Smart Fast Finish: Preventing Overdelivery via Daily Budget Pacing at DoorDash

2025-09-09

AI Summary

To address overdelivery—i.e., budget overspending caused by pacing failure near the end of an advertising campaign—the paper proposes a dynamic budget allocation method. Methodologically, it introduces a historical-data-driven adaptive parameter tuning mechanism that dynamically optimizes throttling initiation time and rate, thereby enhancing budget consumption stability and robustness. The approach integrates online budget-splitting experiments, offline simulation modeling, and closed-loop parameter optimization, trained and validated on real-world advertising campaign data. Deployed in DoorDash’s production environment, the solution reduces overdelivery by XX% and improves budget utilization by X.X%, while simultaneously boosting delivery efficiency and campaign goal attainment. This work provides a scalable, production-ready technical framework for large-scale, real-time advertising budget control.

Technology Category

• Search and Optimization: Algorithm Configuration
• Machine Learning: Online Learning & Bandits
• Planning, Routing, and Scheduling: Planning with Language Models

Application Category

• User Modeling, Personalization and Recommendation: User modeling for targeted and personalized online advertising
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web
• Economics, Online Markets and Human Computation: Advertising auctions, pricing, markets, and exchanges

Abstract

We present a budget pacing feature called Smart Fast Finish (SFF). SFF builds upon the industry standard Fast Finish (FF) feature in budget pacing systems that depletes remaining advertising budget as quickly as possible towards the end of some fixed time period. SFF dynamically updates system parameters such as start time and throttle rate depending on historical ad-campaign data. SFF is currently in use at DoorDash, one of the largest delivery platforms in the US, and is part of its budget pacing system. We show via online budget-split experimentation data and offline simulations that SFF is a robust solution for overdelivery mitigation when pacing budget.

Problem

• Preventing overdelivery in advertising budget pacing
• Dynamically updating system parameters using historical data
• Mitigating budget overdelivery via robust pacing solution

Innovation

• Dynamically updates system parameters historically
• Adapts start time and throttle rate automatically
• Prevents overdelivery via robust budget pacing