Net Promoter Score: How to Calculate NPS From Survey Responses

Step-by-Step Calculation

Step 1: collect scores. Step 2: count promoters, passives, and detractors. Step 3: convert each to a percentage of total responses. Step 4: subtract the detractor percentage from the promoter percentage. Report the result as a whole number without a percent sign. Because NPS is a difference, small changes in either group move it: shifting 5% of respondents from detractor to promoter raises NPS by 10 points.

Report the number of responses along with the score.

Sample Size and Uncertainty

Like any survey statistic, NPS has sampling error. Its standard error is √[(p + d − (p − d)²) ÷ n] where p and d are the promoter and detractor proportions. For p = 0.62, d = 0.14, n = 500: p + d = 0.76; (p − d)² = 0.2304; variance = (0.76 − 0.2304) ÷ 500 = 0.00106, standard error = 0.0325, or 3.3 points. The 95% margin of error is ±6.4 points. With only 50 responses, the margin is about ±20 points, too wide to detect modest changes. Aim for several hundred responses and be cautious when comparing small differences over time or between segments.

Benchmarks and Context

What counts as a good NPS depends on the industry: scores above 0 are positive, above 30 are often considered good, above 50 excellent, and above 70 exceptional, though typical levels differ widely (subscription software, banks, airlines, and retailers each have their own norms). Compare with direct competitors and your own trend. NPS measures loyalty intent, not satisfaction with a specific interaction, and it can be influenced by survey timing, wording, and culture: customers in some countries tend to give lower scores overall.

Acting on the Results

Follow up with an open-ended question asking for the reason for the score. Close the loop with detractors quickly to resolve their problems. Ask promoters for referrals or reviews. Analyze scores by segment, product, and touchpoint to find drivers. Critics note that a single number hides detail and that the 0 to 6 grouping lumps together very unhappy and mildly disappointed customers. Combine NPS with other measures, such as customer effort score, churn, and retention, and with qualitative feedback.