2. Discover, Map, and Curate the Data Estate
Build Unified Catalog, Lineage, Quality, and Data Products
Convert scanned metadata into business domains, trusted products, lineage, access workflows, and measurable data health.
Approx. 3 min reading · Includes diagrams; allow extra time for reflection and linked resources.
Start reading ↓About this learning content and editorial review
Your focus
What you’ll learn
- Apply a repeatable method for build unified catalog, lineage, quality, and data products in a licensed, governed, and testable Purview environment.
This lesson develops a practical understanding of build unified catalog, lineage, quality, and data products and connects design choices to supported capabilities, operational dependencies, user impact, and verifiable evidence.
A catalog becomes valuable when technical assets gain business context
Unified Catalog organizes metadata into governance domains and data products that people can find and understand. A domain establishes an accountability boundary. A data product packages data assets for a business purpose with an owner, description, intended use, access expectations, quality signals, and terms. Glossary terms provide shared language; critical data elements identify fields whose failure would materially affect the business.
Do not begin by importing thousands of glossary terms. Choose one domain with an important decision, identify consumers and stewards, curate a small set of products, and prove that users can discover and safely request the data they need. A “Customer 360” product should explain sources, freshness, quality, permitted use, owner, and lineage—not merely point to a table.
Lineage answers how data moved and changed. It helps assess downstream impact, investigate quality failures, and understand which reports or AI systems depend on a source. Validate automated lineage against real transformations and document manual gaps.
Measure trust rather than catalog volume
Catalog success is not the number of scanned assets. Measure whether important products have owners, descriptions, lineage, quality rules, current classifications, access instructions, and active consumers. A completeness percentage without a business denominator is cosmetic.
Use a product readiness review:
- Is the business purpose and approved use clear?
- Are authoritative and non-authoritative sources distinguished?
- Does lineage cover the transformations that affect meaning?
- Are critical elements linked to quality rules and accountable owners?
- Can a consumer request access without bypassing source authorization?
- Are privacy, retention, and security requirements visible?
- Is there a review date and retirement path?
The catalog governs metadata and workflows; source systems still enforce underlying access. The Data Map and DLP limitations guide prevents the common mistake of assuming a catalog classification automatically blocks Microsoft 365 activity.
Resources
- Microsoft Learn reference for Build Unified Catalog, Lineage, Quality, and Data Products — Official Microsoft documentation supporting the capability, prerequisites, and current product behavior taught in this lesson.