The best customer satisfaction survey questions ask about one thing at a time and connect to a decision. Start with one headline question, such as overall satisfaction on a fully labeled 5-point scale or the 0–10 NPS question, add a few diagnostic questions about quality, value, delivery or support, and finish with one open-ended 'why' question.
A good customer satisfaction survey asks a small number of clear questions that map to decisions you can actually make. Below are 25 customer satisfaction survey questions you can copy, grouped by what they measure, with notes on when to use each one and which answer scale works best. You won't need all 25 in one survey. Most effective satisfaction surveys use one headline metric, a few diagnostic questions, and one or two open-ended follow-ups.
Before you pick questions: three rules
Decide what you'll do with each answer. If no one will act on a question's result, cut it.
Ask one thing per question. "How satisfied were you with the speed and friendliness of our support?" is double-barreled: someone who got fast but rude service can't answer it. Pew Research Center's guidance on writing survey questions recommends splitting questions like this in two.
Keep scales consistent. If you use a 5-point satisfaction scale, use the same one throughout so results are comparable.
Choosing a scale
Scale
Typical use
Notes
5-point, fully labeled (Very dissatisfied to Very satisfied)
CSAT and most diagnostic questions
Easy on mobile; every point has words
7-point, fully labeled
When you need finer distinctions
Krosnick and Presser's review finds reliability tends to level off around 7 points
0–10 likelihood
Net Promoter Score only
Keep the standard 0–10 so your score is calculated correctly
Two research-backed tips from Krosnick and Presser's chapter on questionnaire design: label every point on the scale with words (reliability is higher than when only the ends are labeled), and include a midpoint. They also recommend avoiding agree/disagree statements where you can, because some respondents tend to agree regardless of content. That's why most of the questions below ask directly ("How satisfied...") rather than presenting a statement to agree with.
Overall satisfaction and loyalty (questions 1–5)
These are your headline metrics. Pick one or two, not all five.
1. Overall, how satisfied are you with [company/product]? Very dissatisfied / Dissatisfied / Neither satisfied nor dissatisfied / Satisfied / Very satisfied Use when: you want a single CSAT score to track over time. This is the core question in the CSAT template.
2. How likely are you to recommend [company] to a friend or colleague? (0 = Not at all likely, 10 = Extremely likely) Use when: you want a Net Promoter Score. This is the standard NPS question as described by the Net Promoter System, which classifies 9–10 as promoters, 7–8 as passives and 0–6 as detractors. See the NPS template.
3. How likely are you to buy from us again? Very unlikely / Unlikely / Neither likely nor unlikely / Likely / Very likely Use when: repeat purchase matters more to you than referrals (for example, consumables or subscriptions).
4. How well did [product/service] meet your expectations? Far below / Below / Met / Above / Far above expectations Use when: you've recently changed a product or made a promise in your marketing and want to know if it landed.
5. Compared with other options you've used, how would you rate us? Much worse / Somewhat worse / About the same / Somewhat better / Much better Use when: you want a quick competitive read from your own customers.
Product or service quality (questions 6–10)
These diagnostic questions explain why the headline score is what it is.
6. How satisfied are you with the quality of [product/service]?Use when: quality is a likely driver of satisfaction or complaints.
7. How would you rate the value for money of [product/service]? Very poor / Poor / Fair / Good / Very good Use when: you're considering a price change or hearing price objections.
8. How well does [product] meet your needs? Not at all well / Slightly well / Moderately well / Very well / Extremely well Use when: you want to spot gaps between what customers need and what you offer.
9. How satisfied are you with the reliability of [product/service]?Use when: uptime, consistency, or durability matter (software, appliances, recurring services).
10. How easy is [product] to use? Very difficult / Difficult / Neither easy nor difficult / Easy / Very easy Use when: you have a digital product or anything with setup or instructions. For deeper usability work, see the product feedback template.
Buying and delivery experience (questions 11–14)
Ask these shortly after purchase or delivery, while the experience is fresh.
11. How easy was it to find what you were looking for?Use when: you run an online store or a large catalog.
12. How satisfied were you with the checkout process?Use when: you've changed checkout or see drop-off there.
13. Did your order arrive by the date you were given? Yes / No / Not sure Use when: delivery promises are a common complaint. A factual yes/no question is easier to act on than a rating here.
14. Did your order arrive complete and in good condition? Yes / No, something was missing / No, something was damaged Use when: you need to separate fulfillment problems from product problems.
Customer support (questions 15–19)
Send these after a support interaction closes.
15. How easy was it to get your issue resolved today? Very difficult / Difficult / Somewhat difficult / Neither easy nor difficult / Somewhat easy / Easy / Very easy Use when: you want a Customer Effort Score (CES). The idea of measuring customer effort was popularized by Dixon, Freeman and Toman in Harvard Business Review, who reported that it predicted loyalty better than satisfaction or NPS in their research. See the customer effort score template.
16. Was your issue fully resolved? Yes / Partly / No Use when: always, for support surveys. Satisfaction scores are hard to interpret without knowing whether the problem was fixed.
17. How many times did you contact us about this issue? Once / Twice / Three or more times Use when: you suspect repeat contacts are driving frustration.
18. How satisfied were you with how quickly we responded?Use when: response time is a target your team manages.
19. How knowledgeable was the person who helped you? Not at all / Slightly / Moderately / Very / Extremely knowledgeable Use when: you're evaluating training needs. Keep questions 18 and 19 separate. Combining speed and knowledge in one question would make the answer impossible to interpret.
Open-ended questions (questions 20–23)
Open-ended questions take more effort, so place one or two after the rating questions. Pew's research shows that open and closed versions of a question can produce very different answers, so open text is valuable for finding issues you didn't think to list.
20. What is the main reason for your score?Use when: right after your headline question (CSAT or NPS). This is the single most useful open question.
21. What is one thing we could do to improve your experience?Use when: you want concrete, prioritized suggestions. "One thing" keeps answers focused.
22. Was there anything that almost stopped you from buying?Use when: you want to find friction in the path to purchase.
23. Is there anything else you'd like to tell us?Use when: as an optional last question. Make it optional so it doesn't cause drop-offs.
Context questions (questions 24–25)
These let you segment results without asking for information you already have.
24. How long have you been a customer? Less than 3 months / 3–12 months / 1–3 years / More than 3 years Use when: you don't already know tenure from your records. If you do, pass it in from your customer data instead of asking.
25. May we contact you about your answers? Yes / No Use when: you want to follow up on low scores. Respect a "No," and if you promised anonymity, don't ask this question at all.
Calculating your scores
CSAT. A common approach is the share of respondents who chose one of the top two options (Satisfied or Very satisfied).
Illustrative example (made-up numbers): 200 customers answer question 1. 90 choose Very satisfied and 60 choose Satisfied. CSAT = (90 + 60) ÷ 200 = 75%.
NPS. Subtract the percentage of detractors (0–6) from the percentage of promoters (9–10). Passives (7–8) count in the total but not in the score.
Illustrative example. A post-purchase survey for an online store might use six questions:
Overall satisfaction (Q1)
Main reason for your score (Q20)
Value for money (Q7)
Order arrived on time (Q13)
Order complete and undamaged (Q14)
One thing we could improve (Q21)
That's short enough to finish in a couple of minutes on a phone and covers the headline score, its drivers, and an open-ended "why."
Tips for better responses
Time it right. Send support surveys right after a ticket closes and product surveys after the customer has had time to use the product.
Keep it short. Pew Research Center caps its own online panel surveys at 15 minutes and notes that the longer a survey runs, the fewer people are willing to complete it. A customer satisfaction survey should usually be far shorter than that.
Use logic. If someone says their issue wasn't resolved (Q16), show a follow-up asking what's still outstanding. Skip it for everyone else.
Close the loop. Tell customers what you changed. It makes the next survey more worthwhile to answer.
Build your survey
You can start from the customer satisfaction (CSAT) template, swap in any of the questions above, and use skip logic to show follow-ups only to the people they apply to. SurveyHood's free plan includes 3 surveys and 100 responses a month, and results update live with cross-tabs so you can compare scores by segment. Learn more on the customer satisfaction survey page.
