A firestorm has been brewing ever since Fred Reichheld claimed in his 2003 Harvard Business Review (“The One Number You Need to Grow”) article that the simple Net Promoter Score (NPS) measure of consumer recommendations was a good proxy of customer loyalty and an accurate predictor of business growth.  The publication of the HBR article was followed up with his bestselling book The Ultimate Question.  Many of the largest companies including GE, American Express, T-Mobile, Microsoft and Philips adopted the measure and in many cases, changed the way service is delivered to even tying employee and/or executive compensation to NPS scores.   The beauty of this measure is in its parsimony, consists of one simple question: “How likely is it that you would recommend us to your friends or colleagues?” Calculating NPS scoring is based on a 0-10-point “likely to recommend” scale ranging from “highly unlikely” to “highly likely.”  Those who score between 0 and 6 are considered “detractors”, those who score between 7 and 8 are “passives” and those scoring 9 or above are “promoters”.  The eventual NPS is then calculated by subtracting the percentage of detractors from the percentage of promoters.  For example, if 20% of Company X’s customers are detractors, and 60% are promoters, then Company X has scored an NPS of 40.

Companies have gravitated to NPS due to its straightforward approach to assessing loyalty and providing a clear measure of an organization’s performance through its customers’ eyes.  Further, the growing importance of word-of-mouth communications in driving future growth has made NPS more attractive.   Do higher NPS scores make a difference?   At American Express, for a promoter who is positive, the company sees a 10-15 percent increase in spending, and 4-5 times increased retention—both which drive shareholder value.   Research shows that sustained value creators—companies that achieve long-term profitable growth—have Net Promoter Scores (NPS) two times higher than the average company.  Further, NPS leaders outgrow their competitors in most industries—by an average of 2.5 times.

Yet, NPS has sparked considerable debate during the past 10 years as to the efficacy of the “ultimate question”, where a number of researchers have examined the construct’s validity and now the verdict on NPS is not quite as strong as when Reichheld first introduced the word-of-mouth metric back in 2003.   For example, in an excellent study by Keiningham, et. al (2007), their study’s results undermine a key supposition of NPS, i.e. that it is the single most reliable indicator of company growth.   Their findings, which contradict this supposition, have important implications for managers that have adopted the Net Promoter metric for tracking growth and have consequences as to the potential misallocation of resources.  Further, Mark Molenaar of TNS Research Surveys thinks that the score is too simple, too narrow and no better than other measures of satisfaction or advocacy.

This brings up an ongoing debate when it comes to measurement: “Are multi-item (MI) scales preferred over single-item (SI) scales such as NPS?”  According to conventional measurement theory, the (reflective) items comprising multi-item (MI) measure of a focal construct represent a random selection from the hypothetical domain of all possible indicators of the construct.  Using multiple items helps to average out errors and specificities that are inherent in single items, thus leading to increased reliability and construct validity.  Single-item (SI) measures seem to be a viable option in exploratory research situations where typically weaker effect sizes are expected and smaller samples are used.  Rossiter wrote that “when an attribute is judged to be concrete, there is no need to use more than a single item.”   Many scale development experts recommend following conventional wisdom and use MI scales when conducting survey research.  Back to NPS, Morgan and Rego suggest that if the NPS score is used, it should be supplemented with additional questions which would allow companies to further understand their customers and their reasons for recommending to friends and family.   Bain and Company suggests that research conducted using the Ultimate Question should be followed up with an open-ended question: “Why?”

*NPS Chart: www.davidmitz.com; by David Mitzenmacher, 2011.

William (Bill) Johnson, Ph.D., is a retired Professor of Marketing and Adjunct Professor in Marketing in the H. Wayne Huizenga School of Business and Entrepreneurship, Nova Southeastern University. He can be reached at billyboy@nova.edu