Boosted Second-price Auctions for Heterogeneous Bidders
33 Pages Posted: 12 Aug 2017 Last revised: 13 Mar 2018
Date Written: August 10, 2017
Due to its simplicity and desirable incentive properties, the second-price auction has been the prevalent auction format used by advertising exchanges. However, even with the optimized choice of the 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 ad exchanges, we propose an auction format called the boosted second-price auction, which assigns a boost value to each bidder. The auction favors bidder with higher boost values and allocates the item to a bidder with the highest boosted bid.
We propose a data-driven approach to optimize boost values using previous bids of the bidders. Our analysis of auction data from a large online advertising exchange shows that our algorithm can improve the revenue by up to 6%. Furthermore, we observe that the data-driven algorithm assigns higher boosts to advertisers with more stable bidding behavior. We show how this connects to the Myerson’s optimal mechanism design framework for heterogeneous bidders and propose a boosted second-price auctions, where bid distributions with lower inverse hazard rates receive a higher boost. We establish conditions which guarantee that these boosted auctions will increase revenue over the second-price auctions.
Keywords: Boosted Second-price Auctions, Online Advertising, Heterogeneity, Brand, Retargeting
Suggested Citation: Suggested Citation