Net Promoter Labels Obscure Actual Recommendation Patterns

We recently published a benchmark of Net Promoter Scores of 180 companies across 19 industries. Within that research, we showed that promoters are more likely than detractors to repurchase. In a previous blog post, we examined how promoters and detractors actually recommend companies.

In this post, we go a step further and look at how consumers actually recommend based on the specific response to the NPS question. As you can see in the graphic below:

  • Zero means no. If someone picks the lowest score on this scale, then they rarely recommend a firm.
  • One to five is a neutral zone. Consumers that choose the next five higher responses have about the same frequency of recommending, between 18% and 29%.
  • Everything counts from six on. Thirty-two percent of consumers who selected six on the scale actually recommended those companies; the level of actual recommendations ramps up from there for each score higher on the scale
  • NPS labels hide some insight. The NPS process labels people who select “0” to “6” as detractors, “7” or “8” are called passives, and “9” or “10” are promoters. Theses labels may not accurately describe recommendation patterns. For instance, a detractor that selects “0” is quite different than a detractor that selects “6.”
  • Five may be a negative collector. It appears that consumers may be selecting “5” (the midpoint of the scale) when they are relatively upset. It could be that the selection of a “5” out of “10” is considered a failing score for many people (just as a 50 out of 100 points on a test would be seen as failing). This phenomena could explain the drop-off in recommendations at that level. We’ll continue to study this issue since it might require companies to rethink how they examine their survey results.

The bottom line: Not all promoters and detractors are alike.

About Bruce Temkin, CCXP
I'm an experience (XM) management catalyst; helping organizations improve results by engaging the hearts and minds of their employees, customers, and partners. I enjoy researching and speaking about these topics. I lead the Qualtrics XM Institute, which is the world's best job. We're igniting a global community of XM Professionals who are inspired and empowered to radically improve the human experience. To achieve this goal, my team focuses on thought leadership, training, and community building. My work is driven by a set of fundamental beliefs: 1) Everything starts and ends with human beings, so you need to understand how people think, feel, and behave; 2) XM is a discipline that needs to be woven throughout an organization's entire operating fabric; and 3) Building the XM discipline requires a combination of culture, competency, and technology.

7 Responses to Net Promoter Labels Obscure Actual Recommendation Patterns

  1. Sam Klaidman says:

    This is fascinating, 10’s are twice as likely to have recommended as 6’s and just over 3 times as likely as 1’s.

    So, one conclusion to be drawn is that any effort to earn a higher NPS score will likely result in more actual recommendations. This should cause businesses to rethink their strategy about how to take action around the NPS results and not concentrate on any category.

    Thanks so much for this valuable data.

  2. Hi Bruce — Thanks for publishing this. Not seeing the specific questionnaire/research, are you sure that someone rating a 1-5 is really providing a positive recommendation, on the same level as someone who rates you highly? By looking at customer referral sources including “were you recommended,” we’ve found in some of our longitudinal research that the low scoring “recommenders” rarely provide recommendations that convert prospects into customers.

    By the way, we’ve also found that the anchors on the scale — especially the midpoint — matter quite a lot. We’ve published some of this research at

    And thank you again for leadership in customer experience.

  3. Trevor Speirs says:

    Bruce, this is very interesting.
    Can I get some details:
    How did you define “Actual Recommendations” and what was the method of measurement?

  4. Bruce,

    Trying to link actual recommendations to the “would recommend” score is, in my opinion, unnecessary. The value of the Net Promoter approach is not in determining how many recommendations are made but in the personal brand value that you are willing to lend to the company in question.

    I explore this more, while refuting a well known anti-NPS paper, in this blog post:

  5. Hi Bruce. Thank you so much for this. It adds excellent useful data to the whole system. In fact, one of the doubts people seem to have had over the years about NPS was that it seemed too black and white. This is so much more realistic.

    Can I please add something that I think has always been at the heart of making NPS work in companies? I think that the real power behind NPS has always been the 2nd question which is roughly- what 1 thing could be improve that would increase that score most? In all the companies I have heard of and been involved with where they are making NPS work well, it seems that the insight they gain into what to improve- and the fact that they act on it- that gains their customers’ respect and loyalty. Would you agree?


  6. JulieM says:

    One small comment about the group that selects ‘5’: In my experience many of these responses are from people who are unfamiliar with the product or service you are surveying them about OR are brand new to using the product or service. In essence, they truly are neutral when it comes to recommending since they just don’t know enough about it.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: