In a non-contractual setting, cross-buying is measured as the total number of different product categories that a customer has purchased from a firm from the time of his/her first purchase. Current business trends and past academic research clearly demonstrate the importance of cross-selling--the specific marketing effort by the firm to increase cross-buying--in a retailing context. However, critical questions that warrant answers based on empirical evidence include: (1) why do customers cross-buy from the same firm? who are likely to cross-buy? (2) what product category needs to be promoted? (3) when is the best time to cross-promote a product category? and (4) how much should a firm cross-sell? Or, what is the optimal level of cross-promotion? The two essays presented below address these questions and present answers based on empirical evidence.
The purpose of the first essay titled "Why Cross-buy?" is to understand the motivation of customers to cross-buy, and to identify the key drivers of cross-buy--exchange characteristics, customer characteristics, product characteristics, and the firm's marketing efforts. In other words, the purpose is to identify the characteristics of customers who are likely to cross-buy ( who ). Further, we empirically validate the positive impact of cross-buy on customer-based outcome metrics such as revenue/contribution margin per order, and the number of orders in a given period.
In the second essay titled "What, when, and how much to cross-sell? Optimizing Multi-category Catalog Mailing," we answer the remaining questions-- what, when and how much to cross-sell. We address an existing research gap--lack of models to optimize multi-category mailing--by introducing a multivariate proportional hazard model employed in a Hierarchical Bayesian framework, to jointly estimate probability of purchase and purchase amounts in multiple product categories. In other words, the model integrates when and what components of a customer's purchase decision into how much component of a firm's cross-selling strategy using Genetic Algorithm based optimization. The optimal catalog mailing policy helps to achieve 50% more CLV from top 15% of the households.
The results of the essays have several implications for both practitioners and academics. While a key managerial implication is to use cross selling as a strategic tool to maximize Customer Lifetime Value (CLV), the academic contributions relate to applying a multivariate proportional hazard model to a multi-category catalog retailing context.
