Customer satisfaction surveys: an NPS guide for shops and small businesses

A happy customer comes back and talks about you. An unhappy one often leaves without a word. A customer satisfaction survey lets you hear both before it is too late, and the Net Promoter Score (NPS) is its simplest form: one question, one score you track month after month. This guide covers how to calculate it, when to send the survey, which questions to ask and how to use the answers.

At a glance

• NPS rests on a single question: "How likely are you to recommend us to a friend or colleague?", scored from 0 to 10.
• The calculation: percentage of promoters (9 and 10) minus percentage of detractors (0 to 6). The score runs from −100 to +100.
• The right moment: straight after a purchase, a delivery, an appointment or a repair.
• Comments matter more than the score: they tell you what to fix, and in which store.
• NPS and Google reviews work together: the first is private and measures loyalty, the second is public and drives local visibility.

What is NPS?

The Net Promoter Score (NPS) measures how likely a customer is to recommend your business. It rests on one question: "On a scale of 0 to 10, how likely are you to recommend us to a friend or colleague?" Based on their score, customers fall into three groups, which together give a result between −100 and +100.

• Promoters (9 and 10): loyal customers who come back and recommend you.
• Passives (7 and 8): satisfied, but ready to leave for a better offer.
• Detractors (0 to 6): unhappy customers who may not return and may tell others about it.

The strength of NPS is its simplicity. The question is the same everywhere, it takes two seconds to answer, and the score is easy to compare from one month to the next or from one store to another.

How do you calculate NPS?

You calculate NPS by subtracting the percentage of detractors from the percentage of promoters. Passives count in the total number of responses, but are neither added nor subtracted. The result is a whole number between −100 and +100, written without a percent sign. A positive score means you have more promoters than detractors.

Example with 200 responses:

With no responses there is no NPS: a score of 0 would mean as many promoters as detractors, which is real information. With few responses, the score moves fast. Out of 30 responses, a single customer who goes from promoter to detractor costs almost 7 points. Read the trend over several months rather than the figure for one week.

When should you send a customer satisfaction survey?

Send the survey right after a moment the customer can judge: a delivered order, an in-store purchase, an appointment or a repair. The memory is still fresh and the score reflects a specific experience. If you wait several weeks, customers end up rating your brand in general, and the answer becomes hard to act on.

Depending on your business:

• Online shop: after delivery, once the customer has opened the parcel.
• Physical store: the same day or the day after the purchase at the till.
• Appointments (hair, beauty, consulting): a few hours after the service.
• Repair workshop: when the customer collects their device or bike, not when they drop it off.

Do not ask the same customer after every purchase. A loyal customer who gets a survey every week eventually stops answering or unsubscribes. Set a minimum gap between two surveys for the same person, for example one month.

A customer satisfaction survey template (NPS)

A good NPS questionnaire has two questions: the score from 0 to 10, then an open question asking for the reason behind it. The longer the questionnaire, the fewer customers finish it. If you also want to measure delivery, service or price, keep those questions for a separate survey that you send less often.

The template to copy:

1. "On a scale of 0 to 10, how likely are you to recommend [your business] to a friend or colleague?"
2. "What is the main reason for your score?" (optional, free text)

To go further, adapt the second question to the score: "What did you like most?" for promoters, "What should we improve first?" for detractors.

In the email, three habits help the response rate: a short subject line announcing a single question, a score customers can click directly in the email, and no account to create in order to answer.

How do you read the comments?

Verbatims are the free-text comments that come with the score, and they are what tell you what to fix. Read them by group: detractors describe what spoils the experience, promoters what makes it enjoyable. Tag each comment with a theme, such as welcome, waiting time, price or advice, then count which themes come up most often.

Three habits make the exercise worthwhile:

• Get back to detractors who left a comment. A quick call often solves the problem and shows the customer they were heard.
• Break results down by store and by team member. An average NPS of +30 can hide one store at +50 and another at +5.
• Track the trend month by month after each change (new opening hours, new delivery time, new team) to measure its effect.

NPS, CSAT or Google reviews: which one should you use?

These three measures answer different questions. NPS measures loyalty and willingness to recommend, across the whole relationship. CSAT measures satisfaction with one specific interaction, such as a support reply. Google reviews are public and drive your local visibility. Most shops get the best picture by combining NPS with reviews.

CSAT fits naturally at the close of a ticket, which is what a help desk such as the Support & tickets module does. Google reviews need their own approach and follow their own rules: our guide to Google reviews explains how to get them and reply to them. Keep one Google rule in mind: ask every customer for a review, without first picking out the happiest ones.

Measuring NPS with Eazyclick

The Eazyclick Customer satisfaction (NPS) module sends the survey automatically after a paid online order, an in-store sale to an identified customer, an attended appointment or a repair, depending on the triggers you tick. Customers answer from the email, with their score already selected. The dashboard shows the score, its trend and the breakdown by store and by team member.

• Minimum gap between two surveys for the same customer, adjustable (30 days by default), and an unsubscribe link in every email.
• Manual survey for a specific customer, on top of the automatic sends.
• Import of your Listen360 history from a CSV file, with duplicates flagged.
• Answers can be corrected: customers can change their score or add to their comment.
• An antivirus click does not count: tools that open email links to scan them do not record a score.

The Customer satisfaction module is added on top of the Local reputation module, which already manages your Google reviews, and complements Reviews & testimonials, which displays them on your website. In a repair workshop, the survey goes out when the repair order is marked "returned" in the Workshop module.

Frequently asked questions

How many responses do you need for a reliable NPS?

There is no official minimum, but the arithmetic speaks for itself. Out of 30 responses, each customer weighs more than 3 percentage points, and a promoter who becomes a detractor costs the score almost 7 points. Below a few dozen responses, follow the trend over three months rather than the figure for a single month.

Can NPS be negative?

Yes. NPS is negative when you have more detractors, who score 0 to 6, than promoters, who score 9 or 10. It can go as low as −100 if every respondent is a detractor. A negative score first calls for a careful read of the comments, to find the two or three causes that keep coming back.

Should you reward customers for answering?

Better not. A discount or a prize draw increases the number of responses, but attracts customers who answer for the reward, often with a high score and no comment. The score becomes flattering and loses its value. One question, a score to click in the email and a short message are enough to get honest answers.

How often should you measure NPS?

For a shop, continuous sending works best: one survey after each purchase, appointment or repair, with a minimum gap between two surveys for the same customer. You then read the score by month or by quarter. One large campaign a year gives a snapshot, but does not show the effect of the changes you make.

Can you import the history of another survey tool?

In Eazyclick, yes for Listen360: the history is imported from a CSV file into the Customer satisfaction module, and duplicate responses are flagged before the import. Your past scores then join the dashboard, so you keep your NPS curve without starting from scratch. Other survey tools cannot be imported.