Data Lineage

Data lineage is the record of where data came from and what happened to it along the way: which source produced a field, which transformations touched it, and which tables, dashboards, and models depend on it downstream. Column-level lineage answers the two questions every data team asks weekly — “can I trust this number?” (trace it back) and “what breaks if I change this?” (trace it forward).

Lineage is typically harvested automatically from SQL parsing, orchestrator metadata, and warehouse query logs, and surfaced in a data catalog. Its practical value is impact analysis, incident debugging, and audit — provenance you can query instead of tribal knowledge you have to ask around for.

Go deeper: What Is Data Lineage, and What Is It Actually For?