• How Yelp Built a Back-Testing Engine for Safer, Smarter Ad Budget Allocation - Samuele Mazzanti, Applied Scientist - Feb 2, 2026 Introduction Modern advertising platforms are fast-paced and interconnected: even small adjustments can have ripple effects on how ads are shown, how budgets are spent, and the value advertisers get from their ad spend. • At Yelp, Ad Budget Allocation means splitting each campaignâs spend between onâplatform inventory (our website, mobile site, and app) and offâplatform inventory (the Yelp Ad Network). • We optimize this split to meet advertisersâ performance goals while growing overall revenue. • Due to the complexity of the budget allocation system and its feedback loop, even small changes can lead to unexpected systemâwide effects. • To help us safely evaluate changes, we developed a Back-Testing Engine. • This tool allows us to simulate the entire Ad Budget Allocation ecosystem with proposed algorithm changes, giving us a preview of real-world effects before we run full A/B tests or launch new code.
Article Summaries:
- Yelp has developed a back‑testing engine to evaluate changes to its ad‑budget allocation system before deploying them live. The tool simulates how proposed allocation algorithms would perform using aggregated historical campaign data, allowing the team to see potential impacts on spend, leads, and revenue without affecting real advertisers. By modeling the daily budget‑setting loop-where each day’s spend depends on prior outcomes-the engine helps uncover cascading effects that could arise from even minor tweaks. This safe‑testing approach speeds innovation, reduces risk to advertiser trust, and improves decision‑making across Yelp’s large, complex advertising ecosystem.
Sources: