Measuring Customer Satisfaction on WhatsApp: A CSAT and NPS Guide
Not just collecting ratings, but a complete measurement program. Learn how to calculate CSAT and NPS step by step, when to send the question, what response rate to expect, and how to turn numbers into decisions that improve your customers' experience.
Short answer: Measuring customer satisfaction on WhatsApp rests on two core metrics, CSAT which captures satisfaction with a specific interaction, and NPS which captures how likely a customer is to recommend you. CSAT is the number of satisfied replies (4 and 5 out of 5) divided by total replies, times 100. NPS is the percentage of promoters (9 and 10) minus the percentage of detractors (0 to 6). You send the question right after a conversation closes, through a short WhatsApp message, then turn the results into an improvement plan.
Plenty of stores collect ratings and drop them into a file they never open again. The difference between "collecting ratings" and a real measurement program is that you turn numbers into decisions, know exactly when to ask, calculate your metrics correctly, and track their trend month over month. This guide gives you a practical method from scratch.
Why measure satisfaction on WhatsApp specifically
WhatsApp is already the channel your customer uses to reach you, so the question arrives in the same context without pushing them to an external survey or an email they may never open. A reply takes one tap or a single number from 1 to 5, which lifts participation compared with other channels.
The second advantage is that you tie each rating to a real conversation, so you know which agent served the customer, what the request was, and how long the reply took. An isolated low score tells you little, but a score linked to a conversation lets you diagnose the cause precisely. If you run a WhatsApp bot that sends the rating question automatically after a ticket closes, measurement becomes continuous with no manual effort.
CSAT versus NPS and when to use each
CSAT stands for Customer Satisfaction Score and measures satisfaction with a specific interaction, such as "How was your experience with our service today?" from 1 to 5. It is a moment-in-time metric that fits after every support conversation or after an order is received.
NPS stands for Net Promoter Score and measures long-term loyalty with one question: "From 0 to 10, how likely are you to recommend our store to a friend?" You do not send it after every conversation, you send it periodically (once a quarter, for example) to measure the overall relationship rather than a single touchpoint.
The simple rule is to use CSAT to improve daily operations like reply speed and resolution quality, and use NPS to track the long-term health of your brand. The two complement each other and neither replaces the other.
How to calculate CSAT step by step
The calculation is straightforward:
- Ask the customer to rate their experience from 1 to 5.
- Count replies of 4 and 5 as "satisfied."
- Divide satisfied replies by total replies received.
- Multiply by 100 to get a percentage.
A real example: a small electronics store received 80 replies to its rating question in one week, 60 of them scored 4 or 5. The math is 60 divided by 80 equals 0.75, times 100 equals 75%, so their CSAT is 75%. Anything above 80% is considered good in retail, but the trend of the number month over month matters more than the number on its own.
How to calculate NPS step by step
NPS splits your customers into three groups based on their 0 to 10 answer:
- Promoters: those who gave 9 or 10.
- Passives: those who gave 7 or 8, and they are left out of the formula.
- Detractors: those who gave 0 to 6.
The formula is the percentage of promoters minus the percentage of detractors. Example: a restaurant sent the question to 200 customers, 120 replied, with 66 promoters and 24 detractors. Promoters are 66 divided by 120 equals 55%, and detractors are 24 divided by 120 equals 20%. NPS equals 55 minus 20, which is 35. The result ranges from minus 100 to plus 100, any positive number is acceptable, above 30 is good, and above 50 is excellent.
When to send the rating question (timing makes the difference)
Timing affects the accuracy of the number and the response rate more than the wording of the question itself. Each conversation type has an ideal moment:
| Interaction type | When to send | Suitable metric |
|---|---|---|
| Support conversation | 5 to 15 minutes after the ticket closes | CSAT |
| Order received or delivered | One to two days after delivery | CSAT |
| After the first purchase | One week after the experience | CSAT |
| Overall relationship with the store | Every 3 months to all active customers | NPS |
| After a complaint is resolved | Within 24 hours of closing the complaint | CSAT |
The general rule is to send while the experience is still fresh in the customer's mind but not in the middle of tension. If you ask while the customer is still upset, the result comes out worse than reality, and if you wait two weeks they forget the details. Automating the send guarantees the same timing for every customer without relying on an agent to remember.
Expected response rates and how to raise them
Response rates on WhatsApp surveys are usually clearly higher than email, because the customer is already inside the app. Roughly speaking, a single-tap or short-number question can bring a response rate somewhere between 25% and 45%, against much lower figures for email. The number depends on your business type and how well you time the request.
To raise your response rate:
- Keep the question very short, one number or one button, with no paragraphs.
- Ask only one question in the first message, and request details after they reply.
- Personalize the message with the customer's name and the agent who served them.
- Do not repeat the question to the same customer within a short window so it never feels like a nuisance.
If you want measurement to run without daily intervention, connect it to your WhatsApp customer service system so the question fires automatically the moment an agent closes the conversation.
How to turn results into real improvements
A number with no action is just decoration. A complete measurement program turns every result into a step:
- Every low score (1 or 2) opens an immediate follow-up, you reach the customer within hours and solve what upset them, which recovers customers you were about to lose.
- Categorize the reasons behind low scores monthly: slow reply, missing product, price? Once you see the recurring cause you know where to fix.
- Compare CSAT across agents to learn who needs training and who deserves recognition.
- Track the NPS trend quarterly, and if it drops three quarters in a row you have a structural problem to address.
Tie every improvement to a before-and-after number. For example, if slow replies caused half your low scores, enable an automated reply for first messages and measure the CSAT difference next month. A fast automatic reply usually raises satisfaction because the customer feels their message landed instantly, and you can enable it from your auto reply settings.
A practical step you can start today
Begin with a single metric: enable a CSAT question after every support conversation for one month, calculate the percentage weekly, and categorize the reasons behind any score below 3. Once the process settles, add quarterly NPS. Continuous measurement beats perfection, and a store that tracks one number consistently is better off than a store that gathers ten numbers and never reads them.
WhatsLoop lets you send rating questions automatically after every conversation, tie the result to the agent and the order, and follow your metrics in one place. Try running your first measurement loop and see for yourself how numbers turn into decisions that improve your customers' experience.
