Effectiveness of Mobile Notification Delivery

In this paper, user log data of mobile notifications are collected from a real mobile news application, and notification opening rate and reaction time are identified as key parameters that characterize user behavior. The results indicate that notification opening patterns depend on the content of notifications, the time at which they are sent, the activeness of users, and the reaction time. This paper also demonstrates that there exists an inverse relationship between the notification opening rate of users and their notification reaction time, and between their notification opening rate and notification opening volume. Then, a novel method of quantifying notification usefulness for users is developed, where 3 models are presented to describe usefulness of notifications against the volume of notifications received, each relevant in different contexts. Utilizing this usefulness model, a notification optimization framework is developed for sending notifications from different news categories. The optimized notification allocation improved the average notification opening rate by by 65.24% from 4.92% to 8.13%, and the average reaction time improved by 13.13% from 20.4 hours to 17.68 hours.