Net Promoter Score Imperfect Measure

Net Promoter Score: Flawed? Useful? What Is It?

In the quest for finding a simple and affordable method of determining how good companies are at creating loyal customers, the Net Promoter Score was developed by Fred Reichfield around 1983.

Using only a very few questions, the most important being: “How likely are you to recommend this company to others?, the results for a company are alleged to correlate with monetary, bottom line success, but that’s common for almost any instrument or approach that is marketed and sold AND researched by the for-profit company.

Usefulness?

Net Promoter has good uses, not the least of which is that it enables companies to track their loyaly success, and whether they are going in the right direction. It’s a general way of gauging temperature, not so useful as a thermometer, but about as useful as sticking your arm out the door to see how cold it is.

Flaws? You Bet?

It’s important to understand the flaws and weaknesses of anything you use. That help you makes sense of the data.

• It’s based on customer responses to surveys, so it does NOT measure customer loyalty but measures only survey response behavior. We don’t know if those that SAY they will recommend a company actually do that. That problem is shared by almost every survey method.
• It’s probably not that useful in comparing companies. It’s simply too general and unresponsive to the specifics of each company’s business. Furthermore its questionable whether it is equally applicable to different sectors and niches.
• While there are claims that the Net Promoter Score has a relationship to the bottom line, that does not mean that it causes better business results. This is usually the case for all the metrics available, because it’s nearly impossible to prove that better scores CAUSE better revenue.
• Independent researchers have not necessarily confirmed the claims made by the originator. It’s not really clear whether there IS any predictive link between NPS and business success.
• On a more arcane statistical level, NSP uses a three category system for responses from customers, and while most won’t understand the statistical significance of this, this simplicity may end up invalidating the results. Certainly much potential data is lost.