Mobile AD(D): Estimating Mobile App Session Times for Better Ads

Abstract

While mobile advertisement is the dominant source of revenue for mobile apps, the usage patterns of mobile users, and thus their engagement and exposure times, may be in conflict with the effectiveness of current ads. Users engagement with apps can range from a few seconds to several minutes, depending on a number of factors such as users' locations, concurrent activities and goals. Despite the wide-range of engagement times, the current format of ad auctions dictates that ads are priced, sold and configured prior to actual viewing, that is regardless of the actual ad exposure time. We argue that the wealth of easy-to-gather contextual information on mobile devices is sufficient to allow advertisers to make better choices by effectively predicting exposure time. We analyze mobile device usage patters with a detailed two-week long user study of 37 users in the US and South Korea. After characterizing application session times, we use factor analysis to derive a simple predictive model and show that is able to offer improved accuracy compared to mean session time over 90% of the time. We make the case for including predicted ad exposure duration in the price of mobile advertisements and posit that such information could significantly impact the effectiveness of mobile ads by giving publishers the ability to tune campaigns for engagement length, and enable a more efficient market for ad impressions while lowering network utilization and device power consumption.