What is Data Integration

Definition

Data integration is the practice of combining data from multiple, often incompatible, source systems into a unified, coherent view, resolving differences in structure, semantics, and identity so that consumers can analyse the combined data as if it came from a single source.

« Back to Glossary Index

Creates a single coherent view from systems that store the same entities differently
Resolves conflicting identifiers so a customer or product means the same thing everywhere
Eliminates the manual spreadsheet stitching that produces inconsistent, stale reports
Enables cross-domain analytics that no single source system could answer alone

Real World Example

An insurer integrates policy, claims, and call-centre systems into a unified customer view, letting agents see a policyholder's full history on one screen instead of switching between three applications.

FAQs

How does data integration differ from a data pipeline?

A pipeline moves data; integration focuses on reconciling and unifying data from different systems into a consistent combined view.

What challenges does data integration solve?

It resolves mismatched schemas, conflicting identifiers, and differing semantics so combined data is trustworthy and comparable.

What approaches support data integration?

Techniques include ETL/ELT, data virtualization, change data capture, and master data management depending on freshness and consistency needs.