How Is Mobile User Behavior Different? A Hidden Markov Model of Cross-Mobile Application Usage Dynamics

Mobile application use has become an essential activity in many people’s daily lives in the mobile Internet era. Driven by its ubiquity and strong context dependence, Internet companies are in a race of cross-industry expansion to build a seamless ecosystem incorporating various contexts. Amid such trends, a better understanding of cross-app uses and the impact of contexts becomes critical and imperative. Yet, research on cross-app uses in information systems and marketing is scarce. In this paper, we aim to fill this gap. We develop a hidden Markov model to study cross-app uses (choice and duration), capturing their interdependence and the impacts of contextual factors. We calibrate it using a consumer panel that contains real-time app use information. Our key findings are as follows. (1) In addition to the utilitarian and hedonic states behind consumer decisions identified in prior literature, we uncover a novel social state behind mobile user behavior. App use behavior exhibits large differences across three states. (2) Within-state app interdependence is strongest in the hedonic state, followed by the social and utilitarian states. (3) Social state is the most transient (i.e., mostly likely to switch away), followed by the hedonic and utilitarian states. (4) Contextual factors—in particular, location and time of day—influence the state dynamic. Compared with the intrinsic state transition, being at home or on the way (versus office) and in the morning or evening (versus night) leads to higher volatility. We discuss the managerial implications for a variety of mobile digital strategies.