Skip to content

Data Lineage

Tracking where data comes from and where it goes, and how lineage supports debugging, governance and trust.

Editorial team 1 min read

Data lineage records how data flows from sources through transformations to reports, models and applications.

Why It Matters

  • Debugging: trace a wrong number back to its source.
  • Impact analysis: see what breaks if a source changes.
  • Compliance: show where personal data flows.
  • Trust: users understand how figures are produced.
  • AI governance: know which data trained or feeds each model.

Levels

  • Table-level: which tables feed which.
  • Column-level: which columns derive from which.

Capturing Lineage

  • Parse SQL and transformation code.
  • Tools such as dbt generate lineage graphs automatically.
  • Orchestrators and open standards like OpenLineage emit lineage events.
  • Data catalogues visualise it.

Keep It Accurate

Automated capture beats manual documentation, which quickly goes out of date.

Use It

Integrate lineage into change reviews, incident response and data catalogues.

More in Data engineering

All Data engineering guides →
Data engineering Guide · 2 min

What Is Data Engineering?

What data engineers do, how data flows from source systems to analysis and AI, and the core skills involved.

Data engineering 2 min read 4 Jan 2026

Data engineering Guide · 2 min

ETL Versus ELT

The difference between transforming data before loading and after, and why modern warehouses shifted the default to ELT.

Data engineering 2 min read 3 Jan 2026

Data engineering Guide · 1 min

Data Warehouses, Data Lakes and Lakehouses

The three main architectures for analytical data storage, what each is good for and how they are converging.

Data engineering 1 min read 2 Jan 2026

Data engineering Guide · 2 min

Data Modelling for Analytics

Star schemas, facts and dimensions: how to design tables that make analysis fast, consistent and easy to understand.

Data engineering 2 min read 1 Jan 2026