dbt Mesh is a multi-project pattern for governed cross-team data products
About dbt Mesh
2023 docs introducing dbt Mesh: cross-project refs, Catalog, groups, access, versions, and contracts for independent yet aligned teams.
Based on
About dbt Mesh | dbt Developer Hub
The extract states that organizations outgrow a single dbt project when coordination, dependencies, and governance across teams lack a first-class path. Mesh is introduced as a pattern from converging features: cross-project {{ ref() }} on Enterprise plans, Catalog with cross-project lineage, governance of model access, groups with owners, access configs, model versions, and model contracts on data shape.
For mature data platforms, the page’s trigger conditions include model counts that slow development and teams that have already formed separate domains. The guidance is to treat shared models as stable APIs, using versions for adoption and deprecation and contracts so upstream changes do not silently break downstream products.
Governed shared meaning across systems is the mesh interface: public models, explicit owners, contracted shapes, and versioned change. Autonomy and collaboration are held together by those shared definitions rather than by merging every team into one project.
Put this to work on CoreModels
Related connectors and recipes
See all connectors →Take the next step
Try CoreModels, talk with our team, or explore more resources.