Skip to main content

Data management

The day-to-day activities involved in handling data, collecting it, storing it safely, keeping it accurate, and retiring it when no longer needed.

The Simple Version

The day-to-day activities involved in handling data, collecting it, storing it safely, keeping it accurate, and retiring it when no longer needed.

Detailed Explanation

Data management encompasses eleven knowledge areas in the DAMA Data Management Body of Knowledge (DMBOK): data governance, data architecture, data modelling, data storage, data security, data integration, documents and content, reference and master data, data warehousing and business intelligence, metadata management, and data quality. In the context of AI systems, data management activities directly affect the quality and compliance of training, validation, and test datasets, factors assessed during EU AI Act conformity assessments.

Key Characteristics

  • Operationalises data governance policies through day-to-day processes
  • Spans the full data lifecycle from acquisition through archival
  • Includes both technical operations (storage, integration) and quality activities
  • Enables consistent, repeatable handling of data assets across the organisation

Why It Matters

Investment in data management capability reduces the cost of AI development by ensuring datasets are well-organised, documented, and reliable before modellers begin work.

Real-World Analogy

Like the operations of a well-run library: books (data) are catalogued, shelved correctly, maintained in good condition, lent and returned through a governed process, and eventually deaccessioned when outdated.

Common Misconceptions

  • Data management is solely a database administration task, it covers all aspects of data handling from governance to quality to lifecycle management.
  • Cloud migration solves data management, cloud storage changes where data is kept, not how it is governed, documented, or quality-assured.

Related Terms

Related Articles

Sources & Further Reading