A customer satisfaction survey template is a pre-built set of questions you can copy, customise, and send to customers to measure how they feel about your product, service, or specific touchpoint. A good template includes one core CSAT question (a 1 to 5 satisfaction scale), a Net Promoter Score (NPS) question for word-of-mouth signal, three to five experience-specific questions tied to your funnel, one or two open-ended diagnostics to surface what numbers cannot, and a couple of segmentation fields so you can slice the results. The whole survey should fit in three to four minutes of completion time. Anything longer and you trade response quality for length.
The template below is built around 25 question prompts grouped by intent. You pick the 8 to 12 that fit your touchpoint and ship. Open the PollPe Survey Builder if you want a free editor to drop these into and start sending.
Key takeaways
- Every customer satisfaction survey template should anchor on one CSAT question, one NPS question, and three to five experience-specific questions. The rest is context.
- Length kills response rate. Aim for 8 to 12 questions and under 4 minutes. SurveyMonkey benchmarks suggest completion drops sharply past 7 to 8 minutes.
- Industry CSAT benchmarks: SaaS around 78 percent, retail around 77 percent, financial services around 75 percent (American Customer Satisfaction Index, 2024). Compare yourself to your industry, not to a flat 90 percent target.
- Open-ended questions are where insight lives. Code them by theme weekly, not just at year-end.
- Timing matters as much as content. Send within 24 hours of the experience for the cleanest memory recall.
- PollPe Survey Builder offers unlimited responses on the free plan and 15 languages on Business and above, so you can run the same CSAT template across English, Hindi, Tamil, and Telugu audiences without three separate surveys.
Why most customer satisfaction survey templates collect noise instead of signal
Customer satisfaction surveys are easy to build and hard to do well. The common failure pattern looks like this: a product manager wants to measure something, opens the survey tool, writes 25 questions because each one feels essential, sends it to the entire customer base, and gets 50 responses back, mostly from people who already love or hate the brand. The chart looks tidy. The decision-makers nod. Nothing changes.
Three forces produce that noise.
First, length. Past 12 questions on a consumer survey, you start losing the casual respondent, the very person whose feedback would have been most useful. The vocal evangelists and the bitter churners will push through anything. The middle 80 percent walks. SurveyMonkey's own data suggests completion rates drop noticeably after the 7 to 8 minute mark. A 25-question survey at 20 seconds per question is already past that line.
Second, leading questions. "How satisfied are you with our amazing onboarding experience" is not measurement, it is theatre. The wording primes a positive answer, then you congratulate yourself on a 4.6 CSAT and miss the part where onboarding silently churns 18 percent of trial users.
Third, no segmentation. A CSAT of 78 percent across your entire base sounds fine until you slice it by tier, by region, by lifecycle stage, or by ticket-volume cohort and discover that the 18-month-tenure mid-market segment sits at 61 percent. That segment is your churn risk. A template that does not capture segmentation fields cannot tell you that story.
A template fixes the first problem by keeping things tight. It fixes the second by giving you tested wording. It fixes the third by including the right two or three demographic and lifecycle fields by default.
The anatomy of a customer satisfaction survey template that works
Most strong customer satisfaction surveys share the same shape, regardless of industry.
One anchor metric. A single core CSAT or NPS question that you measure every send. This is what lets you track trends. Pick one, define it precisely, and never change the wording without flagging the methodology change. If you mix CSAT and NPS, pick one as the headline number.
A small set of experience-specific questions. Three to five questions tied to the specific touchpoint you are measuring. For a post-purchase survey, that is delivery, product condition, and ease of checkout. For a post-support survey, it is resolution time, agent helpfulness, and effort. Generic questions on a touchpoint-specific survey water down the signal.
One or two open-ended diagnostics. Free-text questions like "what was the single thing that could have improved your experience" routinely surface issues that no rating scale captures. Yes, they take longer to analyse. Yes, they are worth it.
Segmentation fields. Two or three at most. Customer tier, customer tenure, and the channel they came through are usually enough. If you already capture this in your CRM and can pipe it in as a hidden field, even better. Do not ask customers for information you already have.
A clear opt-out and time estimate. "This takes about 3 minutes. You can skip any question." Sets the expectation, builds trust, lifts completion.
A thank-you screen that does something. Confirm the response is recorded, tell them what you do with the data, and where possible, hint at the change. "Last quarter, your feedback led us to rewrite the welcome email and cut onboarding time by 40 percent" earns the next response.
Once you have that shape locked, the question content slots in naturally.
25 customer satisfaction survey questions
The 25 questions below are grouped by intent. Pick 8 to 12 that match your touchpoint and your goal. Resist the urge to use all 25.
Overall satisfaction
These are your headline metrics. Most templates include the first one (CSAT) and one of the other two.
-
"How satisfied are you with [product or service] overall?" (Scale: 1 = very dissatisfied, 5 = very satisfied.) This is the canonical CSAT question. CSAT score is calculated as the percentage of respondents who picked 4 or 5. Simple, comparable across periods, well understood by stakeholders.
-
"How likely are you to recommend [company] to a friend or colleague?" (Scale: 0 to 10.) The NPS question. NPS is calculated as percentage promoters (9 to 10) minus percentage detractors (0 to 6). NPS tells you about advocacy, not satisfaction, and the two diverge surprisingly often.
-
"Overall, how would you rate your experience with [touchpoint]?" (Scale: 1 = poor, 5 = excellent.) Variant of CSAT scoped to a touchpoint rather than the whole brand. Useful when you are running a touchpoint-specific survey (post-onboarding, post-support) and want a clean read on that moment.
Product or service experience
These get specific about what the customer interacted with.
-
"How well did [product or service] meet your expectations?" (Scale: 1 = much worse than expected, 5 = much better than expected.) Expectations are the hidden anchor in every satisfaction score. This question surfaces gaps between what you promised and what you delivered.
-
"How easy was it to [complete a specific action, like checkout, onboarding, finding a feature]?" (Scale: 1 = very difficult, 5 = very easy.) A Customer Effort Score (CES) variant. CES correlates strongly with retention. For a dedicated walkthrough, see our customer effort score guide.
-
"Which features do you use most often?" (Multi-select, list your top 8 to 10 features.) Tells you what is actually pulling weight versus what is sitting unused. Cross-tab against satisfaction to find the features that drive retention.
-
"Which features do you wish [product] had?" (Open text.) Less for the literal answer (customers describe solutions, not problems) and more for the patterns across answers. Code these monthly.
-
"How often do you use [product]?" (Single select: daily, weekly, monthly, rarely, only when needed.) Frequency is a leading indicator of churn. Heavy users are stickier; weekly users are the watch list.
Support and service quality
If your touchpoint involves support, use 2 to 4 of these.
-
"How satisfied were you with the support you received?" (Scale: 1 to 5.) The support-specific CSAT. Run this on every closed ticket if you can, not just an annual relationship survey.
-
"How quickly did we resolve your issue?" (Scale: 1 = much slower than I needed, 5 = faster than I expected.) Time-to-resolution matters more than first-response time for customer perception in most B2B contexts.
-
"Did we resolve your issue on the first contact?" (Single select: yes, no, partial.) First Contact Resolution (FCR) is a leading indicator of CSAT. Track it weekly.
-
"How would you describe the support agent who helped you?" (Multi-select, with options like knowledgeable, friendly, efficient, professional, frustrating, unclear.) Better than a rating because it gives you actionable language. "Knowledgeable" and "efficient" together suggest a good outcome; "friendly" without "knowledgeable" suggests warm but unhelpful.
Value for money
Three lenses on whether the customer feels you are worth the price.
-
"How would you rate the value for money of [product]?" (Scale: 1 = very poor value, 5 = excellent value.) Value perception and price are different. A customer can rate a $500 product as excellent value and a $50 product as poor value, depending on what they got.
-
"How does [product] compare to similar products you have used?" (Scale: 1 = much worse, 5 = much better.) Direct competitive read. Watch for the open-text variant if you want to capture which alternatives are top of mind.
-
"At your current usage level, do you feel the price you pay is fair, low, or high?" (Single select.) Honest pricing temperature check. Often the answer is "fair" even from churners, which tells you the churn driver is not price.
Likelihood to repurchase or recommend
Forward-looking signals.
-
"How likely are you to continue using [product] in the next 6 months?" (Scale: 1 = very unlikely, 5 = very likely.) Retention prediction. Combine with the NPS question for a two-dimensional read on advocacy versus continued use.
-
"How likely are you to buy from us again?" (Scale: 1 to 5.) The transactional variant for e-commerce or one-off purchase models. For SaaS, use the previous question instead.
-
"What would make you more likely to recommend us?" (Open text, asked only of respondents who scored 7 or 8 on NPS, the "passives".) Detractors will tell you why they are unhappy. Promoters do not need prompting. Passives are the segment you can convert with the right intervention. This question identifies what that intervention is.
Open-ended diagnostic
These are where the qualitative gold sits. Use at most 2 in a single survey.
-
"What is the one thing we could do to improve your experience?" (Open text.) The most useful single question on any survey. Phrasing matters: "the one thing" forces prioritisation. "What could we improve" gets you a list of mild suggestions.
-
"What do you like most about [product or service]?" (Open text.) Captures the language customers use to describe value. Recycle this language in your marketing, your onboarding, and your sales conversations. It will outperform copy your team wrote.
-
"If [product] disappeared tomorrow, what would you use instead?" (Open text.) Competitive intelligence and category positioning in one question. Watch for the answer "nothing" or "go without" because that tells you your category is well-defined.
-
"Tell us about a time [product] really helped you, or really frustrated you." (Open text.) Story-format question. Produces stronger signal than abstract ratings because customers recall specific moments better than aggregated impressions.
Demographics and segmentation
Three fields at most. If you already capture this in your CRM, pre-populate as hidden fields rather than asking again.
-
"How long have you been a customer?" (Single select: less than 1 month, 1 to 6 months, 6 to 12 months, 1 to 2 years, more than 2 years.) Tenure cohorts behave very differently. Segmenting CSAT by tenure regularly surfaces the "valley of disillusionment" around month 4 to 6.
-
"Which plan are you on?" (Single select with your actual plan names.) Plan-level CSAT shows where you are over-serving or under-serving relative to price. Free-tier dissatisfaction is fine. Enterprise dissatisfaction is fire.
-
"How would you describe your role?" (Single select: founder, manager, individual contributor, other.) Role context shapes how someone evaluates the product. A founder rating your reporting feature is judging it differently than an analyst is.
When to send your customer satisfaction survey template
Timing is half the survey. The same question gets a very different answer at the wrong moment.
Post-purchase, 24 to 48 hours. For e-commerce and one-off purchase models. Memory is fresh. Use a 3 to 5 question template focused on the purchase experience and product delivery. Do not ask about loyalty or long-term satisfaction at this point. The customer has not lived with the product yet.
Post-onboarding, 7 to 14 days after signup. For SaaS. The customer has had time to set up, attempt the core job, and form an early opinion. Anchor on a Customer Effort Score and an open text "what got in your way" question.
Post-support, immediately after ticket closure. Auto-trigger when the ticket status changes to resolved. Keep it to 3 questions: CSAT, FCR (yes/no/partial), and open text for anything else. Skip everything else. People who just got off a support interaction have low patience for a 12-question survey.
Quarterly relationship survey, every 90 days. For SaaS and ongoing service relationships. This is your tracker. 8 to 12 questions, anchored on NPS plus a CSAT, with a rotating block of 4 to 6 questions that drill into a different theme each quarter (product, support, value, account management).
Annual deep-dive, once per year. Longer, 15 to 20 questions, includes the segmentation and benchmarking questions you do not need every quarter. This is the one you take to the board.
Event-triggered, after key milestones. First successful use, 10th invoice, plan upgrade. These are moments of high engagement and produce great open-text responses.
For more on send timing and benchmarks, see our survey response rate benchmarks post.
How to distribute it without killing response rates
A customer satisfaction survey template only works if it reaches your customers. Distribution choices matter more than people give them credit for.
Email still wins for relationship surveys. Personalise the subject line with the customer name. Keep the email body under 100 words. Single CTA button to start the survey. Send Tuesday or Wednesday morning in the customer's timezone. Avoid Mondays (inbox overload) and Fridays (already mentally checked out).
In-app is the highest-converting channel for SaaS. A slide-in panel that appears after a successful action ("you just exported your first report, got 30 seconds for a quick question") outperforms email by a factor of 3 to 5 on conversion. Do not over-trigger. Once per quarter per user is the limit.
SMS works in markets where it is still a primary channel. India, Indonesia, parts of Latin America. Keep the message under 160 characters, send during daytime, and link to a mobile-optimised survey. The response window is short, so do not expect responses past 48 hours.
QR codes belong on physical touchpoints: receipts, packaging, in-store signage, restaurant tables. The trick is making it obvious what is in it for the scanner. "Tell us how we did, takes 60 seconds" beats a bare QR with no context.
Embed on the thank-you page after a transaction. The customer is already on your site, the action just completed, and there is zero friction to start.
For a fuller distribution playbook, see how to distribute surveys.
How to actually read the results
A survey is worth what you do with the data. Three habits separate teams that act on feedback from teams that file it.
Calculate CSAT and NPS the standard way. CSAT equals the count of 4 and 5 ratings, divided by the total count, multiplied by 100. NPS equals the percentage of 9 to 10 ratings (promoters) minus the percentage of 0 to 6 ratings (detractors). Do not invent your own formula. The reason CSAT and NPS are useful is benchmarkability, and you lose that the moment you change the math.
Segment everything. Headline numbers hide the story. Run CSAT by tier, by tenure, by region, by acquisition channel. The variance across segments is usually 15 to 25 points, and that variance is where decisions live. A flat 78 percent CSAT is interesting. A 92 percent CSAT in your annual plan customers and 64 percent in your monthly plan customers is actionable.
Code the open-text every time. Pick a coding framework (theme, sub-theme, sentiment) and apply it consistently. Aim for 8 to 12 themes total. Track theme frequency over time. If "onboarding is confusing" jumps from 8 percent to 22 percent of responses across two quarters, that is a leading indicator your latest release broke something.
PollPe's Aria AI assistant handles this analysis on the Business plan and above. The Deep Analysis mode reads through every open-text response in a survey, clusters them into themes, surfaces sentiment shifts, and writes a structured report you can drop into a Notion or Slack thread. For teams running monthly tracking studies, it cuts the analysis loop from a day to less than an hour. Aria's Standard mode is available on every plan, including the free tier, and handles question drafting and quick summaries.
For deeper grounding on the CSAT metric itself, see our CSAT survey guide. For NPS-specific guidance, see NPS survey best practices.
Customer satisfaction survey templates by industry
Generic templates are a starting point. Industry-specific micro-templates ship faster because they skip the questions that do not apply.
SaaS (3 to 4 questions)
- "How satisfied are you with [product] overall?" (1 to 5)
- "How likely are you to recommend [product] to a colleague?" (0 to 10)
- "What is the one thing we could improve in the next 90 days?" (open text)
- "Which features do you use most often?" (multi-select)
Retail or e-commerce (4 questions)
- "How satisfied are you with your recent purchase?" (1 to 5)
- "How would you rate the delivery experience?" (1 to 5)
- "How likely are you to buy from us again?" (1 to 5)
- "Anything we could have done better?" (open text)
Restaurant or hospitality (4 questions)
- "How would you rate your overall experience today?" (1 to 5)
- "How was the food/service quality?" (1 to 5)
- "How likely are you to recommend us to a friend?" (0 to 10)
- "What would you tell a friend about your visit?" (open text)
Professional services (4 questions)
- "How satisfied are you with the engagement so far?" (1 to 5)
- "How responsive has the team been to your needs?" (1 to 5)
- "How clearly have we communicated next steps and expectations?" (1 to 5)
- "What would make this engagement more valuable to you?" (open text)
B2B SaaS (5 questions, run quarterly)
- "How likely are you to recommend [product] to a peer at another company?" (0 to 10)
- "How well does [product] meet the needs of your team?" (1 to 5)
- "How would you rate the value you get for what you pay?" (1 to 5)
- "What would make you more likely to expand your use of [product]?" (open text)
- "How long have you been a customer?" (segmentation)
Each of these can be built in PollPe in under 5 minutes with the Aria AI assistant. Describe the use case in one sentence and Aria drafts the question set; you edit and ship. The free plan supports unlimited responses, so you can run any of these templates at any scale without hitting a cap.
FAQs
How many questions should a customer satisfaction survey have?
For most use cases, 8 to 12 questions. Anything past 12 starts losing response rate sharply. For post-transaction or post-support surveys, drop to 3 to 5 questions. For an annual relationship survey, you can go to 15 to 20 if respondents have a real stake in your product. See our how long should a survey be post for the full breakdown.
What is a good CSAT score?
Industry context matters. The American Customer Satisfaction Index puts software and SaaS around 78 percent, retail around 77 percent, and financial services around 75 percent. Anything in the 80 to 90 percent range is strong for most industries. Anything below 70 percent typically signals systemic issues. Compare to your industry average, not to a flat target.
Is CSAT or NPS better?
They measure different things. CSAT measures satisfaction with a specific experience and is best for touchpoint-level feedback. NPS measures advocacy and loyalty and is best for relationship-level tracking. Most mature CX programs use both, and you should too. Pick one as your headline metric for company-wide tracking, and use the other for deeper diagnostics.
How often should I send customer satisfaction surveys?
For touchpoint surveys (post-purchase, post-support, post-onboarding), trigger them in real time after the event. For relationship surveys (quarterly NPS, annual deep-dive), space them out. The fastest way to kill response rates is to over-survey. If a customer received a survey from you in the last 30 days, do not send another one unless it is a different touchpoint.
Can I send a customer satisfaction survey in multiple languages?
Yes. PollPe Survey Builder supports 15 languages on the Business plan and Enterprise, including Hindi, Tamil, Telugu, Bengali, Marathi, French, Spanish, German, and Arabic. Let respondents pick their language on the first screen and store the choice as a variable so you can segment responses later. Critically, do not just machine-translate the questions. Have a native speaker review for cultural fit. A literal translation of "delighted" can land awkwardly in Hindi or German.
How do I get more people to respond to my customer satisfaction survey?
Three levers: timing (send within 24 to 48 hours of the experience), length (under 4 minutes), and the invitation copy (specific subject line, clear "what is in it for you", and a real estimate of time required). For benchmarks and tactics, see survey response rate benchmarks and our companion piece on how to write survey questions.
How do I share the results with my team?
Build a simple dashboard with three things: the headline metric (CSAT or NPS) over time, the same metric segmented by your top two cuts (tier, tenure, region), and the top 5 themes from your open-text responses for the current period. Share weekly to your CX team, monthly to your leadership team, and quarterly in a written summary to the board. Tools like PollPe's custom dashboards (Business plan and above) automate most of this.
Wrapping up
A customer satisfaction survey template is not a set of 25 questions you blast at every customer. It is a curated short list you adapt to each touchpoint, send at the right moment, and read with discipline. The teams that get value from CSAT and NPS are not the ones with the longest surveys or the prettiest dashboards. They are the teams that pick 8 to 12 questions, send them consistently, segment the results, code the open text, and close the loop with customers when something changes.
Pick a touchpoint to start with. Use 8 to 12 questions from the list above. Send to a defined cohort. Read the results within a week. Decide one action. Ship the action. Re-survey in 90 days.
If you want a free editor to drop these questions into, create a PollPe Survey Builder account. Unlimited responses on the free plan, AI question drafting included, multilingual support on Business and above. Compare plans at pollpe.com/pricing.
