The owner of the cosmetics shop from our loyalty programme example sees an NPS (Net Promoter Score, a measure of willingness to recommend) of +15 in a report and proposes a target of +30 for the marketer. In the survey comments, customers say they could not spend their promised points at checkout. Asking for higher scores will leave that restriction in place. The shop has a different task: find where the promise broke down during the purchase and fix that step. The numbers in this lesson are illustrative.
NPS and CSAT (Customer Satisfaction Score) provide a consistent way to collect feedback. Use them alongside comments, support requests and purchase history. Agree what experience customers are rating, calculate their answers, then assign the work needed to address the problem.
Two questions for different jobs
NPS asks how likely someone is to recommend a company, product or service to a friend or colleague, on a scale from 0 to 10. For the shop, the question might be: "How likely are you to recommend our shop to someone you know?" Keep the object of the rating consistent. An answer about the whole shop and one about the latest delivery do not belong in the same series.
CSAT asks how satisfied someone is with a chosen experience. After a query about points, that could be: "How satisfied are you with the way we resolved your points redemption query?" The person needs a chance to judge the outcome. A survey immediately after an automated support reply collects impressions of the start of the conversation.
| Team's task | Question | When the customer can answer |
|---|---|---|
| Learn how willing people are to recommend the shop | NPS, scale from 0 to 10 | After a purchase and experience of the goods and service |
| Rate the resolution of a points query | CSAT, scale from 1 to 5 | Once the query is closed and the customer can see the outcome |
| Find out how difficult redemption was | An effort question, CES (Customer Effort Score) | After attempting to redeem points |
CES explores effort: how much searching, repetition and asking for help the task required. A team using it defines its question and scale separately. A table of NPS answers cannot be converted into CSAT or CES because people were rating different things.
NPS: passives stay in the calculation
Bain's NPS method divides answers into three groups: 9 and 10 are promoters, 7 and 8 are passives, and 0 to 6 are detractors. Subtract the share of detractors from the share of promoters.
The shop invited 100 customers and received 20 responses:
| Scores | Group | Responses | Share of all respondents |
|---|---|---|---|
| 9 and 10 | Promoters | 8 | 40% |
| 7 and 8 | Passives | 7 | 35% |
| 0 to 6 | Detractors | 5 | 25% |
| Total | All groups | 20 | 100% |
NPS = 8 / 20 × 100 - 5 / 20 × 100 = +15.
The seven passive responses count towards neither share, but they sit in the denominator of both. Removing them gives (8 - 5) / 13 × 100 ≈ +23.1. No customer changed their mind; only the incorrectly calculated score went up.
Report NPS as a number from -100 to +100, without a percent sign. Here +15 is the difference between 40% and 25%; the promoter share is 40%. Group names help with the calculation, while a customer's explanation matters more when following up. A score of 6 may reflect an inconvenient payment option, and a score of 9 may come with a serious complaint about delivery.
With 20 responses, one person moving from detractor to promoter changes NPS by 10 points: one share rises by 5 points and the other falls by 5. A jump from +15 to +25 at this scale calls for reading the responses before attributing it to the team's work.
CSAT: choose the scale before calculating the percentage
For points queries, the shop chose a scale from 1 to 5: 1 is very dissatisfied and 5 is very satisfied. Its CSAT calculation counts 4 and 5 as positive, following the version described by Qualtrics. Record the rule before collecting responses and keep it unchanged for the next survey.
In a separate survey after support, 20 people answered and 15 chose 4 or 5. CSAT = 15 / 20 × 100% = 75%. The report should show the question, scale, positive share and respondent count together. The arithmetic mean of the scores can be reported separately, under a different column name.
These 20 responses concern support queries. They did not come from the earlier NPS table, and the 75% does not follow from NPS +15. One customer might rate the employee's help highly while being reluctant to recommend a shop where that help was needed.
One answer out of 20 changes CSAT by 5 percentage points if it crosses the boundary between 3 and 4. Reporting "75%, 15 out of 20" makes the scale clearer than a large standalone 75%.
Who reached the survey
For NPS, 20 of the 100 invited customers responded, a response rate of 20%. The other 80 have no recorded score. Assigning them 7 or 8 and calling them passives would replace missing data with invented answers.
The invitation list matters too. If the manager sent the link only to people who praised them in chat, the shop is collecting opinions from a preselected group. A link shown only in the customer account may miss buyers who order through a manager. The report needs to name the audience reached and the invitation method.
Before the first survey, the team records:
- who receives the question and which orders qualify;
- the delay after a purchase or support query;
- the channel and exact wording;
- how many times a person may respond during the period;
- when collection ends and who reviews the comments.
Keep these conditions stable when comparing months. Switching from email to a phone call or changing the question calls for a new series or a marked break in the report. Regular customers and newcomers can be shown separately, with the size of each group. That helps reveal whether the overall score moved with the mix of respondents.
A manager's bonus for a score of 10 gives them a reason to suggest the desired answer. A better way to assess feedback work is through completed actions: the issue reached its owner, the cause was investigated and the customer received an outcome. Help does not depend on the rating: the problem still needs solving for someone who no longer wants to answer.
Turn a comment into a task
After the numerical question, one prompt is enough: "What most influenced your rating?" Keep the original wording of the answers and add the topic in a separate field. "They would not let me spend my points" might mean a till error, a product exclusion or an unclear advertising promise.
The manager clarifies the order and desired outcome with a person who left contact details and agreed to discuss the issue. They then examine the rules, balance and transaction log. An anonymous survey still offers a chance to improve the process even when the author cannot be contacted.
| Finding | Who owns it | What completion looks like |
|---|---|---|
| Points failed on an eligible product | Developer and checkout supervisor | Corrected rule and a successful payment test |
| The exclusion appeared only in lengthy terms | Marketer and programme owner | Restriction visible beside the balance and in the basket |
| The cashier did not know how to redeem points | Shop supervisor | Instructions at the workstation and a completed transaction |
Once resolved, the manager tells the customer what was done. The closed task needs an outcome: "Limit corrected, order with points redemption completed." "Passed to the developer" records a handover; work is still in progress.
Take recurring complaints to the process owner. Measure their frequency against purchases or support queries: ten complaints at different volumes of activity call for different interpretations. Customer interviews can clarify the experience, while repeat orders show actual return behaviour.
Discussing the cost with the owner
Suppose fixing redemption and the basket guidance costs 12,000 ₸. For the cosmetics shop, a 10,000 ₸ order leaves 3,000 ₸ after redeeming 500 points and paying 6,500 ₸ in variable costs. Covering 12,000 ₸ takes 12,000 / 3,000 = 4 extra orders with the same contribution.
That is a threshold for discussing costs, not an NPS forecast. Five low scores cannot be translated into four future orders: some people would buy again without the fix, while others will not return for unrelated reasons. Nor can an NPS gain in points be multiplied by average order value. Estimating additional purchases requires a comparison planned in advance.
A fault in a promised redemption feature needs fixing regardless of how many people rated it. The estimate helps choose the repair and explain its cost. The score after the fix, comments and repeat support requests show how the experience changed; orders and contribution provide a separate view of the money.
Exercise: a report that produces work
Take one survey from your project. If none exists yet, prepare a question and collection plan, then practise the calculation on the shop's data above.
The shop's report will show NPS +15 from 20 responses to 100 invitations. If a colleague got +23.1, the seven omitted passive responses explain the difference. CSAT 75% belongs to a separate support survey and is labelled as 15 positive responses out of 20.
The resulting work is to reproduce the redemption failure and assign a fix. "Raise NPS to +30" names neither the fault nor an owner. The completed exercise describes a change a customer will notice and a date for the team to revisit the outcome.