Boosted Second Price Auctions: Revenue Optimization for Heterogeneous Bidders

Abstract

The second price auction has been the prevalent auction format used by advertising exchanges because of its simplicity and desirable incentive properties. However, even with an optimized choice of reserve prices, this auction is not revenue optimal when the bidders are heterogeneous and their valuation distributions differ significantly. In order to optimize the revenue of advertising exchanges, we propose an auction format called the boosted second price auction, which assigns a boost value to each bidder. The auction favors bidders with higher boost values and allocates the item to the bidder with the highest boosted bid. We propose a data-driven approach to optimize boost values using the previous bids of the bidders. Our analysis of auction data from Google's online advertising exchange shows that the boosted second price auction with data-optimized boost values outperforms the second price auction and empirical Myerson auction by up to 6% and 3%, respectively.