The audience for in-app mobile advertising is comparable in size and viewing rate to that for TV but divides its attention across a highly fragmented selection of apps, each competing for advertiser revenue. In market, the assumption is that this audience is deeply segmented, allowing individuals to be contextually targeted on the apps that define their interests and needs. But that assumption is not supported by the Laws of Double Jeopardy and Duplication of Viewing which closely predict usage in most mass media. The purpose of this study is to benchmark in-app audiences against these laws to better understand market structure.
The authors collected nearly 3,000 h of screen time data from a panel of Generation Z respondents and tested the predictive validity of two models against observed interactions with 23 popular apps in six categories over a week.
Results show that contrary to industry assumptions, audience for in-app advertising is not segmented. Engagement on individual apps and audience sharing rates between apps and app formats is predicted well.
Optimising in-app advertising for short-term activation only limits its potential for brand building. These findings encourage advertisers to schedule online campaigns for brand reach as well as sales lift, by advancing current understanding of audience behaviour.
Many authors have called for consistency in metrics to compare on- and off-line media performance. This study bridges that gap, demonstrating how reach and frequency measures could inform digital scheduling.
