Lineage stops at the view boundary and hides the real dependency.
Column-Level Lineage
One unified DAG covers pipelines, views, MVs, dynamic tables, and procedures — at column grain, before data moves.
Discover
Most lineage tools answer questions late, at table grain, or stitched across five products. BoltPipeline computes column-level lineage from a single parser that covers every producer — pipelines, views, materialised views, dynamic tables, stored procedures — and shows the actual transform expression on every hop.
What it solves
Common problems this capability removes from your engineering workflow.
Column maps are guessed at runtime, not verified at certification.
Type mismatches surface as broken deploys, not as caught problems.
Vendor lineage and platform lineage disagree with no way to reconcile.
How it works
The flow this capability runs end-to-end inside the platform.
- 1One parser covers pipelines, views, MVs, dynamic tables, procedures.
- 2Views are transparent — the graph traces through to base tables.
- 3Types flow through expressions so mismatches surface at certification.
- 4Every edge declares its provenance and verification verdict.
Related capabilities
Other capabilities in the BoltPipeline platform that work alongside this one.
1Intelligent Pipeline
Impact Analyzer
Paste any SQL change and see which pipelines, tables, and dashboards break — before you deploy.
2Validation &Schema Drift Detection
Catch schema changes the moment they land — classified by impact, tied to lineage, and shown across environments.
3Lineage &Asset Discovery
Three cooperating tasks keep your catalog fresh within five minutes and reconciled within twenty-four hours.
See it on your data.
Try BoltPipeline against your live database — your data never leaves your environment.