Customer Effort Score Questions: What to Ask, When to Ask It, and Why Wording Matters

Most teams copy the standard CES question from a template and call it done. The problem is that small wording changes produce large score differences. 'How easy was it to resolve your issue?' and 'How much effort did you personally have to put forth?' measure different things and will give you different numbers from the same customers. If you change the question mid-way through a program, you cannot compare results. Get the wording right once, then never touch it.

This guide covers the exact customer effort score questions to ask after support, onboarding, billing, and feature releases, plus the follow-up question that tells you what to fix. All questions are calibrated against the original Gartner wording and tested for SaaS contexts.

The Standard CES Question

The original CES question, developed by CEB (now Gartner) and published in the Harvard Business Review, is:

The company made it easy for me to handle my issue. Rated on a 7-point scale from Strongly Disagree (1) to Strongly Agree (7).

This wording is deliberate. It asks about the company, not the individual agent. It uses an agreement scale, not a satisfaction scale. Both choices matter and both should be preserved.

1-7 scale vs 1-5 scale

Use 1-7. The original research used 7 points, most CES benchmarks are calibrated to 7 points, and switching to 1-5 makes your scores incomparable with industry data. The only reason to use 1-5 is consistency with existing CSAT surveys already running in the same tool. If you do switch, treat it as a fresh baseline and do not compare old scores to new ones.

Why "the company" not "your agent"

Asking about 'the company' rather than 'your agent' captures process friction, not just interpersonal friction. An agent can be professional and helpful while your support process still forces three follow-up contacts to resolve a single issue. You want to catch the process failures. Rating the person misses the systemic problem.

The Follow-Up Question (Never Skip This)

The CES score tells you there is friction. The follow-up tells you where. Always add one open-ended question immediately after the rating. The two options most commonly used:

What made this interaction difficult or easy?

Or, for surfacing friction specifically from low scorers: 'What made this harder than it should have been?' Do not ask both. One open-ended question gets answered. Two gets abandoned.

CES Questions After Support Tickets

Support is the most common CES use case and the easiest to instrument. Send CES within 24 hours of a ticket being marked resolved, not when the agent closes it on their end.

Primary question: "The company made it easy to resolve my support request." (1-7 agree/disagree)

Follow-up: "What made this harder or easier than expected?"

Variations for different support contexts:

• "The company made it easy to get help when I needed it." (broader than one ticket, good for post-conversation surveys)
• "How easy was it to find the answer you were looking for?" (for self-serve and help documentation)
• "The company resolved my issue without requiring extra effort on my part." (emphasises resolution completeness)
• "I was able to resolve my issue without needing to contact support again." (for first-contact resolution tracking alongside CES)

What to look for in support CES data

Look for low scores that cluster around the same product area or queue type. If your billing support CES is 4.8 while your technical support CES is 5.9, the problem is in billing workflows, not team quality. Segment CES by interaction category before drawing any conclusions about people or processes.

CES Questions After Onboarding

Onboarding is the highest-friction moment for most SaaS products. CES here predicts long-term retention more reliably than a 30-day NPS survey because it measures the exact moment users decide whether the product is worth the effort.

Send after the first meaningful activation moment, not at a fixed day-30 interval.

Primary question: "The company made it easy to get started and set up." (1-7 agree/disagree)

Follow-up: "What was the hardest part of getting set up?"

Variations:

• "I was able to get value from the product quickly and without difficulty." (if time-to-value is your core activation metric)
• "The company made it easy to understand how to use [core feature]." (for feature-specific onboarding flows)
• "How easy was the initial setup process?" (simpler phrasing for non-technical users or consumer products)

What onboarding CES tells you

If onboarding CES is below 5.0, users are struggling before they have seen value. This means churn within 30 to 60 days before you have had any chance to demonstrate the product. Fix onboarding friction before investing in any other retention program. The ROI is higher than any email sequence or CS outreach.

CES Questions After Billing and Upgrades

Billing interactions are often the most friction-heavy and the least surveyed. A failed upgrade or confusing invoice is a churn trigger that never appears in NPS because users leave quietly without explaining why. CES on billing interactions catches this before it becomes a cancellation.

Primary question: "The company made it easy to manage my subscription." (1-7 agree/disagree)

Variations:

• "The company made it easy to understand my invoice." (for billing confusion specifically)
• "Upgrading my plan required less effort than I expected." (for upgrade flow)
• "The company made it easy to resolve my billing issue." (for billing support contacts)

CES Questions After Feature Releases

Feature-level CES is the least common use case but often the most valuable for product teams. Send after a user completes a new feature for the first time, not after a product announcement email.

Primary question: "The company made it easy to use [feature name]." (1-7 agree/disagree)

Follow-up: "What would have made [feature name] easier to use?"

Use feature CES to validate releases in limited rollouts before a broad launch. If feature CES is below 5.0 during a limited rollout, you have a usability problem to fix before you announce to your full user base. This is a faster feedback loop than waiting for support ticket volume to spike.

The Most Common CES Mistake: Measuring Satisfaction Instead of Effort

Many teams write CES questions that accidentally measure satisfaction rather than effort. The difference matters because the score changes and the conclusions change.

The scale type also matters. CES uses an agreement scale (Strongly Disagree to Strongly Agree) because you are asking whether a statement is true. CSAT uses a satisfaction scale (Very Dissatisfied to Very Satisfied). Using a satisfaction scale with a CES question gives you a different distribution and makes your scores uninterpretable against benchmarks.

How Many Questions to Include in a CES Survey

Two. The rating question and one open-ended follow-up. That is the complete CES survey.

If your CES survey has more than two questions, you are adding friction to a survey that is supposed to measure friction. Users notice this, response rates drop, and the data you do collect is lower quality because users rush through to finish. The only exception: if you are tracking first-contact resolution alongside CES, you can add that as question two, with the open-ended as question three. Three questions is the absolute maximum for any CES survey.

How to Calculate CES from Your Responses

CES is the mean of all responses on the 7-point scale. The formula:

CES = Sum of all scores / Total number of responses

If you collect 100 responses and the sum of all scores is 560, your CES is 5.6. This is different from NPS, which uses a subtraction formula, and CSAT, which counts only top-box responses as a percentage. CES is a simple average, which makes it straightforward to segment by customer type, plan tier, or interaction category. Track it as a weekly rolling average rather than a point-in-time snapshot to reduce noise from low-volume weeks.

Where to Start

The standard CES question is short. The follow-up is shorter. The entire survey is two questions. The value is not in the survey design, it is in when you send it and what you do with the data.

Send it within 24 hours of a high-friction interaction. Read the open-ended responses in batches. Fix systemic issues first, not individual cases. A CES above 5.5 means you are reducing friction faster than it accumulates. Below 5.0 means users are likely leaving quietly before you have a chance to intervene.

Start with support ticket CES. It is the fastest to instrument, produces the most immediately actionable data, and has the clearest connection to churn. Once you have that baseline, expand to onboarding and then billing.