Placement Disclosure in Ad Auctions: Evidence From a Policy Change
41 Pages Posted: 8 Aug 2019
Date Written: August 5, 2019
Ad exchanges, i.e., platforms where real-time auctions for ad impressions take place, have developed sophisticated technology and data ecosystems to allow advertisers to target users, yet advertisers often do not know which sites their ads appear on. In practice, ad exchanges can easily grant ad placement information to advertisers, allowing advertisers to bid on ads at specific sites. However, ad exchanges are reluctant to disclose placement information due to fears that advertisers will start buying ads only on the most desirable sites leaving inventory on other sites unsold and lowering average revenue. The theoretical literature on information disclosure in auctions suggests that auction revenue is a concave function of information provided and prices will fall when too much or too little information is provided about impressions. This paper explores the empirical effect of ad placement disclosure using a unique data set describing auction outcomes on a major European ad exchange. We find that average revenue per impression rises when ad placement information is provided suggesting that ad context information is important to advertisers. The exception to this are sites which had a low number of buyers prior to the policy change; consistent with theory, these sites with thin markets see prices go down. Our analysis adds evidence that publishers and ad exchanges with thick markets should provide advertisers with site placement information, which can be done at almost no cost.
Keywords: Online display advertising, Real-time bidding, Advertising auctions, Information disclosure, Conflation, Bundling
JEL Classification: M37
Suggested Citation: Suggested Citation