Your wallboard tells you how fast calls were answered. It does not tell you whether the caller left happy. That is what customer surveys are for, and the two scores most call centres use are CSAT and NPS. This guide explains the difference between CSAT vs NPS, how to calculate each one, and how to run a short post call survey on the phone or on WhatsApp without annoying your customers.
CSAT vs NPS: what each one measures
CSAT (customer satisfaction score) measures how a customer felt about one interaction, usually the call that just ended. The question is simple: “How satisfied were you with today’s call?” Customers answer on a scale, often 1 to 5.
NPS (Net Promoter Score) measures loyalty to your business as a whole. The question is: “How likely are you to recommend us to a friend or colleague?” Customers answer from 0 to 10.
| CSAT | NPS | |
|---|---|---|
| Question | How satisfied were you with this call? | How likely are you to recommend us? |
| Scale | Usually 1 to 5 | 0 to 10 |
| Measures | One interaction | The overall relationship |
| Best for | Agent coaching, queue and process fixes | Long-term loyalty, management reporting |
| When to ask | Right after the call | Every few months, or after a key milestone |
In short, CSAT tells a supervisor what to fix on the call floor this week. NPS tells management whether customers are likely to stay and refer others. Most contact centres use both, for different audiences.
How to calculate CSAT
CSAT is the share of satisfied responses:
CSAT (%) = (number of 4 and 5 ratings / total responses) x 100
On a 1 to 5 scale, 4 and 5 count as satisfied. If 400 customers answered and 320 gave a 4 or 5, your CSAT is 80%.
Track CSAT per queue and per agent, not only for the whole centre. A centre-wide score hides the one queue where callers wait too long or the one team that needs training.
How to calculate NPS
NPS splits answers into three groups:
- Promoters: 9 or 10
- Passives: 7 or 8
- Detractors: 0 to 6
NPS = % promoters minus % detractors
If 50% are promoters and 20% are detractors, your NPS is 30. Passives count in the total but not in the score. NPS ranges from minus 100 to plus 100.
What counts as a good score depends on your industry, your customers and how you ask the question. Compare your own score month to month rather than chasing a number from another sector.
Ways to run a post call survey
IVR survey after the call
The caller hears one or two recorded questions and answers with the keypad: “Press 1 to 5.” It is quick, works on any phone and needs no internet. Keep it to one or two questions, because callers hang up on long surveys.
Outbound voice survey
On Voxdial IP PBX, Advanced IVR applications can run outbound voice campaigns such as reminders, notifications or surveys. They control how many calls run at once, how many times a number is retried and when the campaign starts, and the answered call is routed based on the caller’s response. Use this only for your own customers who have just dealt with you, during reasonable hours.
WhatsApp survey
Many customers prefer to tap an answer than to stay on the line. After the call, send a short WhatsApp message with the question, and collect replies in the shared inbox of the BitVoice WhatsApp Business API platform. A chatbot flow can ask the question, record the reply and hand the chat to an agent if the customer gives a low score and wants to talk.
Agent-recorded outcome
Not every call needs a survey. Agents in Voxdial Contact Center pick a disposition at the end of each call, such as “Resolved” or “Call Back”, and can add a remark. That is not a satisfaction score, but it gives you the outcome of every call, including the ones customers never rate.
Write questions that get honest answers
- Ask one thing at a time. “Was the agent helpful and was your issue solved?” is two questions.
- Keep the scale consistent. Use the same 1 to 5 scale everywhere so scores can be compared.
- Use the caller’s language. A survey in Hindi, Tamil or Malayalam gets more answers in those regions than one in English only.
- Ask straight after the call. Memory fades fast. An hour later, the customer is rating their whole day.
- Add one open question, sometimes. “What could we have done better?” gives you the reason behind a score.
Close the loop on low scores
A survey only helps if someone acts on it. Set a simple rule: every detractor or 1 to 2 CSAT rating is reviewed within a working day.
- Find the call in the inbound reports and play the recording.
- Check the wait time and who ended the call.
- Decide: was it the queue, the agent, the process or the product?
- Call the customer back if the issue is still open.
- Use the call in coaching, with listen and whisper on live calls the following week. See call monitoring: listen, whisper and barge.
Low scores often trace back to long waits and repeat calls rather than rude agents. Watch your first call resolution and abandon rate alongside CSAT, and read call centre KPIs and the wallboard for the full set.
Where surveys fit in your contact centre
Surveys work best as part of a human-first service. Your agents answer first; a BitVoice AI voice agent covers busy times and passes callers back with a summary; and a WhatsApp follow-up asks how it went. All of it runs on Voxdial Contact Center, which has customer satisfaction surveys built in, and the scores sit in the same reports.
Want to see a post call survey running on your own queues? Book a demo with BitVoice.
FAQs
What is the difference between CSAT and NPS?
CSAT measures satisfaction with one interaction, such as a single call, usually on a 1 to 5 scale. NPS measures how likely a customer is to recommend your business, on a 0 to 10 scale, and reflects the overall relationship.
How do you calculate CSAT?
Divide the number of satisfied responses (4 or 5 on a 1 to 5 scale) by the total number of responses and multiply by 100.
How do you calculate NPS?
Subtract the percentage of detractors (0 to 6) from the percentage of promoters (9 or 10). The result ranges from minus 100 to plus 100.
How long should a post call survey be?
One or two questions. Short IVR or WhatsApp surveys sent right after the call get far more answers than long questionnaires.
