Anne W. Magi

For retailers, increasing the share of wallet current customers spend in their store is increasingly vital in today's competitive economy. This study explores the effects of two potential avenues for increasing customer share: customer satisfaction and loyalty programs. It also includes both direct and indirect effects of a number of shopper characteristics that have been identified by previous research as potential predictors of customer share.

Achieving and maintaining a high level of customer satisfaction has become an important goal for companies across all industries. This development has likely been propelled by academic research presenting evidence of a strong link between satisfaction and repeat purchase intentions. However, studies linking satisfaction to actual behavior provide mixed results, indicating that satisfaction does not always lead to a high degree of behavioral loyalty. Recent studies also suggest that the effects of customer satisfaction are moderated by such consumer characteristics as age and gender. This implies that the effects of customer satisfaction on behavior are not equal across consumer groups, which should have important consequences for how satisfaction programs are executed.

As for loyalty programs, empirical evidence of their efficacy remains scarce. In fact, it has been suggested that their effects might be limited due to the difficulties of changing habitual consumer behavior patterns. It is also possible that loyalty card programs are only effective in some consumer segments. That is, similar to the effects of satisfaction, the effects of loyalty card programs may be moderated by consumer characteristics.

While the dominant view today seems to be that customer share is something a manager can influence, the store patronage literature suggests that individual difference variables, such as a bargain-hunting propensity or being an apathetic consumer, affect how consumers distribute purchases across stores in a way largely unrelated to the market-ing actions and performance of a store. It is also plausible that these variables would moderate the effects of satisfaction or loyalty programs on behavior. Knowledge about both indirect and direct effects of shopper characteristics on customer share is essential because it needs to be considered when designing and evaluating actions taken to increase share.

To investigate these issues, analyses were carried out on customer share data derived from a store choice diary of grocery purchases. Additional data was collected through a follow-up questionnaire. The analyses focused on the customer share, both in terms of spending and number of visits, of the household's primary store.

The results show that although satisfaction does have a modest effect on primary store customer share, as predicted, this effect is moderated by shopper characteristics. Specifically, satisfaction is clearly a more important determinant of customer share for consumers with a low economic orientation than for consumers with a high economic orientation. The shopper's personalizing shopping orientation also moderates the satisfaction – share relationship. Shoppers who value the social interaction with store personnel were less sensitive to decreases in satisfaction. This suggests that their relationship with store personnel insulates the effects of overall dissatisfaction with the store.

For managers, these results indicate that it is important to incorporate shopper characteristics in surveys of customer satisfaction and behavioral outcomes. Moderating analyses such as those presented in this paper can identify which consumer segments are more, or less, likely to respond to increases in satisfaction by shopping more in their primary store. Based on such analyses, measures can be targeted at enhancing the satisfaction of customers who are most likely to change their behavior.

The analysis regarding the effects of loyalty programs provides only mixed support for a positive relationship between loyalty cards on customer share. On the chain level, program enrollment is associated with a higher customer share for the focal chain, but only for customers who have enrolled exclusively in the focal chain's program. Given the large number of multiple-card holders – in this study 49 percent of respondents had at least two of the available cards on the market – these findings clearly indicate the risk for competing loyalty programs to cancel each other out. This suggests that retailers must take great care when evaluating their loyalty programs, and if they can be identified, use different customer relationship strategies for single-card holders and multiple-card holders. Future studies on why some consumers are more prone than others to enroll in more than one loyalty program will also help managers decide on how loyalty programs should be designed to primarily attract customers who are more likely to become active cardholders.