Satisfaction, NPS, Share of Wallet and other Holy Grails

In the October 2011 issue of the Harvard Business Review, authors Timothy L. Keiningham, Lerzan Aksoy, Alexander Buoye, and Bruce Cooil put forth the case that managing customer loyalty and satisfaction isn’t enough to grow revenue. They view metrics such as customer satisfaction scores and Net Promoter Scores as effective means of measuring the business customers currently do with you but are of little use in growing revenue with existing customers – just because customers are satisfied and would recommend you does not mean that they are not giving money to competitors. They believe many customer satisfaction and loyalty efforts simply make happier customers happier. They advocate using a brand preference metric as the most effective means of growing revenue.

The authors make many valid points regarding the limitations of customer satisfaction and NPS and regarding the value of using a preferred brand metric. However their pitch for a different set of metrics sounds remarkably similar to the case made for the Net Promoter Score by Reichheld in The Ultimate Question: Driving Good Profits and True Growth. It also shares the pitfalls associated with the recent enthusiasm for NPS; that is the idea that a company’s strategy can be boiled down to a single metric. In all due fairness to Riechheld, he presents NPS as a means of integrating customer feedback into a continuous feedback loop and points out that the process is more important than the metric. Unfortunately just as the NPS is about boiling down many factors into a single metric, many managers simply take away the idea of the one metric they need to manage their business. If a brand preference metrics gains momentum in the marketplace it is likely to follow the same path.

At Isurus we believe that the best approach is to keep tabs on many facets of the relationships you have with customers and prospects and use these metrics to help inform business decisions rather than on managing around a set of high level metrics.