CSAT Survey Templates: My 7 Most Used Examples and Questions

A useful CSAT survey template should tell you exactly what to ask, who to ask, and when to ask it. For SaaS teams, that rarely means sending a long customer satisfaction questionnaire to the entire user base. It means asking one focused satisfaction question after a specific moment, then using a short follow-up to understand the score.

As a UX researcher at Userpilot, I use CSAT as a signal, not a diagnosis. A score can show that satisfaction changed after onboarding, a feature release, or a support interaction. It cannot explain the cause on its own. That is why the seven templates below pair each question with a behavioral trigger, a relevant audience, and the product data I would examine next.

Userpilot lets you launch these as in-app surveys, target users by behavior or account data, and change the follow-up based on the score they select.

How CSAT is calculated: Divide the number of satisfied responses by the total number of responses, then multiply by 100. On a five-point scale, “satisfied” normally means the top two responses. Keep the scale and calculation consistent if you want to compare results over time.

7 CSAT survey templates and questions for SaaS

Most generic templates combine product quality, value, support, loyalty, and repurchase intent in one questionnaire. I would not do that inside a SaaS product. Each of those answers belongs to a different moment and may require a different response.

Use this table to choose the template that matches the experience you want to measure:

CSAT template	When to trigger it	Primary question	What to do next
Onboarding completion	After the user completes the activation milestone or finishes the setup checklist.	How satisfied are you with your onboarding experience?	Compare scores with activation and replay low-scoring sessions.
Core workflow	After the user completes an important job for the second or third time.	How satisfied are you with the process of [completing task]?	Look for friction in the steps preceding low scores.
Feature satisfaction	After the user has engaged with the feature enough to form an opinion.	How satisfied are you with [feature]?	Compare satisfaction with adoption frequency and account segment.
New release	After exposed users have tried the updated experience several times.	How satisfied are you with the updated [feature]?	Compare new and returning users, then review what became easier or harder.
Post-support	Immediately after a ticket or chat is marked resolved.	How satisfied are you with the help you received?	Separate issue resolution, agent behavior, and product causes.
Self-service help	After a user reads a help article, watches a tutorial, or exits a resource.	How satisfied are you with the help provided for [task]?	Use low scores to improve the resource or route users to human help.
Implementation and rollout	After an admin completes setup, integration, migration, or team rollout.	How satisfied are you with the implementation process?	Compare admin feedback with setup completion and team adoption.

1. Onboarding completion CSAT survey template

An onboarding CSAT survey should measure a completed onboarding experience, not interrupt it. I would never trigger it on first login or halfway through setup. At that point, the user is still trying to understand the product and does not have enough context to give you a useful satisfaction score.

Instead, define the milestone that represents a meaningful setup outcome. That could be importing data, connecting an integration, inviting a teammate, publishing a first flow, or completing the onboarding checklist. Trigger the survey after that event, or the next time the user returns if asking immediately would interrupt the task.

Primary question:

• How satisfied are you with your onboarding experience?

Follow-up questions:

• For low scores: What was the biggest obstacle during setup?
• For neutral scores: What would have made onboarding easier?
• For high scores: What helped you get set up successfully?

I prefer score-based follow-ups because “What is the reason for your score?” makes the respondent do too much interpretation. A user who selected “very dissatisfied” should be prompted to describe the obstacle. A satisfied user can tell you which part of onboarding is worth preserving.

Do not read onboarding CSAT as proof that onboarding works. Pair it with activation. If users report high satisfaction but still fail to complete the core action, you may have created a pleasant setup experience that does not lead to value. If low scorers also abandon before activation, the two signals point to a more urgent problem.

In Userpilot, you can trigger the survey after the activation event, target only new users or account admins, and use branching logic to send each score range to a different follow-up.

2. Core workflow CSAT survey template

This template measures satisfaction with the job users came to your product to complete. For a reporting tool, that may be building and sharing a dashboard. For an email platform, it may be launching a campaign. For a project management product, it may be creating and assigning a complete workflow.

I would normally wait until the second or third completion. The first attempt is heavily influenced by unfamiliarity, while repeated use gives the user enough experience to judge the workflow itself.

Primary question:

• How satisfied are you with the process of [completing task]?

Follow-up questions:

• What almost stopped you from completing this task?
• Which step took more time than expected?
• What would make this workflow work better for you?

Be precise about what you are measuring. If the real research question is how difficult the workflow felt, use a customer effort score instead. CSAT measures the reaction to the experience; CES measures perceived effort. Using the right label matters because a user can complete an easy task and still dislike the outcome, or struggle through a task and still be satisfied because it solved an important problem.

After collecting responses, look at the events immediately before the score. A low rating becomes much more useful when you can see whether the user repeated a step, encountered an error, opened help content, or abandoned and returned later.

3. Feature satisfaction CSAT survey template

Feature satisfaction surveys work best when the audience has actually used the feature. I often see teams target everyone on a page or everyone included in a release announcement, then wonder why the responses are vague. Visibility is not the same as experience.

Set an eligibility rule based on meaningful usage. For a simple feature, one completed action may be enough. For a workflow with several steps, I would wait for multiple uses or a completed outcome.

Primary question:

• How satisfied are you with [feature]?

Follow-up questions:

• What is the main reason for your score?
• How well does [feature] fit into your workflow?
• What is one thing you would change about it?

The score becomes more useful when you compare it with adoption. High satisfaction and low usage often point to a discovery problem: the people who find the feature like it, but too few users reach it. Low satisfaction and high usage can be more serious because users may depend on a workflow they find frustrating.

I also segment feature CSAT by role and account. An administrator, an occasional contributor, and a daily operator can rate the same feature differently for valid reasons. An aggregate score may hide the exact group the feature is failing.

Userpilot lets you target the survey using feature events and user or company attributes, then filter the results by segment and time period.

4. New release CSAT survey template

A release survey should measure the experience after the novelty has worn off. Triggering it the first time someone opens the redesigned page mostly captures surprise. Some users will rate any change negatively because their habits were interrupted; others will rate it positively because it looks new.

I would target users who have completed the updated workflow several times. For a small interface change, that may happen within a day. For a complex reporting or configuration feature, they may need a week or more.

Primary question:

• How satisfied are you with the updated [feature]?

Follow-up questions:

• What became easier after the update?
• What became harder?
• Is anything missing from your previous workflow?

If you have both new and existing users, analyze them separately. Returning users compare the release with a previous version, while new users judge it without that reference point. Combining the two groups can flatten a real migration problem into an average score that looks acceptable.

I also avoid asking a broad question such as “How satisfied are you with our product?” after one release. A feature update can affect the overall relationship, but a general score will not tell you which change caused the movement. Keep the question tied to the experience you can still improve.

5. Post-support interaction CSAT survey template

Send this survey in the same interaction whenever possible, immediately after the ticket or chat is resolved. Every extra step lowers the chance of a response. Asking the user to open an email two days later also turns a specific service memory into a more general opinion of the company.

Primary question:

• How satisfied are you with the help you received?

Follow-up questions:

• Was your issue resolved?
• What could we have done better?
• Is there anything you would like the support team to know?

I strongly recommend separating satisfaction from resolution. A customer may like the agent but remain unhappy because the product cannot do what they need. Another customer may get the issue resolved but dislike the wait or the number of handoffs. One CSAT question cannot tell you which part failed.

This is also why I would not use a single response as an automatic measure of agent performance. Product limitations, pricing policies, staffing, and ticket complexity all influence the score. Look at CSAT alongside first-contact resolution, reply count, reopened tickets, and resolution time before deciding what an agent needs to improve.

Positive comments matter too. When a user names an agent or describes what was helpful, share it with the team. CSAT should not become a system that only notices people when something goes wrong.

In Userpilot, you can trigger the survey from a ticket-resolved event and route low scorers to a personal follow-up.

6. Self-service help CSAT survey template

This template measures whether your help content supported a specific job. Avoid the generic “Was this article helpful?” unless you only need a rough content rating. Users can like an article and still fail to complete the task.

Primary question:

• How satisfied are you with the help provided for [task]?

Follow-up questions:

• Were you able to complete the task?
• What information was missing?
• Would you like help from our team?

I use the outcome question to distinguish a writing problem from a product problem. If users understand the content but still cannot complete the task, rewriting the article will not fix the underlying friction. If they repeatedly search for the same topic, open several resources, and then contact support, the resource may not answer the question clearly enough.

Make this survey passive or trigger it after the user closes the resource. Do not cover the instructions with a feedback modal while they are still trying to follow them. For negative responses, offer a route to support rather than leaving the user at the end of a survey.

7. Implementation and rollout CSAT survey template

For B2B SaaS, the person implementing the product often has a very different experience from the people who use it afterward. This template is for the administrator managing data imports, integrations, security requirements, permissions, configuration, or the internal rollout.

Trigger it after the implementation milestone is complete, not after the contract is signed and not while the admin is still blocked.

Primary question:

• How satisfied are you with the implementation process?

Follow-up questions:

• Which part of implementation caused the most delay?
• How prepared do you feel to roll the product out to your team?
• What support would make the next stage easier?

I would keep implementation CSAT separate from end-user satisfaction. An admin can be happy with the support they received while the wider team struggles to adopt the product. The reverse is also possible: implementation may be painful, but the end product delivers enough value that users are satisfied.

Analyze the response alongside setup events and account adoption. Did the admin connect the required integrations? Were teammates invited? Did the account reach its rollout target? The score tells you how the process felt; the account data tells you whether the rollout worked.

How I make CSAT data worth trusting

The question template is only one part of the setup. Most weak CSAT programs fail because the score is interpreted too confidently or collected without enough context.

Treat CSAT as a signal

CSAT is useful for benchmarking a specific experience and spotting change. It is not a root-cause research method. If onboarding CSAT drops from 82% to 68%, you know where to investigate. You do not yet know whether the cause was a bug, confusing copy, a new user mix, an integration outage, or a change elsewhere in the product.

Add one optional open-text question, then validate the themes with behavior, session replay, support conversations, or interviews. I prefer that workflow to adding five diagnostic questions to every survey. The longer form may collect more fields from the few people who finish it, but it will lose responses from everyone else.

Read the response rate with the score

A 90% CSAT score with a 10% response rate does not mean 90% of customers are satisfied. It means 90% of the people who chose to answer were satisfied.

Check who responded and who stayed silent. Compare response rates by segment, channel, issue type, lifecycle stage, and outcome. The quiet middle often contains users who were not angry enough to complain and not delighted enough to praise you. Their repeat contacts, incomplete workflows, and declining usage may say more than the score they never submitted.

Userpilot’s survey analytics show completion, question-level drop-off, response trends, and results by segment, company, and time period. That helps you distinguish a satisfaction change from a survey-distribution problem.

Compare equivalent experiences

I do not chase a universal “good CSAT” benchmark. The useful comparison is usually your own score for the same touchpoint, scale, audience, and trigger over time.

A post-support score should not be compared with a feature score. Enterprise administrators should not be mixed with trial users. A survey sent after successful task completion will naturally score differently from one triggered after an error. If the conditions change, document the change and start a new baseline.

Keep the survey short, then branch

For most transactional CSAT surveys, one rating question and one optional follow-up are enough. Every additional question should have a clear owner and a decision attached to it. If you cannot explain what you would do with the answer, remove the question.

Use branching instead of asking everyone the same follow-up. A detractor needs space to describe the problem. A satisfied user may be better asked what worked or whether they are willing to join an interview. Userpilot supports Likert, multiple-choice, and open-text questions with conditional logic, so the survey can stay short without treating every respondent identically.

Control frequency without cherry-picking

Survey fatigue is cumulative. A user does not care that last week’s NPS survey and this week’s feature survey belong to different teams. To them, the product is asking again.

I normally start with a program-level exclusion window and adjust it based on the frequency of meaningful product interactions. Userpilot also lets you set sampling, recurrence, dismissal limits, survey ordering, and throttling.

At the same time, do not over-filter the audience only to increase the response rate. Surveying users who completed the workflow will produce more informed answers, but surveying only your most active or successful customers will inflate the score. Define eligibility around experience, then sample fairly within that group.

Close the loop where users can see it

Customers are more likely to answer again when they can see that feedback changes something. A complete feedback loop can be a personal response to a detractor, an in-app announcement describing the fix, a changelog entry, or a follow-up invitation to test the new version.

Do not wait for a major roadmap release. If the team corrected a help article, changed confusing copy, recognized a support agent, or contacted an account that was blocked, tell the people who raised the issue. The response proves that the survey was part of a conversation rather than a score-collection exercise.

Turn CSAT responses into product decisions

The best CSAT template is not the one with the most questions. It is the one tied to a specific SaaS moment, shown to users with enough experience to answer, and connected to a decision your team is prepared to make.

Start with one of the templates above, keep the survey to one score and one relevant follow-up, and analyze the response alongside what the user did before and after answering.

Book a Userpilot demo to build behavior-triggered CSAT surveys, segment the responses, and turn what you learn into a better in-app experience.CSAT survey template gives you a ready-made way to measure customer satisfaction. The questions are often variations of “How satisfied are you with [product, feature, or interaction]?” They are usually triggered at specific moments in the journey to collect more relevant user feedback.

That said, CSAT surveys are quietly becoming a better source of feedback compared to other methods. Users are fatigued by feedback requests at work and in the rest of their lives, and survey response rates have been falling for years as a result. But CSAT surveys can bypass those issues and help you close the feedback loop, as long as they’re well orchestrated.

So for this article, I didn’t want to show some generic templates and call it a day. Instead, I’ll go over how effective CSAT surveys are today, explain how to use the different templates with multiple question examples, and provide some best practices that shouldn’t be ignored.