The Net Promoter Score: One Number is Not Enough
If you have ever been on a dairy farm around milking time you know all about the herd instinct: once one cow heads for the milking shed the entire herd follows. So it might also seem for the Net Promoter Score (NPS) and the marketing research industry in Australia. In the past 24 months an increasing proportion seem to have adopted the exceptionally accessible NPS metric.
In many cases it has been C-level management who have been ‘sold’ the metric leaving research managers somewhat powerless to properly investigate the veracity of the construct. Indeed, some would say the standout feature of the NPS has been the frightening preparedness of management to adopt a proposition untested by scientific, independent research.
Central to the NPS ‘solution’ is that organisations need only measure one metric in order to predict their future prosperity. The chief proponent of the NPS is Frederick Reichheld, from management consulting firm Bain & Co. Reichheld, who is a strident critic of marketing research, uses examples of the worst possible application of customer satisfaction as a means for repositioning his own ‘solution.’ For example:
‘As surveys grow to 30 or 40 questions or more, the cost per survey creeps up, response rates drop, and sample size shrinks. That introduces sample bias and makes scores volatile and unreliable. But the real problem is that the mountain of customer feedback generated becomes wholly unmanageable without help from the highest-tech statistical programs. These black-box software packages churn out complex analyses that are only intelligible to an elite breed of PhDs. The PhDs interpret the findings for senior executives who in turn pass along their own perspectives to lower level managers. By the time the feedback reaches the front line it is useless.’
Fact over fiction
Against the tidal wave of seemingly unquestioning support for the NPS stood a handful of cautionary voices; two of these voices were published in the marketing research industry magazine, Research News (see December 2006, ‘Value speaks louder than words’ Ken Roberts and February 2007 and ‘Net promoter score: a cautionary tale’ Dr Chris Crook). In July 2007, the first peer-reviewed, independent scientific investigation of the NPS was published. The study failed to replicate or indeed support the assertions regarding the clear superiority of the net promoter score relative to customer satisfaction. Amongst other things, the study found:
- Claims of the NPS' superiority in predicting firm growth or in predicting customers’ future loyalty behaviours were false;
- Without question, ‘Would you recommend us to a friend or colleague?’ is not the ultimate question (as Reichheld claimed) for gauging loyalty;
- Reichheld reported no correlation whatsoever between satisfaction and growth. In two out of three cases tested in the peer reviewed longitudinal study the academics found a stronger correlation between satisfaction and growth than between the NPS and growth.
Advocacy is not a new construct to marketing researchers. Indeed, for many researchers it’s one of a raft of useful constructs the industry has been applying for a decade or more. The NPS innovation is an interesting way to look at the advocacy construct but it is simply not the ‘one measure you need to grow.’
