How to Calculate an NPS Margin of Error
One critique sometimes heard about NPS scores is that they can’t be stat tested. That is not true. Though I know of no statistical packages that do it automatically, it is an easy calculation you can do on your own. Here’s the math:
n = sample size
a = % promoters
b = % passives
c = % detractors
NPS = a – c
First, calculate the standard error:
With that, you can easily calculate the 95% confidence interval (or margin of error) for the NPS score:
NPS ± 1.96 (SE)
If you want to compare two groups: Calculate the difference between the two NPS scores. Then calculate the standard error for each group separately. Then calculate this 95% confidence interval:
If this interval contains the zero value, then the difference is not statistically significant.
Note that applying the typical z-test for percentages normally used in market research is not appropriate, and indeed these calculations are not the same as the margins of error for simple proportions.
So, while there are lots of reasons you might want to avoid using NPS as your be-all and end-all customer satisfaction measure, don’t let your inability to figure out statistical testing be one of them. Toss these formulas into an excel spreadsheet and you’re good to go.
P.S. Many thanks to Mark Schieffer, Kanglung Wong, and Dong Xiaoming at Sprint for pointing out a formula typo in our original post on this topic!