Model Monotonicity in Autobidding Auctions: When Do Better Predictions Lead to Better Outcomes?

2026-05-29

AI Summary

This study investigates whether improving the prediction quality of machine learning models in automated bidding ad auctions invariably enhances platform-centric metrics such as revenue and social welfare. By developing a theoretical framework that formally characterizes model improvements through probabilistic clustering refinement, the work systematically analyzes the interplay among recommendation models, auction mechanisms, and automated bidding strategies, leveraging mechanism design theory, Jensen’s inequality, and constructive numerical examples. The primary contributions include the first characterization of monotonicity properties of platform metrics under various bidding types (e.g., tCPA), auction formats, and budget constraints; specifically, it proves that revenue monotonicity holds in first-price auctions with tCPA bidding absent budgets, whereas second-price auctions or the introduction of budget constraints can violate this monotonicity, for which complete counterexamples are provided.

Abstract

Online advertising platforms rely on machine learning models to predict click-through rates (pCTR) and conversion rates (pCVR) for auction mechanisms. We introduce a novel framework to study the interaction between recommender system model quality, auction format, and autobidder behavior. We formalize when model improvements -- defined via a refinement relation inspired by filtrations in probability theory -- lead to improvements in platform-level Evaluation Criteria Metrics (ECM) such as revenue, welfare, or liquid welfare. Our main contributions are: (1) a formal definition of model improvement based on cluster refinement, and (2) a systematic characterization of ECM monotonicity across different combinations of bidder types (tCPA, max-CPA), auction formats (first-price, second-price, VCG), and budget constraints. We show that first-price auctions with uniform bidding guarantee revenue monotonicity for tCPA bidders without budgets (via Jensen's inequality), while second-price auctions and budget constraints can break this property. We provide full numerical constructions for the non-monotonicity results. Our findings have practical implications for advertising platforms seeking to align model improvements with business outcomes.

Problem

• autobidding
• auction monotonicity
• model refinement
• evaluation criteria metrics
• online advertising

Innovation

• model monotonicity
• autobidding
• auction design
• refinement relation
• evaluation criteria metrics