Net Promoter Score (NPS) is calculated by subtracting the percentage of detractors from the percentage of promoters among your survey respondents. The result is a number between -100 and +100 that measures how likely your customers are to recommend your product or company. The math is simple, but getting useful results depends on asking the right question, categorizing responses correctly, and following up to understand why people gave the scores they did.
The Standard NPS Question
Every NPS calculation starts with a single survey question: “How likely is it that you would recommend [your company] to a friend or colleague?” Respondents answer on a scale from 0 (not at all likely) to 10 (extremely likely). This specific wording and scale are standardized, so resist the urge to modify the question or shorten the range. Consistency is what makes NPS comparable across time periods and against other companies in your industry.
How Respondents Are Categorized
Once responses come in, every person falls into one of three groups based on their score:
- Promoters (9 or 10): Your most enthusiastic customers. They’re likely to refer others and keep buying.
- Passives (7 or 8): Satisfied but unenthusiastic. They won’t actively recommend you and could switch to a competitor.
- Detractors (0 through 6): Unhappy customers who may discourage others from doing business with you.
The wide detractor range surprises many people. A score of 6 might feel decent on a 10-point scale, but in NPS methodology it counts the same as a 0. This is intentional. Research behind the framework found that only customers scoring 9 or 10 reliably drove referral behavior, so the threshold for “promoter” is deliberately high.
The Calculation Step by Step
Here’s how to run the math with a concrete example. Say you surveyed 200 customers and got these results:
- 80 gave a 9 or 10 (promoters)
- 70 gave a 7 or 8 (passives)
- 50 gave a 0 through 6 (detractors)
First, calculate the percentage of promoters: 80 ÷ 200 = 40%. Then calculate the percentage of detractors: 50 ÷ 200 = 25%. Now subtract: 40% minus 25% = +15. Your NPS is 15.
Notice that passives don’t appear in the formula at all. They affect the total respondent count (which changes both percentages), but they’re not directly added or subtracted. Also note that NPS is expressed as a whole number, not a percentage. You’d say “our NPS is 15,” not “15%.”
The score can range from -100 (every single respondent is a detractor) to +100 (every respondent is a promoter). Any positive score means you have more promoters than detractors. Scores above 50 are generally considered excellent, and scores above 70 are exceptional. A negative score signals serious customer satisfaction problems.
Relational vs. Transactional NPS
There are two common ways to deploy the NPS question, and each serves a different purpose.
Relational NPS measures how customers feel about your organization overall. You send it at regular intervals, typically quarterly, every six months, or annually, without tying it to a specific interaction. This gives you a high-level health check of brand loyalty and works well for year-over-year benchmarking. One important timing consideration: avoid sending relational surveys during peak sales periods or right after a specific interaction, since those experiences will color the response and skew your baseline reading.
Transactional NPS measures satisfaction with a specific touchpoint, like a support call, a purchase, or an onboarding session. You send it shortly after the interaction happens. This version helps you identify which parts of the customer experience are strong and which need work. It also gives individual teams a metric they can act on directly, since a support team can own its post-call NPS in a way it can’t own the company’s overall score.
Many companies run both. Relational NPS tracks the big picture while transactional NPS pinpoints where to make improvements.
The Follow-Up Question That Makes NPS Useful
A raw number tells you where you stand, but not why. The most important thing you can do after the 0-to-10 question is ask an open-ended follow-up. The simplest version is: “What is the primary reason for your score?”
You’ll get more actionable responses if you tailor the follow-up to each group. For detractors, ask what was missing or disappointing, or what specific issues led to their rating. For passives, ask what one change would make them rate you higher. For promoters, ask what they love most about your product or service. You can also use the promoter follow-up as a chance to invite testimonials, reviews, or referral program participation.
These qualitative responses are where the real value lives. The score itself is a signal; the comments tell you what to fix, what to protect, and what language your happiest customers use to describe you.
Sample Size and Survey Timing
NPS becomes unreliable with too few responses. With only 20 or 30 answers, a handful of extreme scores will swing the number dramatically. For a reasonably stable score, aim for at least 100 responses per measurement period. If your customer base is small, you may need to survey everyone rather than a sample, or extend your collection window.
Timing matters too. Send transactional surveys soon after the interaction while the experience is fresh, ideally within 24 to 48 hours. For relational surveys, pick a consistent schedule and stick to it so you’re comparing equivalent time periods. If you survey in January one year and July the next, seasonal factors could account for any change you see.
Tracking NPS Over Time
A single NPS reading is a snapshot. The metric becomes powerful when you track it consistently and dig into the trends. Plot your score over several quarters and look for patterns: did a product launch move the needle? Did a policy change create more detractors?
Beyond the top-line number, monitor the underlying distribution. Two companies can both have an NPS of 30, but one might have 50% promoters and 20% detractors while the other has 35% promoters and 5% detractors. The first company has a more polarized customer base. Watching how each group’s share shifts over time reveals dynamics that the single score can mask.
Segmenting your results also adds depth. Break NPS down by customer tenure, product line, geographic region, or acquisition channel. You may find that new customers are highly enthusiastic while long-term customers are growing dissatisfied, or vice versa. Those segment-level insights point directly to where your team should focus.