NPS Calculator — Net Promoter Score from your survey

The NPS calculator below converts your raw survey response counts into a Net Promoter Score — a single number, from −100 to +100, that summarizes how many of your customers would actively recommend you versus warn people away.

Arb Digital tracks NPS and other customer feedback metrics as part of the reporting we build for clients running retention and referral-driven growth programs, because a rising or falling NPS trend is often the earliest signal of a business problem — or a business win — long before it shows up in revenue numbers.

What This NPS Calculator Does

You enter three numbers pulled directly from your survey results: how many respondents scored you 9 or 10 (promoters), how many scored 7 or 8 (passives), and how many scored 0 through 6 (detractors) on the standard "How likely are you to recommend us to a friend or colleague?" question, rated on a 0-10 scale. The calculator works out what percentage of your total responses falls into each group, subtracts the detractor percentage from the promoter percentage to produce your NPS, and gives you a plain-language benchmark verdict for where that score sits.

How to Use It

1. Run the standard NPS survey question — "On a scale of 0 to 10, how likely are you to recommend us to a friend or colleague?" — to your customers.
2. Sort every response into one of three buckets: promoters (9-10), passives (7-8), or detractors (0-6).
3. Enter the count for each bucket into the calculator — not percentages, just raw response counts.
4. Read your NPS and the benchmark verdict, then check the percentage breakdown to see which group is driving your score.

The Formula — How NPS Is Calculated

Net Promoter Score is calculated as the percentage of promoters minus the percentage of detractors, ignoring passives in the subtraction entirely — though passives still count toward your total response base, which means they mathematically dilute both percentages. The formula is: NPS = %Promoters − %Detractors. The result is always a whole number between −100 (every respondent is a detractor) and +100 (every respondent is a promoter) — it is a score, not a percentage, and should always be reported as a plain number. The metric was originally developed and popularized by Fred Reichheld together with Bain & Company and Satmetrix, and it remains one of the most widely used customer loyalty metrics in business today; independent research on the methodology and its use is documented by firms like Qualtrics.

Why Passives Quietly Hurt Your Score

Passives are the most misunderstood part of NPS. They aren't unhappy — a 7 or 8 out of 10 usually reflects a genuinely satisfied customer — but they also aren't enthusiastic enough to actively recommend you, and they're often the first to leave for a slightly better competitor offer since they have no real loyalty pulling them back. Because they count in the total response base but contribute to neither the promoter nor the detractor percentage, every passive response mathematically dilutes your score compared to a world where that same respondent had been a clear promoter. A business with 60% promoters, 30% passives, and 10% detractors scores a +50 NPS; if those same 30% passive respondents had instead been promoters, the same detractor rate would produce a +80 NPS. Passives represent your biggest single opportunity for score improvement, because moving them isn't about fixing something broken — it's about turning "fine" into "great."

Reading Your Score Against Benchmarks

NPS benchmarks are a useful sanity check, but they vary enormously by industry, so treat any global rule of thumb loosely: a negative score generally signals real, urgent problems, since it means you have more detractors than promoters. A score between 0 and 30 is typically considered decent — good, but with clear room to improve. A score between 30 and 70 is generally considered great, reflecting a genuinely loyal, advocacy-driving customer base. Scores above 70 are rare and are usually reserved for a small handful of category-leading brands with an unusually devoted following. These bands are useful directional guides, not hard scientific thresholds — a "good" NPS for a budget telecom provider and a "good" NPS for a premium software product can look very different because customer expectations and switching costs differ so much by category.

Sample Size and the Honest Limits of NPS

NPS becomes unreliable with small sample sizes — under roughly 100 responses, the score can swing wildly from a handful of extra promoters or detractors, so treat scores from small surveys as directional rather than precise, and avoid making big decisions off a 20-response survey. It's also worth being upfront about the methodology's real limitations: NPS is easy to game by prompting happy customers to respond while quietly filtering out unhappy ones, it collapses a lot of nuance into a single number, and published industry-average benchmarks vary so much between sources that comparing your score against a generic "industry average" figure found online is often misleading. The honest, defensible use of NPS is tracking your OWN score's trend over time, segmented consistently the same way survey after survey, rather than chasing a league-table ranking against companies with different customer bases, price points, and survey methodologies.

• Always pair the numeric score with the open-ended "why did you give that score?" follow-up — the score tells you what happened, the follow-up tells you why.
• Segment your NPS by customer type, plan tier, or tenure — a blended company-wide score can hide a serious problem in one segment.
• Re-survey consistently (monthly or quarterly) and watch the trend line, not any single reading in isolation.

When to Send the NPS Survey

Timing affects your results as much as wording does. A "relationship" NPS survey, sent periodically — monthly or quarterly — to your whole customer base regardless of recent activity, measures overall sentiment and is the version most commonly used for the headline company-wide score and trend line. A "transactional" NPS survey, sent immediately after a specific interaction like a support ticket, a purchase, or an onboarding milestone, measures satisfaction with that specific touchpoint and is far more actionable for fixing a particular process, but shouldn't be blended with relationship scores as if they measure the same thing. Mixing the two survey types into a single reported number is a common source of confusing, contradictory NPS trends — a company might see a strong transactional score right after a smooth purchase experience while its relationship score quietly erodes over months due to unrelated product or pricing issues. Decide upfront which version you're running, be consistent about it, and report the two separately if you use both.

What to Do With a Low Score

A low or negative NPS is uncomfortable but genuinely useful information — it means more of your customers would actively warn others away from you than recommend you, and it's far better to know that than not. The first move is almost always to read through the open-ended responses from detractors specifically, since patterns tend to emerge quickly — a recurring complaint about pricing, support response times, a specific feature gap, or onboarding friction. Resist the urge to fix everything at once; pick the single most-repeated detractor complaint, address it, and re-survey to see whether the score moves before chasing the next issue. It's also worth following up individually with detractors where possible — a quick, genuine response to a frustrated customer's feedback can sometimes convert them into a passive or even a promoter, and at minimum it signals that the survey wasn't just going into a black hole.

Common Mistakes to Avoid

• Reporting NPS as a percentage. It's a plain score from −100 to +100, not a percentage — don't add a "%" sign to it.
• Comparing your score to a generic online "industry average." Benchmarks vary hugely by source and methodology; your own historical trend is far more reliable.
• Trusting a score from a tiny sample. Under roughly 100 responses, the score can swing sharply from just a few results — collect a meaningful sample before acting on it.
• Skipping the open-ended follow-up question. The score alone tells you almost nothing about what to actually fix or double down on.
• Only surveying happy customers. Cherry-picking who receives the survey — for example, only emailing recently satisfied support tickets — artificially inflates the score and defeats its purpose.

Related Free Tools From Arb Digital

If customer sentiment feeds into your broader marketing reporting, pair this with our Engagement Rate Calculator for social proof metrics, or the CTR Calculator and AdSense Revenue Calculator if you also run paid campaigns and content monetization. For creator and influencer partnerships that can amplify happy-customer stories, check the Influencer Rate Calculator. Browse everything else in our free online tools hub.

Frequently Asked Questions