Advancing Ad Auction Realism: Practical Insights & Modeling Implications

TL;DR

The paper addresses realism gaps in online ad auctions by modeling advertisers as no-regret learners (Hedge and EXP3-IX) operating under query-dependent values, partial feedback, and imperfect pricing rules. It demonstrates two main contributions: (i) soft-floor pricing can boost revenue in multi-query environments even with ex-ante symmetric bidders, while in single-query settings a well-chosen standard reserve price often dominates; (ii) an inverse-bidding approach infers bidder value distributions from observed bids, validated on synthetic data and applied to production e-commerce data. The results provide a practical, simulation-based framework for evaluating auction formats and for empirical valuation inference, with implications for auction design and revenue optimization in complex ad markets.

Abstract

Contemporary real-world online ad auctions differ from canonical models [Edelman et al., 2007; Varian, 2009] in at least four ways: (1) values and click-through rates can depend upon users' search queries, but advertisers can only partially "tune" their bids to specific queries; (2) advertisers do not know the number, identity, and precise value distribution of competing bidders; (3) advertisers only receive partial, aggregated feedback, and (4) payment rules are only partially known to bidders. These features make it virtually impossible to fully characterize equilibrium bidding behavior. This paper shows that, nevertheless, one can still gain useful insight into modern ad auctions by modeling advertisers as agents governed by an adversarial bandit algorithm, independent of auction mechanism intricacies. To demonstrate our approach, we first simulate "soft-floor" auctions [Zeithammer, 2019], a complex, real-world pricing rule for which no complete equilibrium characterization is known. We find that (i) when values and click-through rates are query-dependent, soft floors can improve revenues relative to standard auction formats even if bidder types are drawn from the same distribution; and (ii) with distributional asymmetries that reflect relevant real-world scenario, we find that soft floors yield lower revenues than suitably chosen reserve prices, even restricting attention to a single query. We then demonstrate how to infer advertiser value distributions from observed bids for a variety of pricing rules, and illustrate our approach with aggregate data from an e-commerce website.

Figures (8)

• Figure 1: Bids in second-price auction under Hedge
• Figure 2: Bidder 1's bids in second-price auction, EXP3-IX
• Figure 3: Expected revenue for soft-floor reserve price (SFRP, blue line) and standard reserve price (RP, orange line) under right-skewed value distributions. We exclude reserve prices between $1.0$ and $1.8$, as they yield lower revenue than reserve prices at $1.8$ or below $1.0$. RP=$0$ is the same as SFRP=$0$.
• Figure 4: Expected revenue for soft-floor reserve price (SFRP, blue line) and standard reserve price (RP, orange line) under symmetric value distributions.
• Figure 5: Expected revenue for soft-floor reserve price (SFRP, blue line) and standard reserve price (RP, orange line) under left-skewed value distributions.