Case Study: Groupe Dubreuil improves product identification and search speed with Looker

Groupe Dubreuil Develops an application for real-time product identification using VisionAI

Groupe Dubreuil, a family-owned French group with businesses across multiple sectors, needed a faster way to help customers find products hidden among more than 300,000 catalog references. Working with Looker and Google Cloud technologies, it set out to build an AI-powered product identification experience for Agrizone that could match a photo to the right part in under 3 seconds.

Looker and the broader Google Cloud solution helped Groupe Dubreuil implement image scanning, OCR, and embedding-based search to identify products from photos and improve result relevance. The solution now recognizes products in an average of 2.13 seconds, with more than 500 scans per month on the Agrizone mobile app, a 90% product-found rate, and improved detection accuracy from 63% to 92% on sample images.

Groupe Dubreuil

Goulpeau Etienne

Data Scientist