What NPS is and how the math works
Net Promoter Score asks customers one question: how likely are you to recommend us to a friend or colleague? Customers answer on a zero-to-ten scale.
Promoters score nine or ten. Passives score seven or eight. Detractors score zero through six. NPS equals the percentage of promoters minus the percentage of detractors, producing a score from -100 to +100.
At enterprise scale, tracked consistently across thousands of responses, the trend line can be useful. That is the context NPS was built for.
Why NPS breaks for small businesses
The sample-size problem is the biggest issue. With 25 or 30 responses, one or two customers can swing the score enough to look meaningful even when it is just noise.
There is also an actionability problem. If NPS drops, what do you change? You still need to ask why. The open-ended reason was the valuable part all along.
For a small business, every survey question is scarce. Spending the first slot on a number that may not be reliable is often the most expensive part of NPS.
When NPS is worth using
- You consistently collect around 100 or more responses per measurement period.
- A partner, marketplace, lender, or franchise requires the number.
- You use low scores as a trigger for personal follow-up instead of treating the score as the main metric.
What to ask instead
- What made you decide to buy from us, rather than somewhere else?
This identifies your real competitive advantage in the customer's own words.
- What almost stopped you?
This reveals conversion blockers from people who bought anyway.
- Have you recommended us to anyone, and what did you say?
This asks about actual behavior rather than predicted likelihood and gives you the language behind the recommendation.
- What is one thing we could do better?
This gives you the detractor-detection NPS promises, with the reason included.
- How disappointed would you be if you could no longer buy from us?
This can be a sharper loyalty read for small samples because the answer carries more practical meaning than a one-point movement on a scale.
The bottom line
NPS is famous, so small businesses inherit it by default. The habit behind it is good: ask customers regularly and act on what you hear.
The instrument itself is often a poor fit for low-volume feedback. Keep the habit, swap the question, and spend your limited survey slots on answers that explain what to do next.
Frequently asked questions
What is a good NPS score?
Above zero means more promoters than detractors, but benchmarks vary widely. For small samples, the confidence interval can be wider than the difference you are trying to interpret.
How is NPS calculated?
It is the percentage of promoters, who score nine or ten, minus the percentage of detractors, who score zero through six.
How many responses do I need for reliable NPS?
As a rule of thumb, at least around 100 per measurement period before trends become useful, and more if you want to detect small changes.
Is NPS worth it for a small business?
Usually not as the main metric. It can be worth using as a follow-up trigger for low scorers or when a partner requires it.
What is the best alternative to NPS for small businesses?
Reason-revealing questions: why customers chose you, what almost stopped them, whether they actually recommended you, and what they said.
Related articles
Post-purchase survey questions
The post-purchase survey questions worth asking after a customer buys, including what drove the decision, what almost stopped it, and how to use the answers.
Customer feedback vs analytics
Analytics shows what happened. Customer feedback explains why it happened. Learn how small ecommerce teams can use both without guessing.