ds@mcpanalytics.aiA data scientist you can email your data and question to, and get a reliable response. Not AI slop.A data scientist you can email. Not AI slop.

Is your Net Promoter Score real?
Calculate it from your survey scores

Drop in the raw 0 to 10 answers from your survey export. See your score with its 95% confidence interval, and how each plan, region or survey quarter compares, using the example file or your own.

Try it on example data

How the score is worked out

Ask customers how likely they are to recommend you, from 0 to 10. Answers of 9 or 10 are promoters, 7 or 8 are passives and 0 to 6 are detractors. Your Net Promoter Score is the percentage of promoters minus the percentage of detractors, so it runs from -100 to 100.

For example, if 200 people answer and 100 are promoters, 60 are passives and 40 are detractors, the score is 50% minus 20%, which is 30. A score from a small sample can move a long way by chance, which is why the tool gives the interval next to the score.

Three questions this answers

  1. What is our Net Promoter Score, and how far could it be off?
  2. Is the gap between two plans, regions or quarters real, or is it noise?
  3. How many responses stand behind each number?

What your file needs

  • One row per survey response.
  • A column with the 0 to 10 answer to the likelihood to recommend question. At least 10 usable responses.
  • Optional: one column to compare groups, such as plan, region or survey quarter.
  • A 1 to 5 satisfaction rating is a different question. The tool stops and says so instead of converting it.

What you get back

  • The score in points, with its 95% confidence interval.
  • The share of promoters, passives and detractors.
  • The full 0 to 10 distribution of answers.
  • The score for each group, each with its own interval.
  • A plain statement of whether the best and worst groups really differ, differ only within noise, or have too few responses to test.
  • The method named and the R code.

Questions people ask

How do I calculate Net Promoter Score?

Subtract the percentage of detractors (0 to 6) from the percentage of promoters (9 or 10). Passives (7 or 8) count in the total number of responses but not in either percentage.

What is a good score?

It depends on your industry and your own history, so one threshold is a poor guide. Compare with your last result and across your groups, and look at the interval before treating a change as real.

How many responses do I need?

The interval answers it for your data: with few responses it is wide. The tool needs at least 10 usable answers to compute a score, and it does not test a gap when either group in the comparison has fewer than 30 responses.

Is a 5 point change real?

Not necessarily. If the intervals of the two results overlap, the tool says the gap must not be called a real difference.

Can I calculate it from a survey export?

Yes. Export the responses as a CSV or Excel file and drop it in. The score column must hold the raw 0 to 10 answers, not counts of promoters and detractors.

Have your survey scores ready?

Try the example file first, then drop in your own export.

Try it on example data

Related