As organizations scale their data governance programs, migrating from Azure Purview (PaaS) to the more advanced Microsoft Purview (SaaS) becomes a strategic priority. However, this migration journey is far from simple—especially when it comes to preserving, cleaning, and enriching metadata across platforms.
To address these challenges head-on, our team developed a Metadata Migration Accelerator Utility —a custom-built, production-grade tool that simplifies and streamlines the entire migration lifecycle.
Why this utility is needed
Many enterprises invested significant time building comprehensive metadata catalogs in Azure Purview PaaS. But Microsoft’s shift to a SaaS-based offering brought a gaps and pain point which need a product accelerator that couldn’t be addressed through native capabilities at this point in time.
The Problem
Migrating metadata between platforms isn’t straightforward due to:
- No out-of-the-box support for transferring metadata between PaaS and SaaS environments.
- Manual limitations in editing metadata at scale—especially for large catalogs.
- Legacy metadata clutter, which introduces governance noise and reduces discoverability.
Migration Challenges
We identified key technical and operational roadblocks:
- API differences between PaaS and SaaS requiring transformation logic.
- Classification and glossary mismatches, complicating 1:1 mapping.
- Ownership & access control inconsistencies between platforms.
- Increased metadata complexity needed by SaaS for governance capabilities like classification, sensitivity labelling, and policy enforcement.
Our Solution: Metadata Migration Accelerator Utility
We built a robust utility that allows organizations to extract, edit, validate, and push metadata during the migration process—all through a flexible and modular pipeline.
Key Capabilities:
- Fetch Metadata from Azure Purview (PaaS)
Using REST APIs, the tool extracts complete metadata catalogs and stores them in a structured format. - Editable Excel Layer for Bulk Operations
Metadata is exported to an Excel file for data stewards and engineers to clean, tag, or remove entries—empowering teams with easy bulk-edit capabilities. - Scheduled Processing with Azure Functions
A serverless compute layer enables automated, rule-driven metadata validation and processing at regular intervals. - Push to Microsoft Purview (SaaS)
Once reviewed and finalized, enriched metadata is loaded into the SaaS environment via Microsoft Purview APIs.
Architecture Overview
Prerequisite:
Ensure assets are scanned in SaaS Purview, and Function App (MSI) has Data Curator role access on the working collection.
Core Components:
- Data Fetching: Extracts metadata from Azure Purview (PaaS) into ADLS Gen2 as Parquet/CSV files.
Original asset in Purview PaaS with metadata
Newly created asset in Purview - SaaS without metadata
Enriched asset in Purview – SaaS after metadata migration
- Editable Layer: Provides downloadable Excel files for manual or programmatic updates.
- Function App: Orchestrates read/validate/write logic for seamless SaaS updates.
- SaaS Update: Writes enriched metadata into Microsoft Purview SaaS environment.
Business Impact & Benefits
- 80% Faster Migration
Through automation, the utility drastically reduces manual effort and speeds up delivery timelines. - Improved Metadata Quality
Editable Excel interfaces allow for governance-ready metadata cleanup and enrichment before go-live. - Future-Proof Governance
Ensures metadata in SaaS is compliant with enterprise data governance policies from Day 1
Future RoadMap
Currently, the utility supports Azure Delta Gen2 Resource sets. Future enhancements will include:
- Support for additional asset types (Parquet, CSV, SQL tables, Synapse, custom assets)
- Enhanced lineage preservation
- Integration with governance workflows (e.g., approvals, alerts)
Conclusion
Transitioning from Azure Purview PaaS to Microsoft Purview SaaS marks a major step in evolving your data governance strategy. Our Metadata Migration Accelerator is not just a utility—it’s a product accelerator built for continuous improvement, ensuring metadata integrity, quality, and compliance throughout the migration journey.
Please find code repository in GitHub with below link:
Microsoft