Glossary

What is data lineage?

Data lineage describes the path information takes from its sources through transformations to its destinations. In product operations, it can show which supplier field populated an attribute, which conversion was applied and where the result was published. Lineage supports tracing a problem backwards and understanding which dependent outputs may be affected by a change.

Trace movement at a useful level

Lineage can describe a whole file, a product record or a specific attribute. Choose a level that answers the operational question. Knowing that a supplier file entered the PIM may be insufficient when only one measurement passed through an incorrect conversion.

Record the systems, mappings and processing steps involved. Where possible, connect those steps to the relevant execution or version, so the team can distinguish the intended design from what happened to a particular batch.

Example: a dimension displayed incorrectly

A retailer notices that a cabinet width is wrong on its website. Attribute-level lineage shows the source column, the import mapping, a unit conversion and the outgoing destination field. The reviewer can identify where the meaning or value changed instead of correcting every destination independently.

If the mapping is repaired, the same record of dependencies helps identify which other products used that transformation and may need to be reprocessed.

Lineage and provenance

Lineage emphasises movement, transformation and dependency. Provenance also concerns the evidence, people and activities behind a value. The concepts overlap, and tools may present them together. Neither guarantees factual correctness: a perfectly documented pipeline can still carry an incorrect supplier specification. Keep source verification and approval alongside the lineage record.

Related terms

Sources and further reading

Product Data Services

Explore practical support for organising, improving and maintaining product information.