A data-driven approach to improve online consumer subscriptions by combining data visualization and machine learning methods

Abstract

Effective online consumer research helps companies on defining a successful strategy to increase user loyalty and shape brand engagement. Digital innovation introduced a dramatic change in businesses, particularly in the online news industry. Content consumers have a wide offer across different channels which increases the digital challenge for online news media companies to retain their readers and convert them into online subscribers. Furthermore, digital news publishers often strive to balance revenue sources in online business models. Thus, this study fills a gap in the literature on media consumer research by proposing a data-driven approach that combines two machine learning models to allow managers dynamically improve their marketing and editorial strategies. Firstly, the authors present an online user profiling to identify consumer segments based on the interplay between several engagement’ variables substantiated in the literature research. Second, as few studies have explored the factors influencing users’ intention to pay for such services, the XGBoost machine learning algorithm identifies the predictors of consumer's willingness to pay. Third, a dashboard presents the key performance indicators across the audience funnel. Thus, practical implications and business suggestions are presented in a two-fold strategy to maximize revenue from digital subscriptions and advertising. Findings provide new insights into an engagement approach and the relation to acquire a digital subscription in online content platforms. We believe that the provided recommendations are potentially useful to help marketing and editorial teams to manage their customer engagement process across the funnel in a more efficient way.