"Customers seem happy" is not a measurement. To know whether your service is getting better or worse, and whether a change worked, you need a number you collect the same way every week.
This guide explains the three measures small businesses use most, how to calculate them, and how to avoid reading too much into small samples.
Average star rating
The simplest measure: ask customers to rate their visit from 1 to 5 stars and take the average. Everyone understands it, and it is easy to compare across weeks, locations and staff members.
Its weakness is that an average hides the spread. A 4.0 could be everyone giving 4 stars, or a mix of delighted and very unhappy customers. Always look at how many low ratings sit behind the average.
CSAT (customer satisfaction score)
CSAT is the share of customers who are satisfied. On a 1 to 5 scale, "satisfied" usually means a 4 or a 5.
CSAT = (number of 4 and 5 ratings / total ratings) x 100
Example: if 120 of 150 responses this month are 4 or 5 stars, CSAT is 120 / 150 x 100 = 80%.
CSAT is good for tracking a specific part of the experience, such as "How satisfied were you with the speed of service?", and it is easy to explain to a team.
NPS (Net Promoter Score)
NPS asks one question: "How likely are you to recommend us to a friend or colleague?" on a 0 to 10 scale. Responses fall into three groups:
- Promoters: 9 or 10
- Passives: 7 or 8
- Detractors: 0 to 6
NPS = % promoters - % detractors, giving a score between -100 and +100.
Example: 200 responses with 110 promoters (55%), 60 passives and 30 detractors (15%) gives an NPS of 55 - 15 = 40.
NPS measures loyalty more than satisfaction with a single visit. It is useful for businesses that rely on repeat customers and word of mouth.
Customer effort score
Some businesses also ask how easy something was, for example "How easy was it to book your appointment?". It is helpful for specific processes like booking, ordering or checking out, but most small venues get more value from ratings and CSAT.
Which measure should you use?
| Measure | Question | Best for |
|---|---|---|
| Average rating | How would you rate your visit? (1 to 5) | A simple weekly number for every location and staff member |
| CSAT | How satisfied were you with ...? (1 to 5) | Tracking one part of the experience |
| NPS | How likely are you to recommend us? (0 to 10) | Loyalty and word of mouth |
For most restaurants, cafes, hotels and shops, start with an overall star rating on every response, then add one CSAT-style question for the area you are improving. Add NPS if repeat business is central to your model. Tracking one or two numbers well beats tracking five badly.
Reading the numbers without fooling yourself
- Small samples swing. With only a handful of responses, one unhappy customer can move the average a lot. Look at trends over several weeks before drawing conclusions.
- Compare like with like. Compare the same form at the same location over time, not a busy city branch with a quiet one.
- Watch the response count. A rising score with falling responses may just mean fewer people are bothering to answer.
- Read the comments. The number tells you something changed. The comments tell you what.
- Change one thing at a time. If you change the menu and the staffing in the same week, you will not know which one moved the score.
Break it down to find what to fix
- By location: which branch needs attention.
- By employee: who to recognise and who to support. See employee ratings.
- By day and time: problems often cluster at peak times or on particular shifts.
- By topic: food, service, cleanliness, value. Complaint categories help here too, as described in our complaint handling guide.
How Custovra helps
Custovra collects ratings through QR code feedback forms and shows average ratings, response counts and trends per form, so each outlet has its own numbers. Employee ratings show averages per team member, and reports can be downloaded as PDF or exported as CSV for your own analysis.