Calculating NPS and Other Difference Between Percentages Variables

The Net Promoter Score (NPS) measures the difference between two percentages. It is possible to calculate the NPS as the average of a recoded variable. This is the best way to calculate the NPS. The same approach can be applied to other problems.

How the NPS calculation is usually described

The table below shows data for Apple.  Fourteen (14.4%) said they were Not at all likely to recommend Apple, 2.6% gave a rating of 1 out of 10, 2.0% gave a rating of 2, etc. If we add up the first seven categories (0 through 6), 51% of people in this data set are Detractors. Adding up categories 7 and 8 gives us 31% Neutrals, and then the remaining 18% are Promoters. If we subtract the Detractors from the Promoters, in this data set, Apple's NPS is -33.3, which is not great. (Amongst Apple customers, the NPS is much higher.)

Calculating the NPS by recoding

The table below shows the raw data for the first 30 people of the 153 people in the data set that rated Apple. The actual ratings, out of 11, are shown in the Recommend: Apple column. The second column shows the recoded data, where the original values are replaced by new values. The trick is to replace values of 0 through 6 with -100, values of 7 or 8 become 0, and values of 9 or 10 get a new value of 100. The column on the right shows the recoded data.

Once you have recoded the data this way, you can compute the NPS by computing the average, and you get exactly the same answer as you do when using the standard way.

It is better to use the recoding approach

This approach is better than the traditional approach as:

• It is much faster.
• When the approach is used in standard analysis software, automated statistical tests will correctly compute differences in NPS scores.

This approach can be reused

This same approach can be used to calculate the difference between any percentages where the percentages are defined by a categorical variable. The basic principle is to assign scores of:

• 100 to the categories that you wish to be subtracted from.
• 0 to categories to be ignored
• -100 to the categories that you wish to subtract.