To measure customer satisfaction, ask customers for feedback through short surveys, score it with CSAT, NPS and CES, then check the scores against behaviour such as repeat business, churn, complaints and reviews. Using several measures together shows how satisfied customers really are.
At a glance
- CSAT is calculated by dividing positive responses (satisfied and very satisfied) by total responses, then multiplying by 100. For example, 80 satisfied out of 100 responses is 80%.
- The three most common survey metrics are CSAT (satisfaction with an experience), NPS (likelihood to recommend) and CES (how easy it was to get something done).
- The standard five-point satisfaction scale runs from very dissatisfied to very satisfied, with neutral in the middle, according to SurveyMonkey.
- Beyond surveys, satisfaction can be tracked through retention rate, churn rate, customer lifetime value, online reviews and complaints, as SurveyMonkey explains.
- Measurement should start with a business question, and transactional measures (after a specific interaction) serve a different purpose from relationship measures (overall loyalty), says the ACSI.
How do you measure customer satisfaction?
Collect feedback with short surveys, score it using CSAT, NPS and CES and compare it with behaviour and complaints. Surveys tell you what customers say. Behaviour and complaints show what they do, and you need both.
Before you pick a metric, pick a question you want answered. “Are patients happy with their appointment?” needs a different measure from “Would clients recommend us to a friend?” The first is a transactional question, asked straight after a specific interaction. The second is a relationship question, about overall loyalty. Mixing the two up is the most common reason small businesses end up with scores they cannot act on.
- Decide the business question you want to answer.
- Score satisfaction with a specific experience using CSAT.
- Score loyalty using NPS.
- Score ease using CES.
- Add behaviour measures such as retention, churn, reviews and complaints.
- Use chat conversations as an extra source of feedback.
- Turn the results into one or two improvements at a time.
What is CSAT and how do you calculate it?
CSAT measures satisfaction with a specific product, service or interaction. Divide satisfied responses by total responses and multiply by 100.
The survey question
Keep the wording plain, for example: “How satisfied were you with our service?” Offer a five-point scale from very dissatisfied to very satisfied.
The formula
CSAT (%) = (satisfied responses ÷ total responses) × 100
Count a response as positive if it is a 4 (satisfied) or a 5 (very satisfied). Say 100 people answer and 80 of them pick 4 or 5. That gives 80 ÷ 100 × 100 = 80%. The other 20 are neutral or unhappy, and it is worth reading their comments first.
When to use it
CSAT works best straight after an appointment, enquiry or purchase, while the experience is fresh. It tells you how a single moment went. It doesn’t tell you how people feel about your business overall. The ABA’s guide to CSAT is a useful extra read if you want another view of the basics.
What are the 5 basic levels of customer satisfaction?
The five levels are very dissatisfied, dissatisfied, neutral, satisfied and very satisfied. Written as a numbered scale:
- Very dissatisfied: the experience clearly failed them.
- Dissatisfied: something fell short of what they expected.
- Neutral: nothing went wrong, but nothing impressed them either.
- Satisfied: the experience met their expectations.
- Very satisfied: the experience was better than they expected.
This scale feeds straight into the CSAT calculation. Scores of 4 and 5 are your positive responses, and everything else sits in the denominator only. A neutral answer is not a complaint, but it does pull your percentage down, so it is a signal that you are doing the basics without giving people a reason to come back.
One tip: keep the labels the same every time. If you change from a five-point scale to a ten-point one, or reword “satisfied” to “happy”, you can no longer compare this quarter with the last.
How do you measure Net Promoter Score (NPS)?
NPS asks how likely a customer is to recommend you from 0 to 10. Subtract the percentage of detractors (0 to 6) from the percentage of promoters (9 to 10).
The question and the three groups
Put one simple question to your customers: how likely, from 0 to 10, are they to suggest your business to a friend or colleague? Then group the replies:
- Promoters: 9 or 10
- Passives: 7 or 8 (they count in the total but not in the subtraction)
- Detractors: 0 to 6
Worked example
NPS = % promoters − % detractors
Suppose 50 clients of an accountancy practice reply. 25 give a 9 or 10, 15 give a 7 or 8 and 10 give 0 to 6. Promoters are 50% and detractors are 20%. NPS = 50 − 20 = 30. The result is a number from −100 to +100, not a percentage.
Add a follow-up
The number alone does not explain anything. Add one open question, such as “What is the main reason for your score?” The comments are often more useful than the score itself.
When to use it
NPS is a relationship measure. Send it a few times a year to your wider customer base rather than after every interaction.
What is Customer Effort Score and when should you use it?
CES measures how easy it was for a customer to get something done, and it highlights friction in your process. A service can be friendly and still be hard to use, and CES picks that up.
The statement and scale
Ask people to respond to a statement such as: “It was easy to book my appointment.” Offer an agree/disagree scale, for example from 1 (strongly disagree) to 5 (strongly agree).
How to calculate it
CES = sum of all scores ÷ number of responses
If 20 customers score a booking process and the scores add up to 90, CES = 90 ÷ 20 = 4.5 out of 5. A higher average means an easier experience. If the average is 3 or lower, look at that journey step by step.
Where it fits
CES suits journeys with a clear task: booking a viewing, sending an enquiry, getting a quote or making a complaint. Ask straight after the task is complete.
Which KPIs track customer satisfaction beyond surveys?
Retention rate, churn rate, repeat bookings, response times, complaints and online reviews show what customers actually do, not just what they say. They need no survey, so they catch the people who never reply to one.
Retention and churn
Retention rate (%) = (customers at end of period − new customers in period) ÷ customers at start of period × 100
If a gym starts the quarter with 200 members, signs up 30 and ends with 210, retention is (210 − 30) ÷ 200 × 100 = 90%.
Churn rate (%) = customers lost in period ÷ customers at start of period × 100
In that example the gym lost 20 members, so churn is 20 ÷ 200 × 100 = 10%.
Customer lifetime value
Customer lifetime value is the total a customer spends with you over the whole relationship. It is a slower signal, but if it rises, people are staying longer or coming back more often. That usually points to satisfaction.
Reviews, complaints and response times
Reviews and complaints are unprompted feedback. Nobody asked for them, so they tend to be honest. Track the number, the average star rating and the repeating themes. First response time is not a satisfaction score, but it is a service input that customers notice, so track it alongside the rest.
Comparison table
| Metric | What it measures | When to use it | Effort to collect |
|---|---|---|---|
| CSAT | Satisfaction with one experience | After an appointment, enquiry or purchase | Low: one question |
| NPS | Loyalty and likelihood to recommend | A few times a year | Low: one question plus a follow-up |
| CES | How easy a task was | After booking, enquiry or complaint journeys | Low: one statement |
| Retention and churn | Who stays and who leaves | Monthly or quarterly | Low if you keep customer records |
| Lifetime value | Long-term worth of a customer | Yearly review | Medium: needs sales history |
| Reviews | Unprompted opinion in public | Ongoing | Low: check regularly |
| Complaints | Specific failures | Ongoing | Low if logged in one place |
| First response time | Speed of your reply | Ongoing | Medium: needs timestamps |
How do you measure customer satisfaction with a survey?
Keep the survey short, send it soon after the interaction, use a consistent scale and add one open comment box. A survey that takes under a minute will get more answers than one that takes five.
Practical steps
- Choose the moment. Right after the appointment, viewing, job or enquiry reply.
- Choose the question. One main score question, plus one open question.
- Choose the channel. Email, text message or a prompt on your website, whichever your customers already use.
- Choose the scale. Use the five-point scale and keep it the same every time.
Example questions
- Dental practice: “How satisfied were you with your visit today?” and “Was it easy to book your appointment?”
- Solicitor: “How satisfied were you with how clearly we explained your next steps?” and “How likely are you to recommend us to a friend or colleague?”
- Letting agent: “How satisfied were you with how we handled your viewing?” and “It was easy to get an answer to my question.”
Adjust the wording of these examples to match your own customers.
UK data protection points
Tell people why you are asking and how you will use their answers. When you collect names, email addresses or anything else that identifies someone, handle that data in line with UK GDPR. Collect only what you need, store it securely and do not keep it longer than necessary. If a survey can be anonymous, say so. For anything specific to your situation, check the Information Commissioner’s Office guidance.
Ways to raise response rates
- Ask one question first, then offer the optional comment box.
- Send it while the experience is fresh, ideally the same day.
- Make it work on a phone with a single tap.
- Tell people what changed because of earlier feedback.
For more on survey design, see SurveyMonkey’s guide to measuring the customer experience.
How can chat conversations help you measure customer satisfaction?
A website chatbot can ask for a quick rating after each chat and show the questions customers ask most, which reveals where they are unhappy or confused. It gives you feedback from people who would never fill in a survey.
A quick rating after each chat
End each conversation with a simple CSAT prompt, such as “How satisfied were you with this answer?” Because it is one tap, it suits busy customers. If the bot is unsure of an answer, the chat can be handed to a human team, so an unhappy customer reaches a person instead of giving up.
Repeated and unanswered questions
Read through the questions people ask most. If many ask where to park, how much a service costs or how to cancel, your website or process may be unclear. Questions the bot could not answer point to gaps in what you have published. Both are satisfaction signals that surveys tend to miss.
How Chati fits in
Chati answers customers from your own website, is trained on your data and offers custom integration. It shows the sources for its answers, so you can check what a customer was told, and it hands chats to your team when it is unsure. We think it offers good value for money for small businesses. Chati is one source of feedback, not a replacement for the measures above.
How do you turn satisfaction scores into improvements?
Set a baseline, compare scores over time, find the biggest causes of low scores and act on one or two priorities at a time. Collecting scores that nobody acts on wastes your customers’ goodwill.
Build a simple layered system
A small business does not need a dashboard of twenty numbers. Start with three:
- One survey score: CSAT after each appointment or job.
- One behaviour measure: repeat bookings, retention or churn.
- One feedback source: reviews, complaints or chat questions.
Having been developing apps since 2009, we know that feedback works best when it is built into the customer journey rather than bolted on afterwards. A rating prompt at the end of a booking or chat is easier to answer than an email a week later.
Benchmark against yourself first
Your own past results are the most reliable comparison. Sector figures can give context, but different businesses word and score their surveys differently, so treat them as a rough guide only.
Close the loop
Contact unhappy customers personally. A short call or message that says “we read your feedback, and here is what we are doing” can turn a poor experience into a reason to stay. Note what you learn and what you changed.
Review on a regular rhythm
Look at the numbers monthly or quarterly. Pick the one or two biggest causes of low scores, fix them and check whether the scores move at the next review.
Would you like to see how a chatbot could help you gather feedback and answer customers from your own website? Take a look at our pricing or contact Chati and we will talk it through with you.
Frequently asked questions
What is a good CSAT score?
There is no single figure that suits every business. Track your own score over time and aim to improve it. Compare like with like, meaning the same question, scale and moment in the customer journey.
What is a good NPS score?
NPS runs from −100 to +100, and a positive score means you have more promoters than detractors. What counts as good depends on your sector, so use your own previous results as the main benchmark.
How often should you survey customers?
Send CSAT or CES surveys after specific interactions, such as an appointment or enquiry. Send relationship surveys such as NPS a few times a year. Avoid asking the same person too often.
What are the 5 C’s of customer experience?
The term is used in different ways by different writers, so there is no single agreed list. This guide focuses on measuring satisfaction, not on a fixed set of five C’s.
What are the four metrics of customer service?
There is no official set of four. Common choices are CSAT, NPS, CES and first response time. They cover satisfaction, loyalty, effort and speed.
How many responses do you need for reliable results?
More responses give a steadier picture, but we cannot give a fixed number. For a small business, focus on a consistent method and look for trends over several months. Read the written comments. They show causes that a score alone cannot.
What is the difference between customer satisfaction and customer loyalty?
Satisfaction is how a customer feels about an experience. Loyalty is whether they stay, return and recommend you. A satisfied customer may still leave, which is why behaviour measures such as retention and repeat bookings matter.
Can you measure customer satisfaction for free?
Yes. You can send a one-question survey by email or text message, read your online reviews, log complaints and work out retention from your own records. Free tools take more manual effort, but the formulas in this guide work the same way.
