Metadata Management strategy is a key component of data strategy. Metadata is information of data that includes business definition, valid values (data domains), business rules for creating the data, transformation and aggregation rules, security, data ownership, source, timelines, and data formats.

Metadata Categories (1)

Descriptive Metadata: Descriptive metadata is information that describes data so people can understand, find, and use it easily.

What descriptive metadata includes (e.g., for a dataset)

• Title of the dataset
• Description of what the data contains
• Purpose for which the data was created
• Subject or domain (e.g., finance, healthcare, sales)
• Owner of the data
• Business definitions of key data fields
• Keywords or tags to support search
  
  

Structural Metadata: Structural metadata describes how data is organized, related, and structured, so systems and people understand how data is put together.

What structural metadata includes (e.g., for a dataset)

• Tables, files in a dataset
• Columns and their order
• Hierarchies (parent–child structures such as database, schema, tables, columns)
• Relationships (primary keys, foreign keys)
• Administrative/ Auditing metadata: created date, created by, update date, etc.
• Lifecycle metadata: status, articulation score, etc.
  
  

Administrative Metadata: Administrative (or Auditing) Metadata describes how data is managed, accessed, and controlled. In other words, the “rules and history” around the data.

• Access and permissions of dataset – who can view, edit, or delete the data
• Ownership and stewardship – who is responsible for the data
• Data sensitivity classification – public, internal, confidential, or restricted
• Audit trails / history – records of who accessed or changed the data and when (create/ update dates)
• Retention and lifecycle rules – how long the data must be kept or archived
• Versioning – tracking different versions of datasets or records
• Compliance information – GDPR, HIPAA, or other regulatory requirements
  
  

Lifecycle metadata: Lifecycle metadata tracks the state and history of a data asset through its lifecycle. Essentially, it documents where the data asset is in its life journey with its status and helps manage it properly.

• Creation details: When and how the data asset was created (in Data Catalog)
• Audit trails / history – metadata records of who changed the metadata and when
• Metadata Quality / Validation status: Has metadata been validated?
• Lifecyle status: Data asset goes through different statues (drafted, initiated, under review, approved)

Metadata Categories (2)

Business Metadata: Business metadata, contributed by SMEs, conveys the business meaning of data to ensure a consistent contextual understanding across diverse data sources and applications.

Technical Metadata: Technical metadata, generated by technology (sometimes provided by technology team), is captured in technical formats within databases and applications, often abbreviations, shorthand, or character delimiters such as underscores, dashes, etc.

Process Metadata: Process metadata functions as technical documentation primarily for the technology community. It specifies actions performed by programs on data elements that include edit-routines, derived column calculations, data quality corrections, or data transformations. The process metadata is typically found in ETL tools, reporting tools, OLAP tools, etc.

Usage Metadata: Usage metadata records data access patterns and frequencies by both users and programs, enabling monitoring of who accesses specific data, how frequently, and at what times data is accessed.