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.

Mining Exploratory Behavior to Improve Mobile App Recommendations.

Mining consuming Behaviors with Temporal Evolution for Personalized Recommendation in Mobile Marketing Apps

User Multi-behavior Enhanced POI Recommendation with Efficient and Informative Negative Sampling.

Influence-Aware Successive Point-of-Interest Recommendation

Structural Analysis of User Choices for Mobile App Recommendation

Give Me Something Unknown: Incorporate Exploration Preference In Cognition Into Recommender System

What Special about Top-N Recommendation for Mobile App Stores

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

Learning Geographical and Mobility Factors for Mobile Application Recommendation

Group recommendation: by mining users' check-in behaviors.

Mining Behavioral Patterns from Millions of Android Users

Modeling User Engagement in Mobile Content Consumption with Tapstream Data and Field Experiment

Research on Personalized Recommendation of Smart Phones Applications Based on Association Rules and Collaborative Filtering Algorithms

Personalized Recommendation Model for Mobile E-commerce Users

A Novel APPs Recommendation Algorithm Based on APPs Popularity and User Behaviors

How is Mobile User Behavior Different? A Hidden Markov Model of Cross-Mobile Application Usage Dynamics

Investigating Fine-Grained Usefulness Perception Process in Mobile Search.

Mobile Phone Recommendation Based On Phone Interest

Diversified Mobile App Recommendation Combining Topic Model and Collaborative Filtering

A Mobile English Learning Platform Based on Data Mining and Personalized Recommendations

Mobile App Recommendation: an Involvement-Enhanced Approach