Adding custom column in schema tab grid in MS Purview

Adding a custom column such as “Transformation Logic” directly into the Schema tab grid for scanned assets isn’t supported.

How the Schema tab is populated:

• For assets with intrinsic schema (for example, Azure SQL tables or Unity Catalog tables), the Schema tab is driven by the asset’s schema elements (columns) and their type definitions.
• The table entity points to its columns via a relationship (for example, schemaElementsAttribute).
• The column type definition can specify which of its own attributes appear in the Schema tab via schemaAttributes. In the Azure SQL example, data_type is surfaced this way in the Schema grid.

In the Unified Catalog experience, data asset attributes created under Custom metadata → Data asset attributes are asset-level key–value pairs. As documented, they extend the fields available for an asset and are shown in places like the Overview tab; they are not used by the Schema grid renderer and therefore do not appear as additional columns in the Schema tab.

Because the current experience does not expose a way to:

• change the column type definition for built‑in types coming from Unity Catalog, or
• bind a data asset attribute so that it is rendered as a column in the Schema grid, there is no supported method to add a custom column such as “Transformation Logic” into the Schema tab grid for Databricks Unity Catalog tables.

Recommended approaches to surface column‑level transformation logic:

1. Continue using data asset attributes at the column level
   • Store “Transformation Logic” as a data asset attribute on each column, as already configured. This is the supported way to attach structured metadata such as governance annotations to assets and columns.
   • Consumers can open the column asset and see the transformation logic in the Overview tab.
2. Use glossary terms for business explanation
   • For any explanatory or definitional aspect of the transformation (for example, what a derived metric means), use glossary terms and link them to the relevant columns. This is the recommended pattern for business explanations rather than structural metadata.
3. Use lineage where applicable
   • For Unity Catalog sources, Purview can capture lineage for tables and columns (subject to the documented limitations). Transformation logic is often best understood in the context of lineage graphs rather than as an extra field in the Schema grid.

At present there is no configuration or API in the documented surface that allows extending the Schema tab grid itself with additional custom columns for Unity Catalog–backed assets.

References:

• Type definitions and how to create custom types
• Create and manage data asset attributes
• Asset management in the Microsoft Purview Data Catalog
• Set up data quality for Azure Databricks Unity Catalog
• Microsoft Purview and Reltio integration for master data management