A truism: Growing companies flourish, shrinking companies die. And customers, of course, are the driving force behind profitable growth.
However, most companies don't manage their relationships with customers well. It's not that they don't care, but that they don't have the information they need to make these relationships more valuable. They lack good customer databases -- ones that link customer survey information to customer behavior information. And in a world in which managers look to analytics to clarify their decisions, this presents challenges. How are managers supposed to measure how customers really feel and what they're most likely to do?
In recent years, various customer metrics have emerged that supposedly illustrate the connections between customer behavior and growth. But they've generated more smoke than heat. Even the best metrics have shown only modest correlations to growth, and none have proven to be universally effective. The weakness of metrics hasn't discouraged companies from adopting new ones, however.
Initially, customer metrics focused on attempts to understand why people buy. The logic was straightforward: If customers preferred a company, managers assumed the customers were more likely to become loyal. In general, metrics tended to boil down to ideas of preference: What did the customers prefer, in terms of perceived quality or value?
Today's most popular metric, the Net Promoter Score, doesn't focus on quality, satisfaction or value, but on how customer word of mouth -- both negative and positive -- can advance growth. It posits that word of mouth can be managed to create a buzz about a brand or product, and that by keeping close track of NPS, companies can learn to be more successful.
The approach works like this: Companies classify customers according to their response to the question, “How likely is it that you would recommend this company to a friend or colleague?” Those who have a high likelihood (9 or 10 on scale of 0-10) are classified as “promoters”; those who rate the likelihood as low (6 or below) are “detractors.” The score is calculated based on the difference: the percentage of promoters minus the percentage of detractors.
A team of researchers recently set out to test claims about NPS. Does NPS really help companies predict customers' future loyalty behaviors? Does it link to growth? Is it better than other commonly used metrics?
The researchers conducted a two-year study of more than 8,000 customers of companies in retail banking, mass merchant retail and Internet services, monitoring individual customer ratings on common satisfaction and loyalty metrics. In the study's second year, they also tracked behaviors associated with customer loyalty.
The links between every customer perception metric and subsequent customer behavior were modest at best. NPS's ability to explain customer behavior ranged from 0% to 20%, which means that something other than what the researchers were measuring caused 80% or more of the differences in customer loyalty behaviors. Furthermore, no metric was well correlated to all the behaviors associated with customer loyalty.
These findings point out an inherent problem: The behaviors associated with customer loyalty are, in fact, varied and distinct from one another. People are motivated by different things. Market factors influence their willingness to buy and be loyal in different ways. Most importantly, the coveted behaviors -- increased spending, retention and positive word of mouth -- contribute differently to companies' successes. The fact that people continue to be customers doesn't mean they will increase (or decrease) how much money they spend on a company's products. And both these behaviors are different from someone recommending a company or product to a friend.
In essence, there is no silver bullet -- nor will one be forthcoming. Many dimensions affect customer loyalty and how it impacts customer behavior and profitability. When companies seek out simple solutions to complex problems, they may come up with answers that are technically correct in a narrow sense -- but overall are substantially wrong.
So what should managers do? First, they must be realistic. They need to accept that metrics are tools that can assist them in making decisions, but understand that metrics don't make decisions -- they are only guidelines. Second, they should be willing to do the level of analysis required to understand their customers and market opportunities. Just as there are significant risks in doing too little -- often pinning your hopes on one number that promises too much -- there are risks in doing too much. Managers must balance their desire for simplicity with their need for accuracy. This is tough. But if it were easy, the debate over which metrics pack the most punch would be a lot less spirited.
For more information on this topic is available at http://sloanreview.mit.edu/smr/issue/2008/summer/14/