Advertisement revenue management: Determining the optimal mix of skippable and non-skippable ads for online video sharing platforms

Abstract

Skippable video advertisements (ads), which allow uninterested users to skip the ad after a few seconds, have witnessed rapid growth in the past few years. While their advantages for viewers and advertisers are obvious, they pose an ad revenue optimization problem for their publishers, i.e., the Video Sharing Platforms (VSPs). The VSPs need to critically balance the higher but uncertain revenue from skippable ads with the lower but guaranteed revenue from non-skippable ads. This problem is particularly challenging because non-skippable ads cause higher disutility to viewers. Moreover, due to network effect, this disutility has a long term impact on the VSPs’ revenue. In this paper we study the revenue management problem faced by a VSP in determining the optimal mix of skippable and non-skippable ads. We model VSP as a two sided platform, identify conditions under which an advertiser would prefer skippable ads over non-skippable ones, and derive the optimality conditions for VSP’s optimal ad mix. Our model reveals the existence of an upper bound on number of non-skippable ads, such that continued violation of this upper bound leads to a cascading effect, resulting in a reduction of both skippable and non-skippable ads over time. Our analysis helps a VSP in determining the incentive it should provide to the advertisers to switch to its preferred ad type. Our study reveals that non-skippable ads are essential for VSPs with niche or low content, and the proportion of skippable ads increases as the content increases or becomes more general.