Customer satisfaction surveys: examples for every stage of the customer journey

Customer satisfaction surveys have been helping businesses listen to customers for years. However, while a score helps, it rarely tells the whole story. A customer can give you a 9, a 6, or a 2 and still leave many questions unanswered: why they responded that way and what you should do with that information. That is why the key is to understand what you want to measure, when it makes sense to ask, and what you will do with the response afterward.

In this guide, you will find examples of customer satisfaction surveys for different moments, when to use NPS, CSAT, or CES and how AI agents for customer satisfaction surveys can turn a static questionnaire into a conversation that can dig deeper into responses and trigger the next action.

What is a customer satisfaction survey and what should it measure?

A customer satisfaction survey is a set of questions designed to understand how a person rates an interaction, a product, a service, or their relationship with a company.

But a score on its own provides limited information. That is why a good survey combines quantitative metrics with questions that help uncover the reason behind the score.

For a survey to truly support decision-making, it should capture four elements:

• Rating: how the customer rates what happened.
• Reason: what explains that rating.
• Context: which interaction, purchase, delivery, or process the rating refers to.
• Actionability: what should happen if a problem, an opportunity, or a sign of churn appears.

NPS, CSAT, and CES: which metric should you use for each survey?

NPS, CSAT, and CES answer different questions:

• NPS (Net Promoter Score): measures willingness to recommend and provides a broader view of the relationship. It is typically used for periodic surveys, relationship tracking, or around renewals.
• CSAT (Customer Satisfaction Score): measures satisfaction with a specific interaction or moment. It works well after an inquiry, a purchase, a delivery, or a customer service interaction.
• CES (Customer Effort Score): measures how easy or difficult it was to complete a task. It is especially useful for resolution processes, onboarding, support, or multi-step processes.

Examples of customer satisfaction surveys by stage in the customer journey

The right question depends heavily on what has just happened in each interaction. Below are some examples of customer satisfaction surveys:

Example 1: Customer satisfaction survey after an interaction or call

• Objective: find out whether the inquiry was resolved and how the customer rates the service.
• When to send it: immediately after the sale closes.
• Recommended metric: CSAT and, when it provides useful information, CES.
• Questions:
  ‍
  
  • Was your inquiry resolved?
  • On a scale from 1 to 5, how satisfied are you with the service you received?
  • How easy was it to resolve your inquiry?
• What to do with the response: if the issue is still open or the rating is low, you can generate an alert or resume the case. AI agents for customer service can capture the reason and route the case to the appropriate team with context.

Example 2. Customer satisfaction survey after a purchase or delivery

Purchase and delivery are related, but they do not measure exactly the same thing:

1. Right after the purchase, it is useful to find out whether the process was easy, the information was clear, and payment was simple.

2. After delivery, on the other hand, it is useful to look at whether the deadline was met, the condition in which the product arrived, and whether it met expectations.

With this in mind, the satisfaction survey can be customized for each case:

• Survey objective: identify potential friction during the purchase or check whether the delivery met expectations.
• When to send it: immediately after the purchase is completed or once the delivery has been confirmed, depending on the moment you want to evaluate.
• Recommended metric: CSAT and, if you want to measure how easy the purchase process was, CES.
• Questions after a purchase:
  ‍
  
  • Was it easy to complete the process?
  • Did you find all the information you needed?
  • Was there any point that caused confusion?
• Questions after a delivery:
  ‍
  
  • Did your order arrive within the expected timeframe?
  • Did it arrive in good condition?
  • Did the delivery meet your expectations?
• What to do after a negative response: identify where the problem occurred and trigger the appropriate follow-up. In addition, some responses reveal new needs that an AI sales agent can record or pass on to the sales team.

Example 3. Customer satisfaction survey for onboarding or service use

• Survey objective: determine whether getting started with the product or service is easy and identify potential blockers.
• When to send it: after the first few days of use or once the first relevant steps have been completed.
• Recommended metric: CES to measure how easy the process is, combined with open-ended questions.
• Questions:
  ‍
  
  • Was it easy to get started with the service?
  • Did you receive the information and support you needed?
  • Which part was the most difficult?
• What to do after a negative response: identify the point of friction and offer help before the difficulty ends up affecting usage or continued use.

Example 4. Customer satisfaction survey tied to renewal, cancellation, or churn risk

• Survey objective: understand the value the customer perceives, identify unmet needs, and find out what could put retention at risk.
• When to send it: before a renewal or immediately after a cancellation.
• Recommended metric: NPS can help measure overall perception before renewal, combined with open-ended questions that help uncover the reason.
• Questions:
• Before a renewal it is useful to ask:
  ‍
  
  • What value have you received from the service?
  • Is there any need we are still not meeting?
  • What could affect your decision to continue?
• After a cancellation, the focus changes:
  ‍
  
  • What was the main reason?
  • Was there any problem we were unable to resolve?
  • What could have changed your decision?
• What to do with the response: identify recurring reasons and determine where there may have been an opportunity to intervene earlier.

Example 5. Periodic NPS survey

• Survey objective: understand the customer’s willingness to recommend the company and which factors are influencing that perception.
• When to send it: periodically or at relevant moments in the relationship, without necessarily tying it to a single interaction.
• Recommended metric: NPS.
• Questions:
  ‍
  
  • On a scale from 0 to 10, how likely are you to recommend us?
  • What is the main reason for your score?
  • What should we improve to earn a higher rating?
• What to do after a negative response: dig deeper into the reason and, if there is a specific issue, trigger a review or route the case to the appropriate team. AI agents for customer satisfaction surveys can handle these situations by routing the conversation to the contact center.

Which channel should you use to send a customer satisfaction survey?

Once you have defined what you want to ask, there is another decision to make: how to reach the customer. The channel should be selected based on the moment, the length of the survey, and the type of response you need to obtain:

• Email: works well for periodic surveys or somewhat longer questionnaires.
• Webchat: useful for collecting feedback during or after a digital process.
• WhatsApp: suitable for short, conversational surveys sent after an interaction.
• SMS: works best for quick questions or when you want to send a link to a questionnaire.
• Voice: can be handled through an IVR (useful for collecting simple scores after a call) or an AI voice agent, which can conduct phone surveys with open-ended responses and questions that adapt to the conversation.

Either way, you do not have to choose a single channel. A company can combine different types of AI agents to increase reach: AI agents for WhatsApp, digital messaging, or AI voice agents. The choice should answer one basic question: which channel makes it easiest to get a useful response at that moment?

How to use AI agents to conduct customer satisfaction surveys

Sending a form, using AI to prepare questions, and conducting a survey through an AI agent are different things.

A form asks the same questions to everyone along a fixed path. AI can help you draft those questions, but that does not change the logic of the questionnaire. With an AI agent, however, the survey adapts to each response: it can dig deeper, ask for clarification, or trigger the next step when it detects something relevant.

During the survey, the virtual agent can:

• Trigger the survey after a call, purchase, or delivery.
• Ask questions by voice or text.
• Adapt the conversation to each person.
• Ask for more details when a response requires it.
• Identify topics, reasons, and sentiment.
• Save the results in the CRM.
• Summarize and structure open-ended responses
• Create alerts or tasks
• Trigger the next action.
• Route a negative rating to the appropriate team with context.

A good customer satisfaction survey helps you decide what to do next

The customer satisfaction survey examples we have reviewed have one thing in common: every question should help you understand something or make a decision.

When a large number of responses come in, Inagent AI agents can classify comments, detect patterns, and record the information in the company’s systems. The company also maintains control over the questions, the rules, and when a person should step in.

With Inagent, that feedback can also be connected to the company’s systems and processes so that a negative rating does not simply end up stored in a report, because an AI agent specialized in customer satisfaction surveys can act based on the information collected. If you want to see how to apply this to your company’s processes, you can request a demo here.

Frequently asked questions about customer satisfaction surveys

How many questions should a customer satisfaction survey include?

It depends on the objective and the moment. A post-interaction satisfaction survey should be brief and focus on a few key questions. Broader surveys can include more questions, as long as each one provides useful information for a later decision.

If you use an AI agent for customer satisfaction surveys, not everyone has to receive exactly the same questions: the agent can dig deeper when a response needs more context.

What is the difference between NPS, CSAT, and CES?

NPS (Net Promoter Score) measures willingness to recommend and provides a broader view of the relationship. CSAT (Customer Satisfaction Score) measures satisfaction with a specific interaction or moment. CES (Customer Effort Score) measures how much effort a person had to make to complete a task.

When should a customer satisfaction survey be sent?

Ideally, the customer satisfaction survey should be conducted close to the moment you want to evaluate so the customer remembers what happened. After a call or service interaction, it can be sent immediately. In other cases, such as NPS or renewal, it makes sense to use a regular cadence or a specific moment.

What open-ended questions can be included?

Some examples are: “What is the main reason for your rating?”, “What could we have done better?”, “Which part of the process was the most difficult for you?”, or “What would need to change for your score to be higher?”. Open-ended questions help uncover the reason behind the score.

How can you increase the response rate for a customer satisfaction survey?

Keeping the survey brief, choosing the right channel, asking close to the moment you want to evaluate, and clearly explaining what is expected of the customer can help.

It can also be useful to select the channel based on the context. For certain processes, AI agents for WhatsApp may work better, while others may require voice or digital messaging.

How are customer satisfaction survey responses analyzed?

In addition to calculating metrics such as NPS or CSAT, it is useful to analyze reasons, recurring topics, open-ended comments, and changes in scores over time. When there is a high volume of responses, AI agents can help structure this information, identify patterns, and make it easier for the team to act on them.

How does a survey conducted by an AI agent work?

The agent follows the questions and criteria defined by the company, but it can adapt the conversation based on each response. It can dig deeper into a negative score, clarify a comment, identify the reason, record the information in the systems, and trigger an alert or transfer when the case requires follow-up.

Can an AI agent conduct customer satisfaction surveys by phone?

Yes, AI voice agents can conduct phone surveys, ask questions, collect scores and spoken comments, and adapt the next questions based on the responses.

Can a customer satisfaction survey be sent through WhatsApp?

Yes, AI agents for WhatsApp can conduct conversational surveys on this channel, adapt the questions, and record the responses in the company’s systems. For other text channels, AI agents for digital messaging can apply the same logic across different touchpoints.

What happens if a customer gives a negative rating?

The agent can dig deeper into the reason, classify the cause, collect the necessary information, and trigger an alert or case. When human intervention is needed, it can transfer the case to the appropriate team with full context. Depending on the response, follow-up can continue through AI agents for customer service or, if a sales opportunity appears, through AI sales agents.