Demetrios Vakratsas and Frank Bass
Previous research on buying behavior for frequently purchased goods has identified two consumer segments that differ substantially with respect to their purchasing pattern: regular and random buyers. Regular buyers make category purchases in relatively fixed purchase intervals producing a highly stable purchasing pattern. Random buyers on the other hand seem to make category purchases in rather irregular intervals.
Although many studies have focused on relating purchase regularity with demographic and shopping characteristics, no effort has been made so far to characterize random and regular households in terms of their marketing mix sensitivities. Such an effort would grant segmentation schemes based on purchase regularity a practical dimension since it would relate retailer marketing mix activities (mainly pricing and promotional events) to the response of random and regular buyers. For example one may conjecture that the irregularity in the purchasing pattern of random households implies a greater flexibility in their purchasing plans and hence suggest that they are more able to take advantage of lower prices and promotions. This would further suggest that random buyers are important drivers of store sales during promotional periods.
In this article we investigate the response of random and regular buyers to retailer marketing mix activities in four product categories (bathroom tissue, margarine, ketchup and sugar). More specifically we examine whether it is the regular or the random purchasing households that exhibit a higher a propensity to accelerate their purchases due to lower prices or the presence of promotions. This investigation is carried out through the use of a fairly parsimonious hazard rate model. This model allows us to characterize simultaneously households with respect to their purchase regularity and their propensity to accelerate while recognizing that they have different (heterogeneous) purchase rates. The four product categories we study differ in terms of the frequency at which they are purchased. Bathroom tissue and margarine are purchased on a regular basis (more frequently) whereas ketchup and sugar are purchased occasionally (less frequently).
The application of the proposed model to purchase timing data on the four product categories showed that in the frequently purchased categories of bathroom tissue and margarine, random buyers do not exhibit any propensity to accelerate while regular buyers do. In the occasionally purchased categories of ketchup and sugar on the other hand, random buyers exhibit at least as much propensity to accelerate their purchases as regular buyers do.
We explain these results by developing arguments regarding the information available by the regular buyers who purchase consistently (in all four product categories) more frequently than random buyers, i.e., they are "heavier" buyers. Specifically we argue that regular buyers, being frequent users of all four categories, are more likely than random buyers to possess more information on prices and should thus be able to respond more efficiently to prices and promotions. Since bathroom tissue and margarine are bought at a much higher frequency than the other two categories they have more at stake in those two categories. They therefore choose to exhibit propensity to accelerate in the frequently purchased categories of bathroom tissue and margarine. Random buyers on the other hand, being less frequent users of all four categories, are less informed about prices. Thus for product categories purchased on a regular basis (bathroom tissue, margarine) where demand has to be timely satisfied and consumers cannot "skip" purchases, random buyers, with more limited information than regular buyers, may not be able to respond efficiently to prices and promotions. In the occasionally purchased categories on the other hand, where demand is more flexible, random buyers take advantage of price cuts and promotions to "buy into" such categories exhibiting therefore opportunistic purchasing behavior.
Regular and random buyers, therefore, do not respond to prices and promotions uniformly across product categories. Furthermore the flexibility implied by the random buyers purchasing pattern does not fully account for their response to prices and promotions as they appear to lack information due to their lower purchase frequency. Both these observations suggest that a segmentation based on purchase regularity can have substantial implications for the retailer as the purchasing behavior of consumers, especially their response to retailer marketing mix activities, depends both on their purchasing pattern and the frequency of a category's demand. The retailer should take such differences in the behavior of regular and random buyers into account when coordinating promotions among different product categories, as the segment (regular or random buyers) that drives sales during promotional events is expected to be different for each category. Extensive replication of the study to different and larger sets of product categories will only strengthen such implications.