Abstract
A challenge in the field of digital media distribution relates to static interruption placements, such as advertisements, which may be determined without considering evolving audience engagement data. This approach can result in interruptions during periods of low viewer interest, potentially increasing session abandonment. Systems and methods are described that can address this by utilizing a quasi-experimental statistical framework to analyze audience engagement data. The framework may compare viewer abandonment rates between sessions where an interruption was presented and sessions where an interruption was not presented for the same temporal location. This comparison can yield a causal abandonment penalty metric for each potential interruption point. Based on a statistical comparison of these penalty metrics, a system can dynamically migrate interruptions from locations with higher abandonment penalties to locations that may correspond with higher viewer engagement, which may improve viewer retention.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
Piega, Joanna; Lepkowski, David; and Socha, Szymon, "Adaptive Media Interruption Placement Using Causal Analysis of Audience Engagement", Technical Disclosure Commons, (August 26, 2026)
https://www.tdcommons.org/dpubs_series/11486