dbt lineage visualization
The dbt lineage visualization in dbt-Workbench provides deterministic, readable graphs so teams can trust what they see. Layouts are built with D3/dagre for consistent positioning, while controls let you explore large projects without losing context.
Core capabilities
- Deterministic D3/dagre layout for consistent model placement.
- Model + column lineage with a shared graph experience.
- Grouping controls by schema, resource type, or tag.
- Depth controls to focus on relevant subgraphs.
- Expand/collapse for large DAGs.
- Impact highlighting for upstream and downstream changes.
dbt lineage graph workflow
- Start on the Lineage page and pick a model.
- Expand upstream and downstream nodes to surface dependencies.
- Turn on column lineage for deeper tracing.
- Use grouping and depth settings to declutter the graph.
For deep column tracing, move to the dbt column lineage page.
FAQ
How is the dbt lineage graph laid out?
dbt-Workbench uses a deterministic D3/dagre layout so the graph is stable between refreshes and across users.
Can I see both model and column lineage?
Yes. Model-level lineage is available by default, and column-level lineage can be enabled for detailed tracing.
Does the graph support grouping?
Grouping by schema, resource type, and tags is supported with collapsible aggregates.
How do I highlight impact?
Upstream and downstream impact highlighting is available at both model and column granularity.
What artifacts power lineage?
Lineage is derived from dbt manifest.json and catalog.json artifacts.