A dbt "logic mesh" with packages: standard model and metrics across 25 autonomous organizations
Blog post from dbt
Greenpeace International used a three-layer dbt package architecture to help a central analytics team support 25 autonomous national and regional organizations with different technology stacks, reporting practices, and metric definitions. After attempts to standardize source systems and repeatedly rebuild transformations proved impractical, the team modeled stable business concepts rather than underlying tools and established shared definitions for global reporting while retaining room for local variations. The architecture combines source packages, organization-specific intermediate transformations, and a standardized dbt package containing shared models, tests, documentation, governance, macros, and semantic definitions, supported by Fivetran, BigQuery, Hex, and Terraform. By October 2026, 14 organizations had adopted the model, with about 75% of their logic delivered through shared packages; automated CI allows most of roughly 50 monthly package releases to propagate across 170 projects without manual intervention. Although the approach requires substantial change management, creates a complex multi-project codebase, and has exposed package-related dbt v2 issues, it has reduced the time needed to deliver global metrics from months to daily insights.
No tracked trend matches for this post yet.
Use this post, company, and trend context to find content marketing opportunities, perform competitive analysis, or address product feature gaps via the Plushcap MCP server or the Plushcap API.