Naming Power: Why NPS® Needs a Clear Taxonomy

A better taxonomy can help Net Promoter System® companies compare results and uncover powerful insights.

By Fred Reichheld

min read


This article originally appeared on LinkedIn. It’s August, and people in the eastern United States have begun to say goodbye to the beach days of summer, their minds turning to the return of school and cooler weather. The Wall Street Journal recently published an interesting piece about another September rite in this part of the world: hurricane season. In it, the author, Jim Sollisch, posits that if we were to name hurricanes after something scary, people might behave with more caution, heeding, for example, the precautionary evacuation notices. Hurricanes Florence and Dolly would be more appropriately called something like “Hurricane Shark,” “Slasher” or “Demon,” he argues. “Names have the power to change reality,” writes Sollisch, a partner at Cleveland advertising agency Marcus Thomas. After his neighbor pointed out that the new flower blooming in one of his beds was not a flower at all but actually a weed called ragwort, Sollisch says his perspective changed entirely. “I went from admiring it to poisoning it, all because it didn’t have a nice name like Susan or marigold,” he writes. Properly naming things has always been important. From the early responsibility of Adam and Eve in Genesis, to the binomial Latin identification of species pioneered by 18th-century Swedish scientist Carl Linnaeus (think Homo sapiens), names are both the basis of identification and critical to learning and accumulating knowledge. The Net Promoter System® has nothing like the complexity of nature, but it has expanded to the point where I think it, too, needs a more specific and agreed-upon taxonomy. Today dozens of companies report Net Promoter Scores® to stock analysts and investors, and many more tie important business decisions to them. If each company uses its own unique definition and calculation methodology, calling all that different output “NPS®,” it creates chaos, not insight. In order to move toward comparable results generated by a consistent and reliable methodology, here are a few terms I suggest as a starting point. Transaction NPS: Feedback gathered shortly after a specific interaction (one sufficiently important that it can sway relationship health). Example: A bank asks for feedback from customers who apply for a credit line increase. Episode NPS: Feedback gathered at the end of a series of interactions required to fulfill a customer need (sometimes called a sub-journey). Example: A lender solicits feedback on the mortgage experience, including shopping for a mortgage, the credit application, the rate-lock process and finally closing on a home. Product/category NPS: Feedback from a sample of customers who purchased or used a product over a recent time period. Example: An online retailer gets feedback from customers who purchased groceries online during the past 18 months and compares its NPS to that of key competitors. Relationship/brand NPS: Feedback triggered at the anniversary of relationship initiation that’s not tied to any specific interaction, but rather reflects feelings about the overall relationship. Example: A company gathers likelihood-to-recommend scores (0–10) from a sample of all customers who have done business across all product lines and categories. Of course, there is also employee NPS, but for now, let’s stick with customers. In addition to naming and understanding what a particular Net Promoter Score measures, there are several other dimensions we need to clarify. Did the calculation use the standard definitions for promoter (9–10), passive (7–8) and detractor (0–6)? What sampling method was used, and what was the response rate? Is it the product of a self-administered survey or generated by double-blind research? Or is it perhaps gleaned not from a survey at all but from interpreting operational data or customer behaviors? These are dimensions we will address in future posts. Without clear names and reporting rules that specify what these names stand for, I worry we will miss out on powerful insights that can occur when we name things carefully. The Wall Street Journal recently wrote about companies reporting NPS results with no clear rules about names and definitions. Let’s begin to fix this problem. Do the four flavors of NPS I’ve listed above accurately describe the ways that you use customer NPS in your company? Net Promoter^®, Net Promoter System^®, Net Promoter Score^® and NPS^® are registered trademarks of Bain & Company, Inc., Fred Reichheld and Satmetrix Systems, Inc.

This article originally appeared on LinkedIn.

It’s August, and people in the eastern United States have begun to say goodbye to the beach days of summer, their minds turning to the return of school and cooler weather. The Wall Street Journal recently published an interesting piece about another September rite in this part of the world: hurricane season.

In it, the author, Jim Sollisch, posits that if we were to name hurricanes after something scary, people might behave with more caution, heeding, for example, the precautionary evacuation notices. Hurricanes Florence and Dolly would be more appropriately called something like “Hurricane Shark,” “Slasher” or “Demon,” he argues.

“Names have the power to change reality,” writes Sollisch, a partner at Cleveland advertising agency Marcus Thomas. After his neighbor pointed out that the new flower blooming in one of his beds was not a flower at all but actually a weed called ragwort, Sollisch says his perspective changed entirely. “I went from admiring it to poisoning it, all because it didn’t have a nice name like Susan or marigold,” he writes.

Properly naming things has always been important. From the early responsibility of Adam and Eve in Genesis, to the binomial Latin identification of species pioneered by 18th-century Swedish scientist Carl Linnaeus (think Homo sapiens), names are both the basis of identification and critical to learning and accumulating knowledge.

The Net Promoter System® has nothing like the complexity of nature, but it has expanded to the point where I think it, too, needs a more specific and agreed-upon taxonomy. Today dozens of companies report Net Promoter Scores® to stock analysts and investors, and many more tie important business decisions to them. If each company uses its own unique definition and calculation methodology, calling all that different output “NPS®,” it creates chaos, not insight.

In order to move toward comparable results generated by a consistent and reliable methodology, here are a few terms I suggest as a starting point.

Transaction NPS: Feedback gathered shortly after a specific interaction (one sufficiently important that it can sway relationship health). Example: A bank asks for feedback from customers who apply for a credit line increase.
Episode NPS: Feedback gathered at the end of a series of interactions required to fulfill a customer need (sometimes called a sub-journey). Example: A lender solicits feedback on the mortgage experience, including shopping for a mortgage, the credit application, the rate-lock process and finally closing on a home.
Product/category NPS: Feedback from a sample of customers who purchased or used a product over a recent time period. Example: An online retailer gets feedback from customers who purchased groceries online during the past 18 months and compares its NPS to that of key competitors.
Relationship/brand NPS: Feedback triggered at the anniversary of relationship initiation that’s not tied to any specific interaction, but rather reflects feelings about the overall relationship. Example: A company gathers likelihood-to-recommend scores (0–10) from a sample of all customers who have done business across all product lines and categories.

Of course, there is also employee NPS, but for now, let’s stick with customers. In addition to naming and understanding what a particular Net Promoter Score measures, there are several other dimensions we need to clarify. Did the calculation use the standard definitions for promoter (9–10), passive (7–8) and detractor (0–6)? What sampling method was used, and what was the response rate? Is it the product of a self-administered survey or generated by double-blind research? Or is it perhaps gleaned not from a survey at all but from interpreting operational data or customer behaviors? These are dimensions we will address in future posts.

Without clear names and reporting rules that specify what these names stand for, I worry we will miss out on powerful insights that can occur when we name things carefully. The Wall Street Journal recently wrote about companies reporting NPS results with no clear rules about names and definitions. Let’s begin to fix this problem.

Do the four flavors of NPS I’ve listed above accurately describe the ways that you use customer NPS in your company?

Net Promoter^®, Net Promoter System^®, Net Promoter Score^® and NPS^® are registered trademarks of Bain & Company, Inc., Fred Reichheld and Satmetrix Systems, Inc.

August 19, 2019

Want to continue the conversation?