This paper proposes an optimal dynamic model for combining guaranteed andspot markets in display advertising. We assume a media seller can well estimatethe total supply and demand of page views from a specific advertisement (inshort ad) slot in a specific future period. The model helps the seller todetermine how to distribute and price those future page views betweenguaranteed contracts and advertising auctions. The former is soldalgorithmically in advance while the latter happens in a few milliseconds aftera user visits the Web page in the future. Therefore, the former is calledprogrammatic guarantee (PG) and the latter is called real-time bidding (RTB).This is one of a few studies that investigate the RTB-based posted price PG fordisplay advertising. The optimization problem is challenging because theallocation and pricing of PG affect the expected revenue from future RTBcampaigns. Several assumptions are made on media buyers' behavior, such as riskaversion, stochastic demand arrivals, and effects of time and guaranteedcontract price. We use dynamic programming to solve the optimization problemand our solution is relatively scalable and efficient. We validate the proposedmodel with an RTB dataset and find it increases the seller's expected totalrevenue by adopting different pricing and allocation strategies according tothe level of competition in future RTB campaigns.
展开▼
