Mobile App Platform Choice

48 Pages Posted: 3 Feb 2017

See all articles by Yongdong Liu

Yongdong Liu

University College London - School of Management

Date Written: February 3, 2017


Since Apple and Google launched their mobile app stores in 2008, the market for mobile apps has experienced rapid growth. Given the existence of multiple app platforms, fundamental questions in the app industry are how app developers choose which app platform to enter and which market designs benefit the platform expansion.

This paper studies these questions using a unique daily-level panel data set that contains information on every app in the two leading app stores, Apple and Google, over a 2-year period. Combining machine learning techniques for big data problems and computationally efficient econometric approaches, I construct and estimate a structural model for heterogeneous app developers' platform choice decisions within an incomplete-information game framework. I find that in general low-quality apps make the platform less favorable for high-quality entrants. In Google app store, the presence of low-quality apps induces more low-quality apps to enter, while Apple app store exhibits strong competitive effects among high-quality apps. Increasing smartphone user base and improving user engagement could accelerate the platform expansion, but also encourage many low-quality apps to enter. Regulations on low-quality apps and attenuating competition are more effective attracting high-quality apps. Platforms can bundle these policies to achieve better marketing evolution.

Keywords: app market; game estimation; competition strategies; machine learning; big data

JEL Classification: L22; L25; C57

Suggested Citation

Liu, Yongdong, Mobile App Platform Choice (February 3, 2017). Available at SSRN: or

Yongdong Liu (Contact Author)

University College London - School of Management ( email )

Gower Street
London, WC1E 6BT
United Kingdom


Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Abstract Views
PlumX Metrics