A computationally scalable, statistically efficient aggregate-disaggregate data fusion method that corrects for selection bias is applied to model customer relationship dynamics at a subscription-based firm.
In 1956, a group of trade associations representing publishers and independent advertising agencies signed a consent decree aimed at ending a set of trade practices that for half a century effectively precluded advertisers from owning and operating in-house agencies. Since then, large firms have internalized more and more of the services formerly performed by external agencies, perhaps as many as half. We use this phenomenon to test a theory of the firm, thereby simultaneously offering an explanation for it. The theory suggests that firms should internalize activities for which their competitive position implies (1) that it is more important for human capital to be firm specific as opposed to function specific and (2) that frequent modifications are desirable. It also predicts (3) that these two effects reinforce each other. This is the first paper to report on a test of the specialization hypothesis, and we find that it is robustly significant in a cross-sectional data set covering nine different agency activities in 79 firms. In addition to the cross-sectional test, we informally present some time-series data suggesting that both specialization and frequency have grown over time along with the level of internalization.
This research aims to demonstrate that the abundant marketing data that companies are using to explore new business opportunities can be an equally fertile source for uncovering an undesirable social attitude or behavior that may be relevant to firms’ business. Companies may benefit from this knowledge when developing innovative new programs that aim to benefit society, such as corporate social responsibility initiatives. In this study, we examine boy-girl gender discrimination in China as manifested in parents’ purchase decisions on behalf of their children across different markets. Our study in itself is significant, because it is the first large-scale empirical work to clearly verify the phenomenon of boy-girl discrimination, taking advantage of e-commerce marketing data. Specifically, we compare the clothing expenditures on boys versus girls using a rich, household-specific data set obtained from two online retailers. We find that the patterns of gender inequality vary systematically across different geographic markets, as the relative expenditure difference on boys versus on girls is bigger in less developed areas as compared with metropolitan areas, and this relative expenditure difference is closely tied with socioeconomic conditions, education levels, and birth rates of a district. Managerial and social implications are discussed.
We propose an interpretable model that combines the simplicity of matrix factorization with the flexibility of neural networks to model evolving user interests by efficiently extracting nonlinear patterns from massive text data collections.
In modern retail contexts, retailers sell products from vast product assortments to a large and heterogeneous customer base. Understanding purchase behavior in such a context is very important. Standard models cannot be used because of the high dimensionality of the data. We propose a new model that creates an efficient dimension reduction through the idea of purchase motivations. We only require customer-level purchase history data, which is ubiquitous in modern retailing. The model handles large-scale data and even works in settings with shopping trips consisting of few purchases. Essential features of our model are that it accounts for the product, customer, and time dimensions present in purchase history data; relates the relevance of motivations to customer- and shopping-trip characteristics; captures interdependencies between motivations; and achieves superior predictive performance. Estimation results from this comprehensive model provide deep insights into purchase behavior. Such insights can be used by managers to create more intuitive, better informed, and more effective marketing actions. As scalability of the model is essential for practical applicability, we develop a fast, custom-made inference algorithm based on variational inference. We illustrate the model using purchase history data from a Fortune 500 retailer involving more than 4,000 unique products.
Managers frequently explore new strategies, and exploit familiar ones, when making decisions on new product development, pricing, or advertising. Exploring for too long, or exploiting too soon, will generate inferior financial returns. Our research describes decision makers’ exploration/exploitation trade-offs and their link to psychometric traits. We conduct an incentive-aligned study in which subjects play a multiarmed bandit experiment and evaluate how subjects balance exploration and exploitation, linked to psychometric traits. To formally describe exploration/exploitation trade-offs, we develop a behavioral model that captures latent dynamics in learning behavior. Subjects transition between three unobserved states—exploration, exploitation, and inertia—updating their beliefs about expected payoffs. Our analysis suggests that decision makers overexplore low-performing options, forgoing over 30% of potential revenue. They heavily rely on recent experiences. Risk-averse decision makers spend more time exploring. Maximizers are more sensitive to payoffs than satisficers. Our research builds the groundwork needed to devise remedial actions aimed at helping managers find an optimal balance between exploration and exploitation. One way to achieve this goal is by carefully designing the learning environment. In two additional studies, we analyze the evolution of exploration/exploitation trade-offs across different learning environments. Offering decision makers repeated opportunities to learn and increasing the planning horizon appears beneficial.
This paper analyzes the build-test-learn cycle of the lean start-up method (LSM) using an analytical model to understand its microfoundation and how best to implement it.
This study empirically examines the effect of copyright piracy reduction on writers’ creative productivity and customer care efforts using data from a digital publishing firm.