Hongjai Rhee and David R. Bell

Successful retailing depends heavily on the ability to locate, maintain and ultimately grow a profitable pool of customers. Fundamentally, some kinds of customers will hold more promise than others and it is critical to focus on the right types of customers, and having identified them, treat them in a way that builds loyalty. In recent years, retailers have become equipped with data that better enables the identification of different groups of shoppers and also the patterns of behavior of individual shoppers. The centrality of the "customer as an asset" concept, and the availability of data have led to loyalty programs and other initiatives designed to deepen the relationship between individual shoppers and their preferred retail outlets. While the results of these efforts have been somewhat mixed, most such initiatives are based on a common implicit assumption: Under the right conditions, shoppers will switch stores.

We believe that this implicit assumption requires more detailed examination, and needs to be viewed in new ways. For example, while it is well accepted that price promotions on certain products may induce shoppers to switch stores, and that some segments (e.g., cherry pickers) will actively look for differences across stores, less is known about the longevity of the switch. In our view, this question needs to be answered if retailers are to effectively manage one of their most important assets – the customer base. Unfortunately, most research studies consider shopper choices as the dependent variable and therefore seek to explain or predict choices as a function of store and shopper characteristics. We believe that this is more likely to encourage retailers to take decisions that have very short-lived effects. Our approach is somewhat different, and more appropriate if one wishes to test the implicit assumption of "mobility" and then develop effective retail strategies. We view a sequence of choices as indicative of an overall relationship with a store. Moreover, we consider that a "switch" takes place only when a shopper moves the vast majority of the share of wallet from one store to another.

We study the shopping behavior of over 500 households shopping over a two-year period among five competing grocery retailers. Using our more stringent definition of a switch – the shopper must move the major share of wallet from one store to another – we find surprisingly little evidence of mobility. It is important to note that this lack of mobility does not arise from lack of alternatives – most shoppers have up to five alternatives within a 2-3 mile radius. On the other hand, there is a wide variation in the behavior of the sample. About 17% of the shoppers change their major allegiance very frequently (almost every other week). Moreover, when changes take place, shoppers are likely to turn over to stores that operate the same basic format (e.g., shoppers will move around among competing EDLP stores but will not cross-shop alternative formats).

Temporary marketing activities (e.g., price specials) don't appear to induce major changes in allegiance and shopper demographic characteristics are largely unrelated to a shopper's level of mobility. All in all, most shoppers show a perhaps surprising preference for the status quo and this is reinforced over time. The longer a customer devotes the majority of her wallet to a store, the more likely she is to continue to do so. In fact, almost three quarters of our sample show this pattern of behavior.

Managerially, our results suggest a clear approach to prioritizing customers. The highest priority customers should be those who have already demonstrated loyalty to the outlet. Next, management should try to focus on newcomers before they "settle in" and third, turn to customers of competitor stores – but only stores that offer a similar overall format. Store traffic alone is unlikely to be a useful metric to manage to. Instead, retailers should focus on increasing the wallet share of the existing customer base. This is because the shopping behavior profile of a particular customer has a strong relationship with the level of mobility. Specifically, customers who visit stores less frequently and purchase larger baskets per trip are less likely to move their affiliation to a competitor. When the "basket size" of a customer is increased, the retailer benefits from a virtuous circle – these customers are the most stable and attractive customers, and once they start buying larger baskets they increasingly favor the status quo. All this suggests that more attention be paid to in-store merchandizing, assortment, category-specific loyalty programs and other initiatives specifically aimed at increasing the transaction value per visit. While the selection of the right activities to achieve this is in itself a non-trivial task, our research suggests this is certainly the correct focus if the goal is to develop a stable and healthy customer franchise.