Net promoter score for small business: worth it or not?

Aug 22
3 min read

Updated: 2 days ago

Introduction

Net promoter score asks one question — how likely are you to recommend us — on a scale of nought to ten, and turns the answers into a single number.

For a large company it is a tracking metric. For a small business the number itself is close to useless, and the free-text answer beside it is worth a great deal. Knowing which half matters is the whole point.

1. Net promoter score for small business: how it is calculated

Respondents are grouped: 9–10 are promoters, 7–8 are passives, 0–6 are detractors.

NPS = % promoters − % detractors

Passives are ignored entirely. The result runs from −100 to +100, which is why the number looks dramatic and moves alarmingly on small samples.

2. Why the number misleads at small volume

With thirty responses, two unhappy customers can move your score by more than ten points. That looks like a trend and is noise.

The scale is also lumpy: a 7 and a 6 are adjacent answers treated as opposite categories. A handful of people rounding differently changes the headline while nothing about the business has changed at all.

If you have fewer than a few hundred responses per period, treat the score as a rough direction and nothing more. Certainly do not set targets against it.

3. The follow-up question is the actual value

Immediately after the score, ask why — as free text, optional, one line.

Those answers are the return on the whole exercise. They arrive unprompted, in the customer's own words, from people with a specific recent experience. Patterns emerge in a few dozen responses, long before the score means anything.

If you only implement half of NPS, implement this half.

4. Ask at the right moment

Timing changes the answer more than wording does. Ask after value has been delivered, not after payment.

For a service, once the outcome is visible. For a restaurant, shortly after the visit. For software, after the customer has actually used it for something real. Asking at the point of invoice measures how they feel about paying.

5. Act on detractors individually

At small scale you can do what large companies cannot: contact every detractor personally.

A short, genuine message asking what went wrong recovers a meaningful share of them, and it produces better information than any survey design. Customers are usually surprised to be contacted, which is itself a differentiator.

This is the strongest argument for running NPS in a small business — not the tracking, the follow-up.

6. Ask promoters for something specific

A 9 or 10 is the best moment you will get to ask for a review or a referral. The person has literally just told you they would recommend you.

Make the request immediate and specific: a link straight to the review form, or a named kind of person you would like introduced. Generic goodwill decays within a day.

7. When to use something simpler

NPS is not always the right instrument. Simpler questions often produce more actionable answers:

"How did you hear about us?" — attribution you cannot get any other way.
"What nearly stopped you buying?" — the objection to fix.
"What should we do better?" — direct and unambiguous.

If you can only ask one question, the attribution question is usually worth more than a score, because it changes where money goes rather than how you feel about the business.

8. Keep it to one question, always

Whatever you ask, the response rate is set by length. A single question with an optional comment gets answered; a five-question survey gets abandoned.

Resist adding "while we have them" questions. Each one costs responses, and the marginal answer is rarely worth the responses lost — particularly for a business whose sample is small to begin with.

Conclusion

Calculate NPS if you like, but treat the number as directional at small volume and never as a target.

The value sits in the free-text follow-up, in contacting every detractor personally, and in asking promoters for a review while they are still enthusiastic. If a score feels like overhead, ask how customers heard about you instead — one question, and it changes decisions rather than sentiment.