses minus percentage of negative responses (Zendesk gives this for emoji scales; neutral answers are left out) | Zendesk |
A made-up example: 40 customers answer on a 1 to 5 scale where 5 is very easy. 28 choose 4 or 5, 7 choose 3 and 5 choose 1 or 2. The share who found it easy is 28 ÷ 40 = 70%. Treating 4 and 5 as positive, 1 and 2 as negative and leaving out the neutral 3s, as Zendesk does for emoji scales, the positive minus negative score is 70% minus 12.5%, which is 57.5. Same answers, different numbers, so always say which method you used.
How is CES different from CSAT and NPS?
CES asks how easy something was, CSAT asks how satisfied the customer was, and NPS asks how likely they are to recommend you. The first two are usually about one interaction, while NPS is about the whole relationship.
| Metric | Typical question | Scale and score |
|---|---|---|
| CES | How easy was it to interact with [company name]? | Very easy to very difficult (Qualtrics); scored as an average (Zendesk) or a share of easy answers (Medallia). |
| CSAT | How would you rate your overall satisfaction with the [goods/service] you received? | 1 to 5; 4 and 5 count as satisfied, divided by all responses. Qualtrics |
| NPS | How likely are you to recommend us to a friend or colleague? | 0 to 10; percentage of promoters (9 or 10) minus percentage of detractors (0 to 6). Bain & Company |
Qualtrics frames the split as CSAT capturing a “here and now” reaction to an interaction, NPS measuring long-term loyalty, and CES being suited to a single touchpoint. It also warns that CES “doesn't always give the full picture and should be used in tandem with NPS”, since a customer may have a good overall relationship with you and still have one hard interaction.
When should you send a CES survey?
Send it straight after the interaction you want to measure. Qualtrics says CES surveys “should be deployed immediately after interactions or specific touchpoints”.
- After a support chat closes: the customer still remembers how many steps it took.
- Once per conversation: repeated surveys are extra effort in themselves.
- Per channel: effort on web chat and effort on email can differ, so do not merge them into one score without saying so.
What makes support feel like high effort?
Effort usually comes from repetition and waiting: telling the same story twice, being passed between people, or chasing a reply. Low CES answers are worth reading alongside the conversation itself, because the cause is often visible there.
- Repeating details. A human handoff that loses context makes the customer start again.
- Slow first replies. Track first response time and resolution time next to CES.
- Switching channels. Being told to email when you asked on WhatsApp adds a step. See omnichannel support.
- Simple questions that need a person. Order status questions (WISMO) are a common example.
Can you measure CES in Fyncall?
Not as a dedicated survey type. Fyncall's web chat widget has a post-chat survey with three types: Stars (a 5-star scale), Thumbs (up or down) and NPS (0 to 10), as set out in the engagement and forms docs.
You write the survey question yourself, so you can ask an effort question on the Stars type, for example “How easy was it to get help today?”, with an optional written feedback box. Answers appear in the widget's Responses section, which shows survey responses and an average rating. That average works as the average-rating method above; for the share who found it easy, count the 4 and 5 star answers yourself. The web chat widget is included on every Fyncall plan, including Free, and the survey covers web chat only.
Common questions
What is a good customer effort score?
There is no single target that fits every store, and we do not quote benchmarks we cannot source. Compare your score with itself over time, using the same question, scale and formula.
Is a high CES good or bad?
It depends on the scale. If 5 means very easy, higher is better. If 1 means easiest, as in some vendors' examples, lower is better. State the direction whenever you share the number.
Should I replace CSAT with CES?
You do not have to choose. They answer different questions: CES shows where support was hard work, and CSAT shows how the customer felt about it. Asking both after every chat, though, adds effort, so consider asking one per touchpoint.
Does CES apply to AI replies?
Yes. Customers answering an AI agent still spend effort, so split scores for conversations the AI resolved and those passed to your team, and compare them.