Customer effort score (CES) is a customer experience (CX) metric that measures how easy customers found it to get something done with a company, such as resolving a problem, making a return or changing an account. It is collected through a short survey, usually one question, sent right after the interaction. Where satisfaction and loyalty surveys ask how customers feel, CES asks how hard the company made them work, which points directly at friction that can be fixed.
At a glance
- CES is usually a single survey question about ease, answered on a numeric or agree–disagree scale.
- It is sent right after a specific task or support interaction, not as a general relationship survey.
- Low scores point to concrete friction: transfers, repeated information, confusing steps or channel switching.
- Scales and formulas vary between companies, so internal trends matter more than outside benchmarks.
- CES is typically used alongside CSAT and NPS rather than instead of them.
What problem it solves
Customers often leave not because of one bad moment but because doing business with a company is tiring: repeating their details, being transferred between departments, waiting on hold, or being told to call back or try a different channel. Broad satisfaction scores can stay acceptable while this friction builds.
CES makes effort visible at the level where it can be fixed. By asking about a specific task, it shows which journeys, channels and processes are hard, for example a password reset that forces customers to call, or a claim process that needs several contacts. Effort also has a direct cost: repeat contacts, escalations and channel switching add volume for the business as well as frustration for the customer.
How it works
Question design. A common form asks customers to rate their agreement with a statement such as “The company made it easy for me to handle my issue,” on a scale from strongly disagree to strongly agree, often 1 to 7. Other versions ask “How easy was it to…?” on a shorter scale. Small wording changes can shift results, so pick one version and keep it.
Timing and trigger. The survey is sent automatically after a defined event: the end of a call or chat, a completed return, a submitted form. Frequency caps keep the same customer from being surveyed too often.
Scoring. The score is usually the average rating or the percentage of respondents who agree that the task was easy. Looking at the distribution helps, because a decent average can hide a group of customers with very high effort.
Analysis. Results are broken down by journey, channel, queue and team, and linked to operational data such as transfers, first contact resolution (FCR) and repeat contacts. Optional comments explain the score; natural language processing (NLP) tools can group them into themes at volume, though people should check the themes before acting.
Action. Findings go to the teams that own the friction, which may be support, product, billing or policy owners, not just the contact center. Customer journey mapping is often used to trace where effort comes from.
When it matters for buyers
- Choosing a contact center or survey platform. Check whether post-interaction CES surveys are built in, across voice and digital channels, and whether results link to the interaction record.
- Launching self-service. CES is a direct way to test whether customer self-service or a new bot made tasks easier or simply moved the effort.
- Contracting with an outsourcer. If CES is part of a performance agreement, define the question, scale, sample and formula in the contract.
- Comparing channels. CES helps show whether moving customers to chat or messaging actually reduced effort.
Our contact center as a service overview covers platforms with built-in post-interaction surveys and analytics.
Questions to ask vendors
- Can we run CES surveys automatically after calls, chats, messages and self-service sessions?
- Can we choose our own question wording and scale, and keep it consistent across channels?
- Are survey results linked to the interaction record, agent, queue and transcript?
- Can we report CES alongside CSAT, NPS, first contact resolution and transfer rates?
- Do you analyze free-text comments, and can we review and correct the themes?
- How do you control survey frequency to avoid over-surveying customers?
How it differs from CSAT
Customer satisfaction (CSAT) asks how satisfied a customer was with an interaction or product; CES asks how easy it was. The two often move together but not always: a customer can be satisfied with a friendly agent yet still have had to call three times, and a quick, easy transaction can still leave them unhappy with the outcome. CES is better at locating process friction; CSAT captures the overall reaction, including outcome and tone. Both differ from Net Promoter Score (NPS), which asks about willingness to recommend and usually tracks the overall relationship rather than a single task. Many teams use all three, each matched to the decision it informs.
Problems this helps with
- Customers say we are hard to deal with but we can't pinpoint where
- Satisfaction scores look fine but customers still leave
- Too many repeat calls and channel switching for simple tasks
- Need to prove a new self-service tool actually made things easier






