Case Study: Tchibo achieves six million daily predictions with Google Cloud Platform

Tchibo Optimizing demand forecasts with AI to match customer needs

Tchibo, the leading German retailer, needed a better way to forecast customer demand because its previous analytics approach was manually maintained, less accurate, and risky for operations and logistics. To address this, Tchibo worked with Google Cloud Platform, using Vertex AI and BigQuery to build a centralized analytics and forecasting platform for its online and retail operations.

Google Cloud Platform implemented the DEMON demand forecasting service on a microservice architecture, with Google Workflows orchestrating data collection, feature creation, model training, prediction, and delivery. The solution now generates about six million predictions per day, supports 84-day item forecasts, helps reduce logistics costs and stockouts, and enables more automated warehouse replenishment and ongoing growth through automation.

Tchibo

Sören Götze

Principal Data Scientist