An empirical study of content-based recommendation systems in mobile app markets

Abstract

Recommendation systems are widely used to promote product visibility and sales. However, past research suggests that they primarily benefit market superstars and therefore, can be detrimental to niche products. This is especially critical in superstar dominated markets such as the market of mobile apps. This study incorporates social network analysis and econometric models to empirically examine the impact of content-based filtering (CBF) recommendation systems on the distribution of demand in mobile app markets. The analysis results of two comprehensive panel datasets from App Store and Google Play suggest that CBF recommendations favor niche items and effectively boost the long tail of the market. Moreover, consistent with signaling theory, relative quality signals provided by recommendation systems can impact consumer decision making process, leading to a spillover of demand. Our findings provide important implications for developers and market operators to better promote their products in the highly competitive market of mobile apps.