Case Study: Yahoo Japan Corporation achieves faster, more accurate ad screening with Google Cloud Platform

Rapidly developing fraudulent ad screening with the Tech Acceleration Program using AI

Yahoo Japan Corporation, a major Japanese internet and media company, needed a better way to screen the huge volume of ads it handles each day. Its existing manual and automated checks were struggling to keep up with increasingly sophisticated fraudulent ads, creating pressure on the trust and safety team. Yahoo Japan Corporation turned to Google Cloud Platform, using services such as BigQuery and Cloud AutoML to build a more scalable, AI-driven ad screening system.

With Google Cloud Platform, Yahoo Japan Corporation created a high-speed machine learning workflow using Cloud Run, Vertex AI, Firestore, and Cloud Functions to classify ads as OK or NG and support non-technical staff in building models. The company said the new system improved guideline violation detection efficiency, reduced harmful ads served to users, and helped block approximately 9.5 million fraudulent ads in one month.

Yahoo Japan Corporation

Ichijo

General Manager