metricData, experimentation & growth

Customer Effort Score (CES)

Customer Effort Score measures how easy or difficult customers perceive a specific task, interaction, or service experience to be.

Download Markdown (.md)

01What it means

Customer Effort Score is a survey metric for perceived ease or difficulty. It is commonly collected after a service interaction or task using an agreement or ease scale, but wording and scale direction vary, so the exact instrument must accompany the result.[1][2]

Perceived ease or difficulty after a defined interaction Place responses on one declared scale, then keep wording and direction stable
  1. 01Very low effortThe task or interaction felt easy
  2. 02Lower effortMinor friction without material obstruction
  3. 03Neutral midpointMeaning depends on the chosen scale
  4. 04Higher effortNoticeable work, confusion, or repetition
  5. 05Very high effortThe experience was difficult to complete

CES has no universal wording or scale direction. Report the instrument, distribution, response rate, sample, and context rather than a detached average.

02Why it matters

Customers often want to solve a problem with minimal friction rather than be delighted by an elaborate interaction. CES focuses teams on reducing avoidable work such as repetition, channel switching, unclear instructions, and repeated contacts that can damage an otherwise successful journey.

03When to use it

Use CES after a clearly bounded task or service interaction, such as onboarding, resolving a support issue, returning an item, or changing a plan. It is most useful when paired with behavioral evidence and when the survey arrives close enough to the experience for reliable recall.

04How it works

  1. Define the interaction, eligible population, survey timing, and the exact construct: ease, difficulty, or agreement that the company made the task easy.
  2. Use one consistent question and response scale; state clearly which direction represents lower effort.
  3. Summarize valid responses with the chosen statistic and report distribution, response rate, sample, and wording rather than publishing a context-free number.
  4. Segment results by journey, channel, customer group, and outcome to locate friction.
  5. Investigate causes with operational data and qualitative research, then remeasure after changes using the same instrument.

05A practical example

Example

After a support case closes, a telecom asks whether the company made the issue easy to resolve and invites a comment. Low scores cluster among customers transferred between billing and technical support. The team redesigns routing and ownership, then compares the same CES item and repeat-contact rate after release.

06Where it breaks

There is no single universally administered CES instrument, so scores from different questions or scales may not be comparable. Surveys capture perception and suffer from timing, cultural, and nonresponse biases. Low effort can coexist with a poor outcome, and some valuable tasks legitimately require thought.

Common mistakes

  • Comparing CES values without matching question wording, scale, and direction.
  • Treating effort as a substitute for task success or outcome quality.
  • Surveying long after the interaction or only sampling customers who reached completion.

07Sources

  1. Stop Trying to Delight Your CustomersHarvard Business Review · Foundational research articleStop Trying to Delight Your Customers
  2. Stop Trying to Delight Your CustomersHarvard Business Publishing · Original article recordStop Trying to Delight Your Customers
View saved topics

Start typing to search titles, definitions, categories, and frameworks.