Not Just a Fad: Optimal Sequencing in Mobile In-App Advertising

• 成果类型: Article
• 署名作者: Sun, Zhen; Dawande, Milind; Janakiraman, Ganesh; Mookerjee, Vijay
• 署名单位: George Washington University; University of Texas System; University of Texas Dallas
• 刊物名称: INFORMATION SYSTEMS RESEARCH
• ISSN/ISSBN: 1047-7047
• DOI: 10.1287/isre.2017.0697
• 发表日期: 2017
• 页码: 511-528
• 关键词: scheduling advertisements; web page; clickstream; performance; algorithm

摘要

In this paper, we address the challenge faced by ad networks in managing the fading ads (or fads) shown to an end user during a session of a mobile application (app). A fad is an ad that disappears if the user does not interact with it for some length of time. The withdrawn ad could be replaced by another ad. The goal of the ad network is to determine the sequence of fads shown to the user in an ad space to maximize the expected revenue generated over the user's app session. Mobile in-app advertising is uniquely suited for the sequencing of fads because user sessions are typically longer (than web sessions), and a single ad is displayed at any given point in time. We consider two factors that affect the probability of a click on an ad during a session: (i) the sojourn effect, the influence of the passage of time, and (ii) the exposure effect, the influence of the number of prior exposures of the ad to the user during that session. We provide simple and optimal policies for the ad-sequencing problem when either of these two effects dominates. For the general case in which both effects are significant, we offer a provably near-optimal heuristic policy. The following two enhancements to the basic sequencing problem are also analyzed: (a) consideration of both click ads (which generate revenue for the ad network only through clicks) and display ads (which generate revenue only through exposures) and (b) the presence of a constraint imposed by the publisher (i. e., the owner of the app) that the expected revenue in each time slot exceeds a certain threshold.