eNPS Calculator (Employee Net Promoter Score) with Margin of Error

On a scale of -1 to +1 the variance of a single answer is p + d - (p - d)², where p and d are the shares of promoters and detractors. The standard error is the square root of that divided by the number of answers, times 100 to put it in points; the margin of error is 1.96 standard errors at 95% (Rocks, 2016; Engelhardt, 2016).

Customer NPS formulas assume a huge population. An employee survey is different: if 120 of 150 people answered, there is little left to be uncertain about. The finite population correction, the square root of (N - n)/(N - 1), shrinks the margin accordingly. Enter the number of people invited to use it.

100 answers: 50 promoters, 30 passives, 20 detractors. eNPS = 50 - 20 = +30. Variance = .50 + .20 - (.30)² = .61. Standard error = √(.61/100) x 100 = 7.81 points; 95% margin = 1.96 x 7.81 = ±15.3. If those 100 people were out of a company of 150, the correction is √(50/149) = .579 and the margin falls to ±8.9.

Last time the same company scored 40 promoters, 35 passives, 27 detractors: +12.7. The rise of 17.3 points has a standard error of about 11.1 (the test leaves out the team-size correction, to stay on the safe side), z = 1.55, p = .12: it could be chance. Report it as "up, but not yet a clear change", and look at the comments.

Any score above 0 means more promoters than detractors. Published "benchmarks" come from survey vendors' own client bases and use different questions and samples, so they are a weak guide. The more useful comparison is your own score over time, with its margin of error, and the comments people give with their score. One question cannot tell you why people feel as they do.

The variance and interval follow Rocks (2016) and Engelhardt (2016); the finite population correction is the standard one for sampling without replacement. Checked on 2 November 2026.

1. Rocks, B. (2016). Interval estimation for the "net promoter score". The American Statistician, 70(4), 365-372: https://doi.org/10.1080/00031305.2016.1158124
2. Engelhardt, A. (2016). Net Promoter Score confidence intervals 2: maths and stats: https://www.r-bloggers.com/2016/01/net-promoter-score-confidence-intervals-2-maths-and-stats/
3. Reichheld, F. F. (2003). The one number you need to grow. Harvard Business Review, 81(12): https://hbr.org/2003/12/the-one-number-you-need-to-grow

Free to use and free to quote. If a figure here is useful in something you are writing, a link back is all that is asked.

Each one answers a single question and refuses the ones it cannot. All free, none asks for an email address.