Social Learning and Diffusion of Pervasive Goods: An Empirical Study of an African App Store

📝 Abstract

In this study, the authors develop a structural model that combines a macro diffusion model with a micro choice model to control for the effect of social influence on the mobile app choices of customers over app stores. Social influence refers to the density of adopters within the proximity of other customers. Using a large data set from an African app store and Bayesian estimation methods, the authors quantify the effect of social influence and investigate the impact of ignoring this process in estimating customer choices. The findings show that customer choices in the app store are explained better by offline than online density of adopters and that ignoring social influence in estimations results in biased estimates. Furthermore, the findings show that the mobile app adoption process is similar to adoption of music CDs, among all other classic economy goods. A counterfactual analysis shows that the app store can increase its revenue by 13.6% through a viral marketing policy (e.g., a sharing with friends and family button).

💡 Analysis

In this study, the authors develop a structural model that combines a macro diffusion model with a micro choice model to control for the effect of social influence on the mobile app choices of customers over app stores. Social influence refers to the density of adopters within the proximity of other customers. Using a large data set from an African app store and Bayesian estimation methods, the authors quantify the effect of social influence and investigate the impact of ignoring this process in estimating customer choices. The findings show that customer choices in the app store are explained better by offline than online density of adopters and that ignoring social influence in estimations results in biased estimates. Furthermore, the findings show that the mobile app adoption process is similar to adoption of music CDs, among all other classic economy goods. A counterfactual analysis shows that the app store can increase its revenue by 13.6% through a viral marketing policy (e.g., a sharing with friends and family button).

📄 Content

1 Social Learning and Diffusion of Pervasive Goods: An Empirical Study of an African App Store

Meisam Hejazi Nia

Brian T. Ratchford

Norris Bruce

*** Please do not cite, quote, or distribute without permission ***

Meisam Hejazi Nia is a senior data scientist at Staples (meisam.hejazynia@gmail.com), Brian T. Ratchford (972- 883-5975, btr051000@utdallas.edu) is Charles and Nancy Davidson Professor in Marketing, and Norris I. Bruce (972.883.6293, nxb018100@utdallas.edu) is an Associate Professor at the Naveen Jindal School of Management at the University of Texas at Dallas, P.O. Box 830688, Richardson, TX 75083. 2 Social Learning and Diffusion of Pervasive Goods: An Empirical Study of an African App Store

ABSTRACT

In this study, the authors develop a structural model that combines a macro diffusion model with a micro choice model to control for the effect of social influence on the mobile app choices of customers over app stores. Social influence refers to the density of adopters within the proximity of other customers. Using a large data set from an African app store and Bayesian estimation methods, the authors quantify the effect of social influence and investigate the impact of ignoring this process in estimating customer choices. The findings show that customer choices in the app store are explained better by offline than online density of adopters and that ignoring social influence in estimations results in biased estimates. Furthermore, the findings show that the mobile app adoption process is similar to adoption of music CDs, among all other classic economy goods. A counterfactual analysis shows that the app store can increase its revenue by 13.6% through a viral marketing policy (e.g., a sharing with friends and family button).

Keywords: mobile app store, social learning, state-space model, structural model, semi- parametric Bayesian, MCEM, unscented Kalman filter, hierarchical mixture model, genetic optimization 3 Smartphones pervade the global telecommunications market to such an extent that in the United States, for example, a consumer can adopt a smartphone handset with a postpaid contract, no matter which mobile operator (e.g., T-Mobile, Verizon, AT&T) he or she selects. The smartphone handsets and mobile apps are complements. A mobile app store (e.g., Google play, Apple and Microsoft app stores) acts as a two-sided platform that matches consumers to mobile app publishers/developers. The mobile app platform revenue comes from two sources: selling the paid apps or advertising on freemium apps. As a result, for the app store platform, consumer adoption of mobile apps represents a critical problem. The app store platform has a large amount of information about consumers’ download behavior, enabling it to customize its marketing actions to target different consumers according to their behaviors. For example, a mobile app platform can decide to offer viral-referral or free-trial strategies. A viral-referral strategy is useful when consumers’ preferences are interrelated because of the psychological benefits of social identification, learning, and inclusion and the utilitarian benefits of network externalities. A free- trial strategy is useful when consumers have learning costs or are uncertain about a mobile app. It is common for customers to have interrelated preferences for mobile apps. Online forums are filled with questions about requests for mobile app recommendations,1 and app stores try to inform users about the popularity of mobile apps. The interdependence of mobile app choices is important because customers often do not know what mobile app they want, so they rely on offline family, friends, and colleagues to find new apps. App stores have tried to facilitate this process by creating “Tell a Friend” and “Share This Application” (WonderHowTo 2011). Therefore, to design policies to influence consumers’ mobile app choices, an app store platform

1 See “Title of actual blog here” at http://www.cnet.com/forums/mobile-apps/ . 4 needs a framework to quantify not only the effect of mobile app characteristics but also the effect of online and offline social influences on customer choices.
Given this context, we ask the following questions:

1. How can we design a targeting approach for an app store platform?
2. How does the social learning process of mobile app customers differ from that of classic economy goods, such as a color television?
3. How can an app store platform capture the heterogeneity of its customers and the variation in mobile apps to customize its marketing actions?
4. What are the key elements of consumers’ utility of adopting a mobile app that allows an app store platform to group and target potential customers?
   To answer these questions, we combine a macro social learning diffus

This content is AI-processed based on ArXiv data.