Optimizing Floors in First Price Auctions: an Empirical Study of Yahoo Advertising

• URL: http://arxiv.org/abs/2302.06018v2
• Date: Fri, 9 Feb 2024 17:44:51 GMT
• Title: Optimizing Floors in First Price Auctions: an Empirical Study of Yahoo Advertising
• Authors: Miguel Alcobendas, Jonathan Ji, Hemakumar Gokulakannan, Dawit Wami, Boris Kapchits, Emilien Pouradier Duteil, Korby Satow, Maria Rosario Levy Roman, Oriol Diaz, Amado A. Diaz Jr., Rabi Kavoori
• Abstract summary: We present a model to set floors in first price auctions, and discuss the impact of its implementation on Yahoo sites. Our solution induces bidders to change their bidding behavior as a response to the floors enclosed in the bid request. The annualized incremental revenue is estimated at +1.3% on Yahoo display inventory, and +2.5% on video ad inventory.
• Score: 0.0
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Floors (also known as reserve prices) help publishers to increase the expected revenue of their ad space, which is usually sold via auctions. Floors are defined as the minimum bid that a seller (it can be a publisher or an ad exchange) is willing to accept for the inventory opportunity. In this paper, we present a model to set floors in first price auctions, and discuss the impact of its implementation on Yahoo sites. The model captures important characteristics of the online advertising industry. For instance, some bidders impose restrictions on how ad exchanges can handle data from bidders, conditioning the model choice to set reserve prices. Our solution induces bidders to change their bidding behavior as a response to the floors enclosed in the bid request, helping online publishers to increase their ad revenue. The outlined methodology has been implemented at Yahoo with remarkable results. The annualized incremental revenue is estimated at +1.3% on Yahoo display inventory, and +2.5% on video ad inventory. These are non-negligible numbers in the multi-million Yahoo ad business.

Related papers

• A pragmatic policy learning approach to account for users' fatigue in repeated auctions [47.75983850930121]
  ML models can predict how previously won auctions the current opportunity value. A policy that uses this prediction tomaximize the expected payoff of the current auction could be dubbedimpatient. We provide two empirical arguments for the importance of this cost ofimpatience.
  arXiv  Detail & Related papers  (2024-07-15T07:53:29Z)
• Advancing Ad Auction Realism: Practical Insights & Modeling Implications [2.8413290300628313]
  This paper shows that one can still gain useful insight into modern ad auctions by modeling advertisers as agents governed by an adversarial bandit algorithm. We find that soft floors yield lower revenues than suitably chosen reserve prices, even restricting attention to a single query.
  arXiv  Detail & Related papers  (2023-07-21T17:45:28Z)
• Leveraging the Hints: Adaptive Bidding in Repeated First-Price Auctions [42.002983450368134]
  We study the question of how to bid in first-price auctions. Unlike in second-price auctions, bidding one's private value truthfully is no longer optimal. We consider two types of hints: one where a single point-prediction is available, and the other where a hint interval is available.
  arXiv  Detail & Related papers  (2022-11-05T19:20:53Z)
• A novel auction system for selecting advertisements in Real-Time bidding [68.8204255655161]
  Real-Time Bidding is a new Internet advertising system that has become very popular in recent years. We propose an alternative betting system with a new approach that not only considers the economic aspect but also other relevant factors for the functioning of the advertising system.
  arXiv  Detail & Related papers  (2020-10-22T18:36:41Z)
• Dynamic Knapsack Optimization Towards Efficient Multi-Channel Sequential Advertising [52.3825928886714]
  We formulate the sequential advertising strategy optimization as a dynamic knapsack problem. We propose a theoretically guaranteed bilevel optimization framework, which significantly reduces the solution space of the original optimization space. To improve the exploration efficiency of reinforcement learning, we also devise an effective action space reduction approach.
  arXiv  Detail & Related papers  (2020-06-29T18:50:35Z)
• Real-Time Optimization Of Web Publisher RTB Revenues [10.908037452134302]
  This paper describes an engine to optimize web publisher revenues from second-price auctions. The engine is able to predict, for each auction, an optimal reserve price in approximately one millisecond.
  arXiv  Detail & Related papers  (2020-06-12T11:14:56Z)
• Optimal Allocation of Real-Time-Bidding and Direct Campaigns [10.888918892489638]
  We consider the problem of optimizing the revenue a web publisher gets through real-time bidding (i.e. from ads sold in real-time auctions) and direct (i.e. from ads sold through contracts agreed in advance) This paper presents an algorithm to build an optimal strategy for the publisher to deliver its direct campaigns while maximizing its real-time bidding revenue.
  arXiv  Detail & Related papers  (2020-06-12T10:44:56Z)
• Do Interruptions Pay Off? Effects of Interruptive Ads on Consumers Willingness to Pay [79.9312329825761]
  We present the results of a study designed to measure the impact of interruptive advertising on consumers willingness to pay for products bearing the advertiser's brand. Our results contribute to the research on the economic impact of advertising, and introduce a method of measuring actual (as opposed to self-reported) willingness to pay in experimental marketing research.
  arXiv  Detail & Related papers  (2020-05-14T09:26:57Z)
• Online Joint Bid/Daily Budget Optimization of Internet Advertising Campaigns [115.96295568115251]
  We study the problem of automating the online joint bid/daily budget optimization of pay-per-click advertising campaigns over multiple channels. For every campaign, we capture the dependency of the number of clicks on the bid and daily budget by Gaussian Processes. We design four algorithms and show that they suffer from a regret that is upper bounded with high probability as O(sqrtT) We present the results of the adoption of our algorithms in a real-world application with a daily average spent of 1,000 Euros for more than one year.
  arXiv  Detail & Related papers  (2020-03-03T11:07:38Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.