Kogan.com transformed their data platform from a request-based Tableau model to a self-service architecture using dbt, Looker, and Acryl DataHub. The new platform features modular data flows with layered dbt modeling, governed exploration through Looker's LookML, and comprehensive metadata management via DataHub. Key improvements include version-controlled business logic, CI/CD pipelines, row-level security, and AI-powered natural language queries. The migration process involves not just recreating dashboards but improving data models and user experience. The platform emphasizes governance through transparency rather than restrictions, enabling teams to explore data independently while maintaining trust and quality.
Table of contents
Rethinking Our BI ModelA New Architecture: Modular, Transparent, ScalableWhy Looker Was the Right Fit for Self-ServiceBridging the MigrationData Discovery and Observability with AcrylLessons We’ve LearnedFinal ThoughtsSort: