Net promoter scores in small business work on the same arithmetic as anywhere else: you ask customers how likely they are to recommend you on a nought-to-ten scale, then subtract the percentage who answer nought to six from the percentage who answer nine or ten. The result runs from −100 to +100.
What changes at small scale is interpretation, not method. With thirty respondents instead of three thousand, the number moves around for reasons that have nothing to do with your business, and the useful information sits in the comments underneath it rather than in the score. A small business that treats NPS as a precise tracking metric will chase noise. One that treats it as a structured way to hear from customers gets the benefit.
The Bottom Line
- NPS = % promoters (9–10) − % detractors (0–6). Passives at 7–8 count in the denominator only.
- SurveyMonkey’s data across 150,000+ organisations puts the average at 32 and the top quartile at 72 or above.
- With a small list, the follow-up question matters more than the score. “Why?” is where the usable information is.
- Survey twice a year rather than monthly. Small samples need time to accumulate and the score barely moves in between.
- A twenty-point swing on thirty responses is usually statistical noise, not a change in your business.
Metrobi drivers are rated 4.97/5
Trusted by local businesses for:
- Background-checked professionals
- Specialized in business deliveries
- Same drivers for consistency
- 4.97/5 average delivery rating
What the net promoter score actually measures
The survey is one question: how likely are you to recommend us to a friend or colleague, on a scale of nought to ten?
Answers split into three groups. Promoters score nine or ten and are the people who recommend you unprompted. Passives score seven or eight; they are satisfied, not attached, and will leave for a better offer without much hesitation. Detractors score nought to six and range from mildly disappointed to actively warning people off.
The calculation deliberately throws away the passives. If 50% of your respondents are promoters, 30% passives and 20% detractors, your NPS is 50 − 20 = 30. The passives still count in the denominator, so they drag the score toward zero without contributing to it, which is the intended behaviour: the metric is built to measure advocacy, and indifference is not advocacy.
What it is measuring, in substance, is whether customers will do your marketing for you. That makes it a leading indicator rather than a lagging one. People stop recommending you a good while before they stop buying from you, which is why a falling NPS tends to arrive before a falling revenue line, and why it belongs in the same review as the revenue growth metrics rather than being filed under customer service.
What is a good net promoter score?
Any positive score means you have more promoters than detractors. Beyond that, the context does most of the work.
The broad convention is that anything from 0 to 30 is good, 30 to 70 is strong and above 70 is exceptional. SurveyMonkey’s benchmark data puts the cross-industry average at 32, with the top quarter of organisations at 72 or higher. Retently’s industry breakdown shows how much the figure varies by sector, which is the more useful way to read it.
Two cautions on benchmarks. First, industry matters enormously: sectors where customers enjoy the experience run far higher than sectors where the interaction is a chore, regardless of how well any individual company performs. Second, small businesses often score better than large ones in the same trade, because the owner knows the customers and the service is personal. Comparing your florist shop to a national grocery chain’s NPS flatters you without telling you anything.
The benchmark that actually helps is your own, measured the same way twice.
How to run an NPS survey with a short customer list
The mechanics are simple enough that the tooling barely matters. An email, a form, or a question at the point of delivery all work.
Ask one question, then one more. The score question, then an open “what is the main reason for your score?” The second question is where the value sits for a small business, because thirty comments are readable and thirty scores are not very informative.
Send it to everyone, not a sample. Sampling exists to control cost at scale. With a hundred customers, survey the hundred. A low response rate is the constraint you are fighting, so do not make it worse by starting with a subset.
Time it to the experience, not the calendar. Shortly after an order arrives is when the customer has something specific to say. A survey landing six weeks after their last purchase gets a general impression, which gives you less to act on.
Keep it short. Two questions. Every additional field costs responses, and with a small list you cannot afford them.
Ask twice a year. Quarterly at most. Small businesses change slowly, small samples move randomly, and asking monthly mostly teaches customers to ignore your emails.
Why small samples make the score unreliable
This is the part most NPS guidance skips, and it is the thing that matters most at small scale.
With thirty responses, each individual answer is worth 3.3 percentage points of your score. One customer moving from promoter to detractor shifts your NPS by nearly seven points on its own. Three bad deliveries in the week you happened to survey can move the number twenty points, and three good ones can move it back, without anything about your business having changed.
The practical consequences:
Do not set targets on small-sample NPS. A target of “NPS above 45” on thirty responses is a target on the weather. You will hit it and miss it in alternating quarters and learn nothing.
Do not compare quarter to quarter. Compare year to year, where you have accumulated more responses and seasonal effects cancel out.
Count the raw numbers, not just the percentage. “Four detractors this quarter, up from one” is a real and investigable fact. “NPS fell from 52 to 31” is the same fact dressed up as precision.
Read every comment. With thirty responses this takes fifteen minutes and is where the return on the whole exercise lives. Patterns in free text are visible at a sample size where patterns in scores are not.
The benefits of tracking NPS in a small business
Given all those caveats, why bother? Four reasons, and none of them is the number.
It creates a channel customers will actually use. Most unhappy customers say nothing; they simply stop ordering. A survey is a low-friction invitation to complain to you rather than about you, which is the cheapest market research available.
It identifies your promoters by name. This is the most underused output. You finish the survey knowing exactly which customers would recommend you, which is the list to ask for reviews, referrals and testimonials. Asking a confirmed nine-out-of-ten for a review has a completely different hit rate from asking everyone.
It catches problems before revenue does. Advocacy falls first, repeat orders second, revenue third. A detractor count rising while sales look flat is advance warning, and the gap is usually a quarter or two.
It connects to money. Promoters order more often and refer, which raises lifetime value and lowers acquisition cost. The connection runs through order frequency and average order value, both of which tend to be higher among customers who would recommend you. That is also what makes retention spend justifiable: you can run the ROI calculation on a retention initiative once you know what a retained customer is worth.
What to do with detractors, passives and promoters
Different groups need different responses, and treating them identically wastes the survey.
| Group | Score | What they need | What you get |
|---|---|---|---|
| Detractors | 0–6 | A direct reply, fast, from a person | Recovery, and a specific fault you can fix |
| Passives | 7–8 | A reason to prefer you over the alternative | The largest available pool of new promoters |
| Promoters | 9–10 | An easy way to tell other people | Reviews, referrals, repeat orders |
Detractors first, and personally. A phone call about a specific complaint resolves a surprising proportion of them, and even when it does not, you learn what went wrong in enough detail to prevent a repeat. Much of what turns a detractor around is having a clear policy to fall back on rather than improvising, and return policy tips that build trust covers how to write one that settles complaints instead of prolonging them.
Passives are the group most businesses ignore and the one with the most upside. They are not unhappy, which means there is no fault to fix; there is a gap in distinctiveness. Moving a passive to a promoter is usually about reliability and small courtesies rather than discounting, which matters because discounting the problem away damages your profit margin in exchange for a score.
Promoters need a prompt, nothing more. Most people who would happily recommend you never get round to it unless asked at the moment they are feeling positive.
How NPS fits with your other metrics
On its own, NPS answers one question: are customers willing to recommend us? That is valuable and narrow, and it needs company.
Pair it with repeat order rate, which is the behavioural version of the same question and immune to survey bias. A high NPS with a falling repeat rate means people like you and are buying elsewhere anyway, which usually points at price or convenience rather than service.
Pair it with your cash numbers, too. Service quality often degrades under cash pressure, because a business short of working capital cuts the things customers notice, so a falling NPS alongside a lengthening cash conversion cycle is a connected story rather than two problems.
NPS sits in the customer group of the small business metrics worth tracking, which also sets out how often to review each one.
Frequently asked questions
How do you calculate net promoter score?
Subtract the percentage of respondents scoring 0–6 from the percentage scoring 9–10. Those scoring 7–8 are excluded from both figures but still count in the total used to work out the percentages.
What is a good NPS for a small business?
Any positive number is a reasonable starting position, and the conventional reading is that 0–30 is good, 30–70 strong and 70+ exceptional. Judge against your own previous score and your specific trade rather than a cross-industry average, which blends sectors that are not comparable.
How many responses do you need for a reliable NPS?
Reliability improves sharply past about a hundred responses. Below that the score is indicative rather than precise, and the free-text answers are the more trustworthy output. This is not a reason to skip the survey, only a reason not to set targets on the number.
How often should a small business survey for NPS?
Twice a year for most, quarterly at the outside. Small samples need time to accumulate, the underlying score moves slowly, and over-surveying reduces response rates on the rounds that matter.
Is NPS better than a star rating?
They do different jobs. Star ratings on public platforms influence buyers directly. NPS is a private diagnostic that tells you what to fix and who to ask for a review. Most small businesses benefit from both, used for those separate purposes.
Getting started
Write the two questions, send them to your whole customer list, and read the replies. Calculate the score so you have a baseline, then put the number aside and work on the comments, which is where the instruction is.
Six months later, do it again. The second score is the first one that means anything, because it has something to be compared to, and by then you will also know which of your customers to ask for a referral.
