Understanding consumer usage learning mechanisms in the retail industry.

摘要

Consumers have long been subjected to promotional activities such as price cuts, features and in-store displays offered by retailers. This long term exposure to retailer promotional activities has given consumers an opportunity to understand the promotional strategies of retailers and has enabled them to time their purchases to coincide with the retailer's promotions. In other words, consumers engage in a learning process to effectively gauge the promotional mechanisms of the retailer. Furthermore, this inherent learning in consumers may not be restricted to promotional strategies alone. Every time a consumer purchases a product at a particular price from a particular retailer, he/she learns about the quality, the attributes and the admissible prices for that product. Thus consumer learning extends to categories, brands within a category, products within each of the brands and also the different attributes of interest.;Consumer usage learning mechanism can be addressed using econometric models incorporating learning through Bayesian updation of the parameter for which learning occurs. The dynamic econometric learning models offer a useful tool for addressing such behaviors since the specification itself accounts for updating the parameter after each occasion of learning involved i.e., the parameter value is bound to change after each learning occasion as warranted in consumer learning scenario. In this thesis, we use the dynamic learning models in addressing consumer usage learning mechanisms in the retail scenario. We also study the mechanism involved when consumers learn about a particular product/category attribute, the differential learning prevalent in categories and finally, the impact and spillover of such learning across categories.;Retailers use promotional strategies, such as price promotions feature and display, extensively, in an attempt to increase store traffic, expand basket sales, and improve store sales and profits. Consumers often are able to understand and time their purchases to their benefit thus moderating the retailer's objective in offering promotions. We first use dynamic leaning models to understand the impact of the various promotional strategies on consumers purchase behavior. More specifically, we expand the literature on price promotions by adopting the Bayesian updating framework to understand the effects that price promotions can have on the price sensitivity of consumers. Such a study helps (a) disentangle the effect of ever changing price sensitivities on consumer brand choice (b) identify the differential effect of promotion depth on price sensitivities across categories (c) identify the relationship between the effects observed and category characteristics, and (d) help retailers reassess their promotional strategies by determining the optimal promotional depth for different categories.;Retailers also put forth new products or products with new attributes to draw the consumer's attention. Through repeated purchases of these products, consumers learn about the new products or the new attributes that they have encountered. We use the learning model to address the understanding of consumer learning through repeat purchases. We strive to understand the preference evolution of relatively newer attributes that transcend across categories, using learning models. To this end we study the learning in consumers for the organic attribute starting from their first purchase incidence of an organic product in a category and also study how the learning for organics is different across different categories. The growth of organic products has been phenomenal in recent years. 'Organics' have fast become an important attribute across a wide variety of categories. And for the same reason the preference of this attribute in one category might be influential in the preference for an organic product in other categories as well. We supplement this study by addressing the role played by consumer psychographic variables in the consumer learning process for organics.;It is evident that the learning for an attribute in one category is bound to impact the perceptions of the same attribute in another category as well. Hence in consumer usage learning for organics the learning about perceptions is bound to have spillover effects. It then becomes important for retailers to study how the spillover occurs in order to identify the dominant categories influencing organic preference evolution in a basket. We employ the learning mechanism to formulate a framework that investigates the spill-over effects of organic preference across categories as well. That is, we further study the evolution of organic preferences by addressing the spillover of preference for organics across categories. The results of this study would help retailers understand the strong and weak impacts for organic preference evolution among categories purchased. It would also help retailers disentangle the effects of carryover across categories, as well as marketing mix variables. Finally, it would enable retailers to derive strategies for distinguishing the categories which need to be promoted together and the ones which require individual strategies in terms of organic sales promotion. (Abstract shortened by UMI.)