40+ CSAT Survey Questions (+ Templates & Scoring Guide)
Johannes
CEO & Co-Founder
12 Minutes
June 5th, 2026
Most CSAT programs fail not because teams stop caring but because they ask the wrong questions at the wrong time and then calculate the score incorrectly. Fornell, Johnson, Anderson, Cha, and Bryant (1996), in their foundational paper introducing the American Customer Satisfaction Index published in the Journal of Marketing, found that customer satisfaction drives retention, reduces price sensitivity, and lowers the cost of future transactions when measured at the right level of specificity. The specificity part is what most CSAT surveys miss. A vague "how did we do?" survey produces vague data. This guide gives you 40+ CSAT survey questions that produce data you can actually act on, organized by touchpoint with scoring formulas and industry benchmarks.
What Is a CSAT Survey and How Is It Different from NPS and CES?
CSAT (Customer Satisfaction Score) is a transactional metric. It measures how satisfied a customer was with a specific interaction, such as a purchase, a support call, or an onboarding session.
It does not measure overall brand loyalty (that is NPS). It does not measure how hard an interaction was to complete (that is CES). Each metric answers a different question.
| Metric | What It Measures | Scale | Formula | Best Used After | Good Score |
|---|---|---|---|---|---|
| CSAT | Satisfaction with specific interaction | 1-5 | (4+5 responses / total) x 100 | Purchase, support, onboarding, product use | >75% |
| NPS | Overall loyalty and likelihood to recommend | 0-10 | % Promoters minus % Detractors | Quarterly relationship check | +30 or higher |
| CES | Effort required to complete a task | 1-7 | Average of all scores | Support resolution, task completion | 5.5+ (out of 7) |
The three metrics work together. CSAT tells you how a moment went. NPS tells you where the relationship stands. CES tells you where friction is hiding. For deeper coverage of NPS question design, see our NPS question examples guide.
CSAT's structural limitation is that it only captures the customer's view of a single touchpoint. A customer can give 5/5 for a support interaction and churn the next week because the product is not delivering value. This is why CSAT should always be one part of a broader customer experience measurement program.
The CSAT Scoring Formula (and What Scores Actually Mean)
The formula is simple. The interpretation is where teams go wrong.
CSAT formula:
- Ask customers to rate their satisfaction on a 1-5 scale
- Count responses of 4 (Satisfied) and 5 (Very Satisfied)
- Divide by total number of respondents
- Multiply by 100
Example: 150 customers respond. 60 select 5, 55 select 4, 20 select 3, 10 select 2, 5 select 1. CSAT = (60 + 55) / 150 x 100 = 76.7%.
Industry Benchmarks
The following benchmarks come from the American Customer Satisfaction Index (ACSI), which has tracked satisfaction across US industries since 1994:
| Industry | ACSI Average CSAT |
|---|---|
| Ecommerce / Online Retail | 78-82% |
| Software / SaaS | 74-78% |
| Banking and Financial Services | 75-78% |
| Hotels | 73-76% |
| Airlines | 72-75% |
| Telecom / Internet Service | 62-68% |
| Government Services | 60-65% |
| Healthcare | 71-75% |
Score Interpretation
- Below 60%: Critical problems, immediate investigation required
- 60-70%: Below industry average, improvement is urgent
- 70-75%: Average, most customers are satisfied but the margin is thin
- 75-80%: Good, meeting expectations for the majority of customers
- 80-85%: Excellent, customers are genuinely happy
- Above 85%: World-class performance, protect what you are doing
A single score is a snapshot. The trend is what matters. Track CSAT monthly at each touchpoint and look for sudden drops (investigate what changed) or gradual declines (segment by customer type and channel to find the source).
The Ceiling Effect: Why High CSAT Scores Are Often Misleading
Most CSAT scores cluster above 70%. This is not because most companies are excellent at customer satisfaction. It is a measurement artifact called the ceiling effect.
Three forces drive CSAT scores upward regardless of actual experience quality:
Response bias. Satisfied customers are more likely to complete surveys than dissatisfied ones. Customers who had a frustrating experience often leave before a survey can be triggered, churn before it is sent, or simply ignore it.
Social desirability bias. People lean toward positive ratings when interacting with a company representative or when they feel their identity is attached to the response. Oliver (1980), in his foundational paper on the cognitive antecedents of satisfaction in the Journal of Marketing Research, showed that satisfaction judgments are influenced not just by the actual experience but by expectations and the framing of the evaluation itself.
Scale compression. On a 1-5 scale, most customers who bother to respond choose 3, 4, or 5. The bottom two options are psychologically associated with strong negativity and require active effort to select.
What this means in practice: A CSAT of 72% from a carefully designed, anonymized, post-interaction survey is more meaningful than a CSAT of 84% from a survey embedded in a happy-path confirmation email sent only to recent purchasers. Methodology matters as much as the number.
To counter the ceiling effect: anonymize surveys where possible, survey at multiple touchpoints including friction points, and segment your CSAT by customer outcome (did they churn within 90 days?) to understand what scores actually predict.
40+ CSAT Survey Questions by Category
Each question below includes a question type and a priority rating: Essential (include in every survey of this type), Recommended (include when relevant), or Nice-to-have (include if survey length allows).
Core CSAT Questions (Use Across All Survey Types)
These are the foundational questions that appear in most CSAT programs. Use them as your baseline metrics.
1. How satisfied are you with your overall experience with [company/product]?
- Type: Likert (1-5) | Essential
- Your primary CSAT metric. Track this quarterly across all touchpoints to establish a baseline and measure whether changes move the number.
2. How would you rate the quality of [product/service] you received?
- Type: Likert (1-5) | Essential
- Quality perception is the strongest driver of satisfaction scores. Fornell et al. (1996) found that customization and reliability are the two primary quality dimensions that explain CSAT variance across industries.
3. How well did [product/service/interaction] meet your expectations?
- Type: Scale (Exceeded / Met / Fell short) | Recommended
- Expectation gaps are the root cause of low CSAT scores. "Fell short" responses identify where your marketing or onboarding over-promises.
4. How likely are you to continue using [product/service]?
- Type: Likert (1-5) | Essential
- A leading indicator of churn. Customers who score 1-2 here are at serious risk and warrant immediate outreach.
5. What is the main reason for your satisfaction rating?
- Type: Open-ended | Essential
- Always pair a score question with this. The number tells you where you stand. This tells you why. Without the "why," CSAT data cannot drive action.
Post-Purchase CSAT Questions
Send these 3-7 days after delivery or product activation, once the customer has had time to experience what they bought.
6. How satisfied are you with your recent purchase from [company]?
- Type: Likert (1-5, CSAT) | Essential
- The core post-purchase metric. Cross-reference with delivery and product quality questions to diagnose whether dissatisfaction is about the product itself or the buying process.
7. How easy was it to complete your purchase?
- Type: Likert (1-7, CES) | Recommended
- Checkout friction is a top driver of cart abandonment and reduces post-purchase satisfaction even when the product is good.
8. How satisfied are you with the delivery time and condition of your order?
- Type: Likert (1-5) | Recommended
- Delivery is the moment of truth between purchase and product use. A damaged or late delivery colors the entire experience.
9. How confident do you feel about your purchase decision?
- Type: Likert (1-5) | Recommended
- Post-purchase cognitive dissonance is real. Low confidence scores indicate your onboarding or post-purchase communication is not reinforcing the value of their decision.
10. How would you rate the packaging and presentation of your order?
- Type: Likert (1-5) | Nice-to-have
- Most valuable for direct-to-consumer or premium brands where unboxing is part of the product experience.
11. Based on your purchase experience, how likely are you to buy from us again?
- Type: Likert (1-5) | Essential
- Repeat purchase intent is the most direct post-purchase retention metric. Segment by product category, price point, and customer tenure to find patterns.
12. What could we have done to make your purchase experience better?
- Type: Open-ended | Recommended
- Post-purchase is the highest-intent moment for specific, actionable feedback. Customers have just committed money and have strong opinions about the experience.
Post-Support CSAT Questions
Send these within 1-2 hours of ticket resolution while the interaction is fresh. For chat support, trigger immediately after the session closes.
13. How satisfied are you with the support you received?
- Type: Likert (1-5, CSAT) | Essential
- The primary support quality metric. Benchmark separately from your product CSAT because support satisfaction has different drivers. For a complete set of support survey questions, see our customer service survey questions guide.
14. Was your issue fully resolved?
- Type: Scale (Yes / Partially / No) | Essential
- Resolution rate is the strongest predictor of support CSAT. "Partially" responses often indicate systemic problems where agents lack the authority or tools to complete a resolution.
15. How much effort did you have to put in to get your issue resolved?
- Type: Likert (1-7, CES) | Essential
- Customer effort in support is a stronger predictor of disloyalty than satisfaction alone. Customers who experience high-effort resolutions report being disloyal at much higher rates than those with low-effort resolutions, as documented by Dixon, Freeman, and Toman (2010) in their Harvard Business Review research on customer effort.
16. How knowledgeable was the support agent who helped you?
- Type: Likert (1-5) | Recommended
- Agent knowledge is consistently the top driver of support satisfaction. Low scores here point to training gaps or an inadequate knowledge base.
17. How satisfied are you with the time it took to resolve your issue?
- Type: Likert (1-5) | Recommended
- Response time expectations vary by channel: chat (under 5 minutes), email (under 4 hours), phone (under 2 minutes). Measure channel-specific satisfaction separately.
18. How satisfied are you with how the agent communicated with you throughout the process?
- Type: Likert (1-5) | Recommended
- Communication quality predicts satisfaction independently of resolution speed. Customers tolerate slow resolutions better when they are kept informed.
19. Would you use this support channel again for a similar issue?
- Type: Binary (Yes/No) | Nice-to-have
- Channel preference data. "No" responses indicate customers feel their issue was a mismatch for the channel they used.
20. What one thing could we do to improve our support experience?
- Type: Open-ended | Essential
- Support feedback is the most specific and actionable feedback customers give. They know exactly what went wrong. Group responses by theme to identify patterns.
Post-Onboarding CSAT Questions
Send these 7-14 days after signup or activation, once the customer has had time to attempt their first meaningful outcome.
21. How satisfied are you with the onboarding experience so far?
- Type: Likert (1-5, CSAT) | Essential
- Baseline onboarding satisfaction. Low scores at this stage are an early churn signal that requires immediate intervention.
22. How easy was it to get started with [product]?
- Type: Likert (1-7, CES) | Essential
- Onboarding effort predicts early churn better than almost any other metric. If customers struggle in the first week, many never reach the value that would retain them.
23. Have you been able to accomplish your main goal with [product] yet?
- Type: Binary (Yes/No) + conditional follow-up | Essential
- Time-to-value is the most important onboarding metric. "No" responses need follow-up: "What is blocking you?" This is the highest-ROI support conversation you can have.
24. How helpful were the setup guides and documentation?
- Type: Likert (1-5) | Recommended
- Documentation quality directly affects onboarding effort. Low scores here correlate with high early-stage support ticket volume on the same topics.
25. Did [product] meet the expectations you had when you signed up?
- Type: Scale (Exceeded / Met / Fell short) | Recommended
- Early expectation alignment is the strongest predictor of 90-day retention. "Fell short" responses in the first two weeks almost always become churned customers by month three.
26. What was the biggest challenge you faced in getting started?
- Type: Open-ended | Essential
- Identifies the friction points blocking adoption. Group by theme to prioritize onboarding improvements with the broadest impact.
Product Experience CSAT Questions
Use these in periodic in-product surveys, 30-90 days after activation, once the customer is past initial setup.
27. How satisfied are you with [product] overall?
- Type: Likert (1-5, CSAT) | Essential
- Your primary product satisfaction metric. Segment by plan, use case, and tenure to find which segments are most and least satisfied.
28. How reliable is [product] in your day-to-day use?
- Type: Likert (1-5) | Essential
- Reliability is a non-negotiable baseline. Inconsistent performance erodes satisfaction faster than any missing feature.
29. How satisfied are you with the features available in [product]?
- Type: Likert (1-5) | Recommended
- Feature satisfaction captures whether your roadmap matches what customers need. Low scores paired with high quality scores signal missing features, not broken ones.
30. How easy is [product] to use?
- Type: Likert (1-7, CES) | Essential
- Usability is a key satisfaction driver. Products that are powerful but complex lose customers to simpler alternatives. Cross-reference with tenure to see if usability perception improves over time.
31. How satisfied are you with the value you get relative to what you pay?
- Type: Likert (1-5) | Essential
- Price-quality perception determines whether customers feel the product is worth renewing. Low scores here are a stronger churn predictor than product quality scores alone.
32. How satisfied are you with the frequency and quality of product updates?
- Type: Likert (1-5) | Nice-to-have
- Too few updates signal stagnation. Too many cause change fatigue. This question calibrates your release cadence against customer expectations.
33. If you could change one thing about [product], what would it be?
- Type: Open-ended | Essential
- The single-constraint format forces prioritization. More actionable than "what would you improve?" which produces scattered wish lists.
General Relationship CSAT Questions
Use these in quarterly relationship surveys to measure how the overall experience is trending. Keep to 5-7 questions maximum.
34. How satisfied are you with your overall relationship with [company]?
- Type: Likert (1-5, CSAT) | Essential
- The relationship-level counterpart to transactional CSAT. Tracks whether your brand is building or eroding goodwill over time.
35. How well does [company] understand your needs?
- Type: Likert (1-5) | Recommended
- Perceived understanding predicts loyalty. Customers who feel understood by a vendor are significantly less likely to evaluate alternatives.
36. How likely are you to recommend [company] to a colleague or peer?
- Type: Rating (0-10, NPS) | Essential
- Include NPS in relationship surveys to pair satisfaction data with loyalty data. A high CSAT paired with a low NPS is the signature of the satisfaction-loyalty gap described later in this guide.
37. How has your satisfaction with [company] changed over the past 6 months?
- Type: Scale (Decreased significantly / Decreased somewhat / Stayed the same / Increased somewhat / Increased significantly) | Recommended
- Satisfaction trajectory matters more than a point-in-time score. "Decreased somewhat" is an early warning sign that needs investigation before it becomes "decreased significantly."
38. Have you considered switching to a competitor in the past 6 months?
- Type: Binary (Yes/No) + conditional open-ended | Recommended
- A "Yes" paired with "Why?" reveals your competitive vulnerabilities. Even satisfied customers evaluate alternatives. Understanding the trigger helps you address it proactively.
39. What do you value most about your experience with [company]?
- Type: Open-ended | Recommended
- Identifies your true differentiators from the customer's perspective. These are the things to protect and amplify in your messaging and product roadmap.
40. What is the one thing [company] should do differently to better serve you?
- Type: Open-ended | Essential
- The single most actionable question in any relationship survey. Group responses by theme to identify systemic gaps that no single touchpoint survey surfaces.
Bonus questions for specific contexts:
41. How satisfied are you with our communication about [specific change or update]? (Use after pricing changes, product updates, or policy changes)
42. How satisfied are you with the renewal or upgrade process? (Use at contract renewal or upsell moments)
43. How well did [event/webinar/training] meet your learning objectives? (Use post-event for education touchpoints)
44. How satisfied are you with the speed and accuracy of our invoicing and billing? (Use for B2B customers where billing is a friction point)
Question Types and When to Use Them
| Question Type | Best For | Pros | Cons |
|---|---|---|---|
| Likert 1-5 | CSAT, quality, satisfaction | Easy to benchmark, familiar to respondents | Can be gamed by survey design |
| Likert 1-7 | CES, usability, effort | More granular, research-validated for effort | Slightly more cognitive load |
| NPS 0-10 | Loyalty, relationship | Widely benchmarked, segmentation-ready | Does not explain individual interactions |
| Binary Yes/No | Resolution, preference | Simple, high response rate | No nuance |
| Open-ended | Root cause, discovery | Surfaces unknown issues | Low completion, requires qualitative analysis |
| Scale with labels | Expectation alignment | Clear anchors reduce ambiguity | Hard to aggregate across surveys |
For open-ended questions: Pew Research Center's survey methodology guidance documents that open-ended questions have significantly higher non-response rates than closed-ended questions. Limit them to 1-2 per survey and always place them after closed-ended questions, never before.
The Response Scale Problem: Why 1-5 and 1-10 Produce Different CSAT Scores
This is one of the most underappreciated issues in customer satisfaction measurement, and it directly affects how you interpret benchmark data.
The same experience, measured on a 1-5 scale versus a 1-10 scale, produces systematically different CSAT scores. This is not a statistical quirk. It is a documented phenomenon in survey methodology.
On a 1-5 scale, responses of 4 and 5 count as "satisfied" (the standard CSAT formula). On a 1-10 scale, there is no consensus definition of "satisfied." Some teams count 8-10. Some count 7-10. The choice changes the reported score by 10-20 percentage points without any change in actual customer experience.
Preston and Colman (2000), writing in Current Psychology on the optimal number of response categories, found that 5-point and 7-point scales consistently outperformed 2-point, 3-point, 10-point, and higher scales on reliability and validity measures, with no meaningful advantage from scales beyond 7 points for most satisfaction measurements.
Practical implication: Never compare your CSAT score against a benchmark from a company using a different scale. If a competitor reports 85% CSAT on a 1-10 scale (counting 8+) and you report 78% CSAT on a 1-5 scale (counting 4+), you are not measuring the same thing. Your internal trend, on a consistent scale and methodology, is your most reliable benchmark.
When you change your scale, your historical data becomes incomparable. Treat a scale change as a measurement restart and build a new baseline.
When to Send CSAT Surveys
Timing determines whether you get accurate feedback or noise. Send too soon, and customers are reacting to immediate emotion rather than considered judgment. Send too late, and recall decay reduces accuracy.
| Touchpoint | Ideal Send Time | Why |
|---|---|---|
| Post-support | Within 1-2 hours of resolution | Issue is fresh, emotions have settled |
| Post-purchase (digital) | Immediately after confirmation | Checkout experience is fresh |
| Post-purchase (physical) | 3-7 days after delivery | Customer has had time to use the product |
| Post-onboarding | 7-14 days after activation | Customer has attempted first meaningful outcome |
| Post-product use | 30-90 days after activation | Customer is past setup, using the product regularly |
| Relationship survey | Quarterly | Enough time has passed to reflect on the full relationship |
Never survey the same customer more than once per month. Surveying too frequently reduces response rates and trains customers to ignore survey requests.
For more on timing, frequency, and channel strategy, see our guide on how to increase survey response rates.
The Satisfaction-Loyalty Gap: When CSAT Goes Up but NPS Goes Down
This is one of the most counterintuitive patterns in customer experience data, and it signals a strategic problem that transaction-level CSAT cannot detect on its own.
The pattern: your CSAT scores improve across support, onboarding, and product touchpoints. Your NPS drops in the same period. Customers say each interaction is fine. They would not recommend you.
This is the satisfaction-loyalty gap. It happens when:
- Customers are satisfied with individual touchpoints but the cumulative experience does not deliver enough distinctive value to warrant advocacy
- Competitors have improved and the relative comparison has shifted, even if absolute satisfaction is stable
- Pricing has increased faster than perceived value, eroding the emotional surplus that drives promoter behavior
- The product has become commodity: it works, but so does the alternative
What to do: Stop optimizing individual interactions and look at the full journey. Use journey mapping to identify the gap between "meeting expectations at each step" and "delivering something worth recommending." Compare your retention cohort data with your NPS segmentation: if your promoters retain at much higher rates than your passives, the loyalty gap is actively costing you revenue.
Closing the feedback loop with customers who have given low NPS scores despite high CSAT scores is one of the most valuable conversations you can have. They will tell you exactly what is missing.
CSAT Survey Best Practices
Match the question to the touchpoint. A post-support survey should ask about the support experience, not general product satisfaction. Generic satisfaction questions at specific touchpoints produce averages that explain nothing.
Keep surveys short. Transactional CSAT surveys: 3-5 questions maximum. Relationship surveys: 5-8 questions. Completion rates drop sharply as length increases. If you need more data across a customer relationship, run multiple short surveys at different touchpoints rather than one comprehensive survey.
Always pair a score with an open-ended follow-up. "How satisfied are you?" gives you a number. "What is the main reason for your score?" gives you something to act on. Without the qualitative context, CSAT data identifies that something is wrong without explaining what.
Anonymize where possible. Named surveys tied to a customer account inflate scores because of social desirability bias. For in-app and email surveys, use neutral framing: "Your response is used to improve the experience for all customers."
Set internal benchmarks first. Your most meaningful CSAT benchmark is your own score last quarter. External industry benchmarks provide context, but your internal trend is what drives roadmap and operational decisions.
Close the loop visibly. Tell customers what changed because of their feedback. "You told us onboarding was unclear, so we rebuilt the first-run experience." This is the single most effective practice for increasing future response rates and building trust that feedback matters.
Common CSAT Survey Mistakes
Surveying only happy customers. If CSAT is triggered only on success flows (order confirmation, resolved ticket), you miss the customers who abandoned or churned. This structurally inflates your score.
Leading questions. "How satisfied were you with our excellent support team?" primes the respondent to rate positively before they have even thought about their answer. Keep question wording neutral.
No follow-up action. Collecting feedback without acting on it is worse than not collecting it. Customers who complete surveys and see no change disengage from future surveys. More critically, they conclude the company does not listen.
Measuring CSAT in isolation. CSAT without NPS tells you whether individual moments are good but not whether the relationship is healthy. Add NPS to quarterly relationship surveys and use CES at effort-heavy touchpoints. For a comprehensive look at customer experience survey questions across metrics, see our dedicated guide.
Ignoring the middle of the scale. A score of 3 (neutral) is often treated as acceptable. It is not. Neutral customers are indifferent, and indifferent customers switch to competitors without warning. A 3 deserves the same investigation as a 1 or 2.
Changing scales without resetting baselines. If you move from a 1-10 scale to a 1-5 scale, your historical scores are not comparable. Treat the change as a measurement restart and communicate it clearly to stakeholders.
How to Analyze CSAT Survey Results
Step 1: Calculate your score. (Responses of 4 and 5 / total responses) x 100. Do this for each touchpoint separately, not just as a company average.
Step 2: Segment by dimension. Minimum segments: touchpoint, customer type (new vs. tenured), channel, and plan or product tier. A company-wide CSAT of 76% hides the fact that enterprise customers may be at 85% and SMB customers at 64%.
Step 3: Cross-reference with open-ended responses. Low scores without qualitative context are hard to act on. Group open-ended responses by theme (wait time, product quality, communication, pricing) and frequency.
Step 4: Prioritize by frequency and churn correlation. An issue mentioned by 30% of low-score respondents is more urgent than one mentioned by 5%, even if the 5% group is more vocal. If you have churn data, correlate CSAT scores with 90-day churn to identify which score thresholds actually predict departure.
Step 5: Assign ownership and set targets. Each CSAT touchpoint should have a clear owner (support lead, product team, onboarding team). Set quarterly targets that are specific and incremental: +3 percentage points per quarter is more realistic and trackable than "improve CSAT."
Step 6: Close the loop with customers. Route 1-2 scores to a follow-up workflow within 24 hours. Acknowledge the feedback. Communicate changes. This recovers some detractors and demonstrates the program is real. For a systematic approach, see our guide on closing the feedback loop.
Free CSAT Survey Template
Skip the blank page. Formbricks offers free, open-source CSAT survey templates you can deploy in minutes.
Related templates:
How to get started:
- Sign up at formbricks.com (free tier available, no credit card required)
- Choose the CSAT survey template or start from scratch
- Set targeting rules to trigger after interactions
- Launch and monitor responses in real time
Formbricks is open source and supports self-hosting for teams that need full data control. It works across in-app, website, email, and link survey channels, so you can run the same CSAT question at every touchpoint with consistent methodology.
Get Your Free CSAT Survey Template →
Frequently Asked Questions
Try Formbricks now
