The Benefits of Segmentation in Trial-Offer Markets with Social Influence and Position Bias

The purchasing behaviour of consumers is often influenced by numerous factors, including the visibility of the products and the influence of other customers through their own purchases or their recommendations. Motivated by trial-offer and freemium markets and a number of online markets for cultural products, leisure services, and retail, this paper studies the dynamics of a marketplace ran by a single firm and which is visited by heterogeneous consumers whose choice preferences can be modeled using a Mixed Multinomial Logit. In this marketplace, consumers are influenced by past purchases, the inherent appeal of the products, and the visibility of each product. The resulting setting generalizes recent models already verified in cultural markets. We examine various marketing policies for this market and analyze their long-term dynamics and the potential benefits of social influence. In particular, we show that the heterogeneity of the customers complicates market optimization significantly: Many of the optimality and computational properties of the corresponding homogeneous market no longer hold. To remedy these limitations, we explore a market segmentation strategy and quantify its benefits. The theoretical results are complemented by Monte Carlo simulations conducted on examples of interest.