Interval Estimation for the 'Net Promoter Score'

TL;DR

This paper investigates statistical methods for estimating confidence intervals for the Net Promoter Score, a popular customer loyalty metric, evaluating their performance on extensive real-world data.

Contribution

It adapts and assesses various interval estimation techniques for NPS, identifying the most effective methods based on empirical analysis.

Findings

Adjusted Wald method performs well

Iterative Score test shows superior accuracy

Provides practical guidance for NPS interval estimation

Abstract

The Net Promoter Score (NPS) is a novel summary statistic used by thousands of companies as a key performance indicator of customer loyalty. While adoption of the statistic has grown rapidly over the last decade, there has been little published on its statistical properties. Common interval estimation techniques are adapted for use with the NPS, and performance assessed on the largest available database of companies' Net Promoter Scores. Variations on the Adjusted Wald, and an iterative Score test are found to have superior performance.