Personalized Mobile Targeting with User Engagement Stages: Combining Structural Hidden Markov Model and Field Experiment

40 Pages Posted: 5 Oct 2016 Last revised: 22 May 2018

See all articles by Yingjie Zhang

Yingjie Zhang

University of Texas at Dallas - Naveen Jindal School of Management

Beibei Li

Carnegie Mellon University - H. John Heinz III School of Public Policy and Management

Xueming Luo

Temple University

Xiaoyi Wang

Zhejiang University

Date Written: May 15, 2018

Abstract

Low engagement rates and high attrition rates have been formidable challenges to mobile apps and their long-term success, especially for those whose revenues derive mainly from in-app purchases. To date, little is known about how companies can scientifically detect user engagement stages and optimize corresponding personalized-targeting promotion strategies so as to improve business revenues. This paper proposes a new structural forward-looking Hidden Markov Model (FHMM) as combined with a randomized field experiment on app notification promotions. Our model can recover consumer latent engagement stages by accounting for both the time-varying nature of users' engagement and their forward-looking consumption behavior. Although app users in most of the engagement stages are likely to become less dynamically engaged, this slippery slope of user engagement can be alleviated by randomized treatments of app promotions. The structural estimates from the FHMM with the field-experimental data also enable us to identify heterogeneity in the treatment effects, specifically in terms of the causal impact of app promotions on continuous app consumption behavior across different hidden engagement stages. Additionally, we simulate and optimize the revenues of different personalized-targeting promotion strategies with the structural estimates. Personalized dynamic engagement-based targeting based on the FHMM can, compared with non-personalized mass promotion, generate 101.84% more revenue for the price promotion and 72.46% more revenue for the free content promotion. It also can generate substantially higher revenues than the experience-based targeting strategy applied by current industry practices and targeting strategies based on alternative customer segmentation models such as k-means or myopic HMM. Overall, the novel feature of our paper is its proposal of a new personalized-targeting approach combining the FHMM with a field experiment to tackle the challenge of low engagement with mobile apps.

Keywords: User engagement, Mobile content consumption, App platforms, Hidden-state model, Forward-looking behavior, Structural econometric model, Field experiment

Suggested Citation

Zhang, Yingjie and Li, Beibei and Luo, Xueming and Wang, Xiaoyi, Personalized Mobile Targeting with User Engagement Stages: Combining Structural Hidden Markov Model and Field Experiment (May 15, 2018). Available at SSRN: https://ssrn.com/abstract=2847634 or http://dx.doi.org/10.2139/ssrn.2847634

Yingjie Zhang

University of Texas at Dallas - Naveen Jindal School of Management ( email )

P.O. Box 830688
Richardson, TX 75083-0688
United States

Beibei Li

Carnegie Mellon University - H. John Heinz III School of Public Policy and Management ( email )

Pittsburgh, PA 15213-3890
United States

Xueming Luo (Contact Author)

Temple University ( email )

1810 N. 13th Street
Floor 2
Philadelphia, PA 19128
United States

HOME PAGE: http://www.fox.temple.edu/mcm_people/xueming-luo/

Xiaoyi Wang

Zhejiang University ( email )

38 Zheda Road
Hangzhou, Zhejiang 310058
China

Here is the Coronavirus
related research on SSRN

Paper statistics

Downloads
461
Abstract Views
1,929
rank
64,999
PlumX Metrics