CSAT measures satisfaction with one moment, such as a support conversation, and is asked right after it closes. NPS measures how likely a customer is to recommend the company on a 0 to 10 scale, and is asked a few times a year about the whole relationship. A support team owns CSAT, the people who own the account or the product own NPS, and a small team runs both because each answers a question the other cannot.
Use CSAT to judge conversations and the people who answer them, with a one-question survey sent when a conversation closes. Use NPS to judge the relationship, with the recommend question sent to each customer two to four times a year and never inside a support thread. Report CSAT weekly per teammate and per inbox, and NPS quarterly per account or segment. Read a high CSAT with a low NPS as a sign that the product or the price, rather than the service, is the problem.
What CSAT measures
CSAT, the customer satisfaction score, is the share of responses that rated a specific interaction as satisfied. The survey asks one question, usually how satisfied the customer was with the help they received, on a scale of three or five points. The score is the percentage who chose the top one or two. Because it is asked at the close of a conversation, it reflects that conversation and the person who handled it. The formula, the scales and the limits are set out in what is CSAT.
What NPS measures
NPS, the net promoter score, asks how likely the customer is to recommend the company to a friend or colleague, on a scale from 0 to 10. Customers who answer 9 or 10 are promoters, 7 or 8 are passives, and 0 to 6 are detractors. The score is the percentage of promoters minus the percentage of detractors, which gives a number from minus 100 to 100. The question was introduced by Fred Reichheld in a 2003 Harvard Business Review article, and the method is a registered trademark of Bain & Company, Satmetrix and Reichheld.
Because the question is about the relationship rather than an interaction, it is sent on a schedule instead of after a conversation. The comment that follows the number is where the value sits, since a detractor who explains the reason has given the account owner something to act on. A score without comments is a trend line and little more.
CSAT vs NPS compared
| CSAT | NPS | |
|---|---|---|
| Question | How satisfied were you with the help you received? | How likely are you to recommend us to a friend or colleague? |
| Scale | 3 or 5 points, sometimes stars or faces | 0 to 10 |
| Score | Satisfied responses as a percentage of all responses | Percentage of promoters minus percentage of detractors, from minus 100 to 100 |
| Timing | Right after a conversation closes | On a schedule, two to four times a year |
| What it predicts | Whether the last reply landed well, and whether a teammate's replies do | Whether the customer will stay and refer others |
| Unit | A conversation, rolled up per teammate and per inbox | A customer or account, rolled up per segment |
| Who owns it | The support team | Account management, customer success or the product team |
| Typical response | One tap, often with a comment | A number and a comment, with a follow-up call for detractors |
| Where it misleads | It scores only the customers who answered and says nothing about the quiet ones | It moves slowly and is affected by price and product as much as by service |
When to use CSAT
CSAT is the right measure when the question is about service quality at the level of a conversation. It shows whether a change to a template, a rule or a rota changed how replies landed, within a week of the change. It is also the fair way to look at individual teammates, read above a floor of responses so that one rating does not move the figure by ten points. A team with several inboxes uses it to compare each inbox with its own past months.
It is the wrong measure for the state of an account. A customer can rate every reply as satisfied and still be preparing to leave, because the replies were good and the product was not. CSAT also goes quiet when the customer does, since a customer who has stopped writing receives no survey and leaves no trace in the score.
When to use NPS
NPS is the right measure when the question is whether customers will stay and whether they would bring others. Sent to every customer on a quarterly or twice-yearly cycle, it gives a number per segment that can be compared over a year. The comment from each detractor is a reason to call. It is also the measure a board or an investor expects to see, since it has been used across industries for two decades.
It is the wrong measure for day-to-day service. It moves slowly and mixes price, product and service into one number. A quarter is also too long to wait to learn that a new teammate's replies are landing badly. Sending the recommend question inside a support thread confuses the two measures, because the customer answers about the thread they are in.
How small support and account teams combine CSAT and NPS
A team of three to twenty runs both without much overhead. CSAT is sent by the inbox when a conversation closes, capped at one request per customer per week, and read weekly per teammate and per inbox. NPS is sent from the account owner's address twice a year, timed away from any open conversation, and read per account and per segment at the quarterly review. The two are kept on separate scales and separate reports so neither is mistaken for the other.
The combination is where the reading gets useful. High CSAT with low NPS points at the product, the price or the onboarding, since the service is landing well and the customer still would not recommend the company. Low CSAT with high NPS points at the support queue, because the customer values the company and is unhappy with the replies. Both low is an account that needs a call this week, and both high is the baseline the team protects.
A third measure, the customer effort score, asks how easy it was to get the help and is sometimes added to the same survey. It is useful for a help center or a self-service flow, and a small team rarely needs it alongside CSAT. For account teams, two measures predict retention sooner than either score: whether promises were kept on time and which accounts have gone quiet. Both are described in customer success metrics beyond response time.
Where Sektra fits
Sektra sends a CSAT survey when a conversation in a shared inbox closes, on email, WhatsApp and website chat. It reports the scores per teammate and per inbox in its analytics beside response and resolution times. It does not run an NPS program, which belongs in the tool that owns the relationship, such as the CRM or a survey product. What it adds for the account side is Lens, the relationship panel beside each thread. Lens keeps each client's open commitments and history and shows how long an account has been quiet. A team reads CSAT in the inbox and NPS in the account review, with Lens as the record between them.
Which to start with
A support team that measures nothing starts with CSAT, because it can be switched on in a day, read within a week and tied to a teammate and a conversation. An account team starts with the quiet-account and open-question measures and adds CSAT to the inbox. It runs NPS from the second quarter, once there are enough customers to make a segment meaningful. A company with both teams runs both scores and compares them in the quarterly review, where the pairing says more than either number on its own.
Common questions
What is the difference between CSAT and NPS?
CSAT measures satisfaction with one interaction, such as a support conversation, and is asked right after it on a three or five point scale. NPS measures how likely a customer is to recommend the company on a 0 to 10 scale, and is asked a few times a year about the whole relationship.
Is CSAT or NPS better for customer support?
CSAT is the better measure for a support team, because it is tied to a conversation and a teammate and moves within a week of a change. NPS is better for the people who own the account or the product, since it reflects the relationship and moves slowly.
Can you use CSAT and NPS together?
Yes, and teams that measure both keep them separate. CSAT is sent by the inbox when a conversation closes and read weekly, and NPS is sent on a schedule away from open conversations and read quarterly. Reading the two side by side shows whether a problem sits in the service or in the product.
How is NPS calculated?
Customers answer the recommend question from 0 to 10. Those who answer 9 or 10 are promoters and those who answer 0 to 6 are detractors. NPS is the percentage of promoters minus the percentage of detractors, which gives a number from minus 100 to 100.
