Analyzing Platforms Goods Using Multiple-Discrete Continuous Demand Models
47 Pages Posted: 28 Mar 2012 Last revised: 28 Sep 2019
Date Written: March 5, 2013
Platform goods are commonly seen in practice. They occur when a company offers a durable product (platform) and a consumable product (component). Platform goods are desirable because they create local monopolies and increase the switching cost of products. Platform goods have been extensively studied in the marketing literature because they introduce a unique form of complementarity between the platform and the component.
Previous models of platform goods have largely ignored two important aspects of platforms goods. The first aspect, called multi-homing, occurs when consumers purchase multiple platforms. This eliminates many of the benefits of introducing a platform good as the consumer can readily switch between platforms as their needs and wants dictate. The second aspect is brought about when there are multiple, differentiated component products tied to the same platform. This can increase the utility of the platform, but also complicates the marketing decisions. In some cases, competitors introduce components that can disrupt the local monopoly and change the competitive structure. In this paper, we develop a model of platform goods that allows for both multi-homing and multiple component goods. We test this model on a conjoint analysis of single serve coffee machines, a canonical platform goods. The model is compared to a naive model where the the cost of the platform goods is not directly modeled. The model we propose provides better fit and allows for exploration of a number of counter-factual situations. We then discuss other potential uses for the model as well as possible extensions.
Keywords: Utility Theory, Bayesian Estimation, Nonlinear Pricing, Tied Goods, Platform Goods, multi-homing
Suggested Citation: Suggested Citation