Originally published 17 April 2024. Updated 18 November 2026 after Microsoft announced it is retiring Purview Data Governance.
The question used to be which data catalog tool is better - Microsoft Purview or Dataedo. On 12 November 2026 Microsoft announced that Purview Data Governance will be retired and supported only until 12 November 2027. Data governance moves into Microsoft Fabric's OneLake catalog. Purview stays, but as a security and compliance product. If your business glossary, data owners, lineage or data quality rules live in Purview today, you have twelve months to choose where they go next. We compared Dataedo with Purview, and now also with the OneLake catalog, to help you make that choice.
What Microsoft announced in November 2026
Microsoft is moving data governance out of Purview and into Fabric. Purview keeps the security and compliance work. The announcement was made on 12 November 2026, and end of support is next year on 12 November 2027.
| What | Before | After the announcement |
|---|---|---|
| Purview Data Governance: Unified Catalog, Data Map, Workflows and classic governance (formerly Azure Purview) | Microsoft's data governance product | Retiring (Microsoft Partner Center). No new accounts and maintenance mode until 12 Nov 2027 (Microsoft Learn) |
| Governance home | Purview, a separate service next to the data platform | OneLake catalog in Microsoft Fabric |
| Billing | Separate Purview governance meters | Included in Fabric capacity |
| Non-Fabric data | Scanned by Purview Data Map | Reached through Fabric shortcuts, mirroring and supported connectors, without copying |
| Labels, classification, DLP, compliance | Purview | Still Purview |
One date comes earlier: Purview access policies for SQL retire on 30 October 2027.
Microsoft says detailed transition guidance is available. It asks regulated customers and those with contractual obligations to contact their account team early. It also names Atlan, Alation, Collibra and Informatica as third-party catalogs Fabric integrates with.
In other words, Microsoft itself acknowledges that some organizations need more than Fabric offers today. For mid-sized companies without million-dollar budgets, as well as organizations in highly regulated industries like healthcare, financial services, or government, Dataedo is another tool worth considering.
Dataedo overview
Dataedo is an end-to-end data governance tool for mid-sized organizations: data catalog, data lineage and data quality in one product. It works across Microsoft and non-Microsoft platforms and does not depend on Fabric.
- For business users: a business-friendly data catalog, a business glossary, data profiling, ERDs, and Data Community with threaded comments, reviews with ratings, and reactions.
- For data stewards: Steward Hub with suggested quality rules, glossary terms and keys, plus bulk assignment of owners and domains; workflows; and AI Autodocumentation.
- For technical teams: 100+ connectors, including open-source databases such as MySQL, MariaDB and PostgreSQL on any cloud. Automated column-level lineage for SQL Server, Power BI and Fabric, dbt, Databricks with Unity Catalog, Snowflake and more.
- Integrations: a public REST API, a Purview connector, a read-only MCP server (beta) for AI assistants, and notifications to Teams or Slack.
- Deployment: on-premises, self-hosted on AWS, Azure or Google Cloud, or hosted by Dataedo.
- Price: public pricing from $18,000 a year for Essentials (data catalog, 3 editors). Data lineage starts at $24,000 a year, and the full suite with data quality is $32,000. Every plan has unlimited viewers and connectors.
Microsoft Purview overview
Microsoft Purview is Microsoft's family of data security, compliance and governance tools. After November 2026, only the security and compliance part has a future.
- Governance (retiring): Data Map scans 45 source types, and Unified Catalog adds a glossary, governance domains, data products, data quality and health management reports. Workflows and the classic governance portal (formerly Azure Purview) retire with them. None of it takes new accounts, and it stays in maintenance mode until 12 November 2027.
- Security and compliance (continuing): automated classification with built-in sensitive information types, sensitivity labels, Information Protection, data loss prevention and compliance across Microsoft 365, Azure and Fabric.
- Strengths: deep integration with Azure and Microsoft 365, and automated detection of sensitive data.
- Limits: outside Azure, many sources get metadata only. Tableau, Qlik Sense, Salesforce, SAP HANA, MongoDB and Amazon Redshift have no automatic classification or lineage. Curation is limited to assigned roles, and lineage is richest for Azure sources and Data Factory or Synapse pipelines.
Dataedo vs. Purview vs. OneLake catalog at a glance
Purview's governance features no longer develop, so the real comparison for 2027 is Dataedo against the OneLake catalog, with Purview kept for protection.
| Area | Dataedo | Purview (governance, until 12 Nov 2027) | OneLake catalog |
|---|---|---|---|
| Role after 2027 | Governance catalog across all platforms | Security and compliance only | Microsoft's governance product for Fabric |
| Product status | Actively developed, releases every few weeks (release notes) | Maintenance mode, no new accounts | Actively developed (Fabric roadmap) |
| Scope | 100+ connectors, Fabric not required | 45 Data Map sources (list) | Fabric items; other data via shortcuts, mirroring and connectors |
| Business glossary | Yes, with bulk create and update | Yes, until retirement | Tags, descriptions and endorsement; no full glossary (Microsoft transition guidance) |
| Domains and data products | Yes | Yes, until retirement | Domains and subdomains for Fabric items |
| Lineage | Object and column level across platforms | Asset and column level, richest for Azure and Data Factory or Synapse pipelines | Lineage of Fabric items |
| Data quality | Rules, schedules, failed rows, history | Yes, until retirement | Governance health reports; no data quality rules engine (Microsoft transition guidance) |
| Data profiling | Yes | Within Unified Catalog data quality, retiring | No |
| ERDs and PK/FK discovery | Yes, reverse-engineered and suggested | No | No |
| Stewardship workflows | Steward Hub suggestions, workflows | Curation by assigned roles; workflows, retiring | Curation completeness and recommended actions |
| Sensitivity labels, DLP | Not a goal; documents classifications | Yes, continues | Shows Purview label and DLP coverage |
| AI access | MCP server (beta) for Copilot Studio, Claude, ChatGPT | Copilot search in Unified Catalog, retiring | Fabric MCP, public APIs, Copilot |
| Deployment | On-premises, self-hosted on AWS, Azure or Google Cloud, or Dataedo-hosted | Microsoft-hosted SaaS on Azure | Fabric SaaS |
| Pricing | Published: from $18,000 a year; lineage from $24,000; lineage and data quality $32,000 (pricing) | Separate meters, ending | Included in Fabric capacity |
Capabilities as of 18 November 2026. Check both vendors' current docs, and test your own scenarios in a proof of concept.
Dataedo vs. Purview by persona
The tables below update our 2024 comparison. Purview cells describe governance features as they stand in maintenance mode, so treat them as what you would migrate away from, not as a roadmap.
Data stewards
Stewards add business context to harvested metadata. This is the work most at risk in a migration.
| Feature | Dataedo | Purview |
|---|---|---|
| ERDs and keys | Reverse-engineers data models and ERDs with PK/FK. Steward Hub suggests missing keys | Discovers schema only; relationships must be documented by hand |
| Curation | AI Autodocumentation, now with several AI providers. Power users can edit descriptions in the web catalog | Limited to assigned roles: curators, data stewards and owners |
| Glossary | Glossary with relationships, dashboards, and bulk create and update (26.4) | Unified Catalog glossary, retiring |
| Steward portal | Steward Hub: suggested quality rules, glossary terms and keys; bulk assignment of owners and domains | Health actions and controls, retiring |
| Data quality | Rules on columns, schedules, saved failed rows, per-rule status and failed-row trends, overview dashboard | Unified Catalog data quality, retiring |
| Export | HTML and Excel export (Data Lineage and Data Quality plans) | No documentation export; glossary CSV export only in the classic portal |
| Collaboration | Data Community: threaded comments, reviews with ratings, reactions | Asset ratings |
Data workers
Business users, analysts and BI developers use the catalog to find and understand data. Purview was never the best tool for them. Its catalog was folded into Microsoft's security and compliance suite, and its search, browse and lineage views assume technical knowledge that most business users do not have.
| Feature | Dataedo | Purview |
|---|---|---|
| Interface | Business-friendly web catalog | Search, browse and lineage aimed at technical users |
| Search | Search with filters, plus Keyword Explorer, which surfaces frequent terms across the catalog | Search with filters |
| Reports | Report catalog: reports, glossary and datasets together, including Power BI, Tableau, Qlik and SSRS | Power BI and Fabric reports; Tableau and Qlik scanned without lineage |
| Lineage | Object and column-level lineage across platforms, redesigned diagrams (26.4) | Asset and column-level lineage, richest for Azure and Data Factory or Synapse pipelines |
| Profiling | Data profiling for 24 sources, including SQL Server, Oracle, Snowflake, Databricks, BigQuery and Fabric Lakehouse and Warehouse. Shows value distribution, completeness, formatting patterns, and top and random values, without writing a query. Full, Statistical or None mode controls how much sensitive data users see | Within Unified Catalog data quality, retiring |
| Learning | Tutorials and Academy | Microsoft Learn docs |
Data catalog owners
Catalog owners run the data governance program and answer for its budget, adoption and long-term plan. For them, the vendor's future, support and pricing matter as much as features.
| Feature | Dataedo | Purview |
|---|---|---|
| Product future | Active development; 26.4.1 shipped 16 Nov 2026 | Governance in maintenance mode until 12 Nov 2027 |
| Onboarding | Dedicated Data Success Manager on every plan | Microsoft account team and partners |
| Support | Email support with a 2-business-day response time | Microsoft support plans |
| Roadmap | Public product roadmap | Governance roadmap moves to the Fabric roadmap |
| Pricing | Published: Essentials $18,000, Data Lineage $24,000, Data Quality $32,000 a year; 3 editors included, unlimited viewers and connectors | Governance meters replaced by Fabric capacity |
Data source admins
Data source admins are the IT staff, such as cloud engineers and DBAs, who connect sources to the catalog and keep scans running. They care about connector coverage, APIs and where the tool can be deployed.
| Feature | Dataedo | Purview |
|---|---|---|
| Native connectors | 100+ (list) | 45 Data Map sources |
| Unsupported sources | Generic connectors, SQL DDL, OpenAPI and OpenLineage imports | Custom entities through the Atlas-based Data Map API |
| API | Public REST API since 24.2, plus a read-only MCP server (beta) | Data Map REST API (Apache Atlas) and Unified Catalog API (preview) |
| Import from Purview | Purview connector since 24.4 | — |
| Deployment | On-premises, self-hosted on AWS, Azure or Google Cloud, or Dataedo-hosted | Microsoft-hosted SaaS on Azure; scans on-premises and multicloud sources |
Connector differences
Both tools now cover the main Microsoft and cloud databases. The gaps are in BI, ETL, open-source and SAP sources. Checked against Dataedo docs and Microsoft Learn on 18 Nov 2026.
| Source | Dataedo | Purview Data Map |
|---|---|---|
| SQL Server | ✅ | ✅ (lineage only through ADF or Synapse pipelines) |
| Azure SQL Database | ✅ | ✅ (lineage in preview) |
| Oracle | ✅ | ✅ |
| Snowflake | ✅ | ✅ |
| Google BigQuery | ✅ | ✅ |
| Amazon Redshift | ✅ | ⚠️ Metadata only, no classification or lineage |
| Databricks Unity Catalog | ✅ | ✅ |
| Microsoft Fabric (Power BI, Warehouse, Lakehouse, Data Factory) | ✅ | ✅ |
| Tableau | ✅ | ⚠️ Metadata only, no classification or lineage |
| Qlik Sense | ✅ | ⚠️ Metadata only, no classification or lineage |
| Salesforce | ✅ | ⚠️ Metadata only, no classification or lineage |
| SAP HANA | ✅ | ⚠️ Metadata only, no classification or lineage |
| MongoDB | ✅ | ⚠️ Metadata only, no classification or lineage |
| Cassandra | ✅ | ✅ |
| dbt | ✅ | ❌ |
| SSIS | ✅ | ⚠️ Lineage only for packages run in the Azure Data Factory SSIS integration runtime; no on-premises SSIS |
| Fivetran | ✅ | ❌ |
| Coalesce | ✅ | ❌ |
| MariaDB | ✅ | ❌ |
| ClickHouse | ✅ | ❌ |
| Vertica | ✅ | ❌ |
| Neo4j | ✅ | ❌ |
| Elasticsearch | ✅ | ❌ |
| Starburst/Trino | ✅ | ❌ |
| SSRS | ✅ | ❌ |
| Power BI Report Server | ✅ | ❌ |
| SAP BO | ✅ | ❌ |
| Amazon QuickSight | ✅ | ❌ |
| Microsoft Dynamics 365 | ✅ | ⚠️ Through the Dataverse source |
| NetSuite | ✅ | ❌ |
| Oracle e-Business Suite | ✅ | ❌ |
| Looker | ❌ | ✅ |
| Airflow | ❌ | ✅ |
| Erwin | ❌ | ✅ |
| Teradata | ❌ | ✅ |
| SAP ECC | ❌ | ✅ |
| SAP S/4HANA | ❌ | ✅ |
| SAP Business Warehouse | ❌ | ⚠️ Metadata only, no classification or lineage |
| Azure Data Explorer | ❌ | ✅ |
| Azure Files | ❌ | ✅ |
| Azure Data Share | ❌ | ✅ |
| HDFS | ❌ | ✅ |
When Dataedo is a good Purview replacement
Dataedo replaces the governance half of Purview: the catalog, glossary, ownership, lineage and data quality. It is a strong fit when one or more of these is true.
| Your situation | Why Dataedo fits |
|---|---|
| Your data lives beyond Azure and Fabric: SQL Server on-premises, Oracle, Snowflake, Databricks, Tableau, Qlik or SAP BO | Native connectors for 100+ sources, no Fabric needed |
| Business users, not only stewards, must use the catalog | Business-friendly web catalog, glossary, Keyword Explorer and Data Community |
| You need column-level lineage from source to report | Automated column-level lineage across platforms, including dbt, SSIS and Power BI |
| Data quality and profiling should sit next to documentation | Quality rules, schedules, failed rows and profiling in the same tool |
| Your stewardship team is small | Steward Hub suggestions, AI Autodocumentation and edits by power users |
| Metadata must stay on-premises or in your own cloud | Self-hosted on-premises or on AWS, Azure or Google Cloud |
| You want a cost that is not tied to Fabric capacity | Published annual pricing with unlimited viewers and connectors |
| AI assistants should answer from governed metadata | Read-only MCP server (beta) for Copilot Studio, Claude and ChatGPT |
When Dataedo is not the right replacement
- Almost all your data is in Fabric and Power BI, and the OneLake catalog covers your needs within Fabric capacity.
- You use Purview mainly for sensitivity labels, DLP and compliance. Keep Purview for that; Dataedo does not replace it.
- You depend on sources Dataedo does not support natively, such as SAP ECC, SAP S/4HANA, Teradata or Looker. Check generic connectors and imports first.
Frequently asked questions
Is Microsoft Purview being discontinued?
No. Only Purview Data Governance is retiring: Unified Catalog, Data Map, Workflows and the classic governance features. Support ends on 12 November 2027, and Purview access policies for SQL end earlier, on 30 October 2027. Purview's security and compliance features, such as sensitivity labels and data loss prevention, continue.
What replaces Purview Data Governance?
Microsoft names the OneLake catalog in Microsoft Fabric as its data governance product. Non-Fabric data is reached through shortcuts, mirroring and supported connectors.
Can I still create a new Purview governance account?
No. Microsoft Learn says new accounts can no longer be created, and existing capabilities stay in maintenance mode until 12 November 2027.
Is the OneLake catalog free?
There is no separate meter: governance is included in Fabric capacity. Sensitivity labels and DLP on Fabric still need Purview licensing.
Does Dataedo replace Purview?
Dataedo replaces Purview's governance catalog: glossary, ownership, lineage, documentation and data quality. It does not replace Purview's sensitivity labels, DLP or compliance features, and it does not need to.
Do I need Microsoft Fabric to use Dataedo?
No. Dataedo connects directly to 100+ sources. If you use Fabric, Dataedo documents Power BI, Data Warehouse, Data Lakehouse and Data Factory.
Can I move my Purview content to Dataedo?
Partly automatically. The Purview connector imports assets and technical metadata, and builds object-level lineage within each import. Classifications and managed attributes come across too, but you map them manually. Glossaries load from Excel or CSV. Domains, data products, workflows and quality rules are rebuilt.
Can Dataedo run on-premises?
Yes. Dataedo runs on-premises, self-hosted on AWS, Azure or Google Cloud, or hosted by Dataedo.
Does Dataedo work with Microsoft Copilot?
Yes. Dataedo's read-only MCP server (beta) works with Copilot Studio, Claude and ChatGPT, and respects each user's Dataedo permissions.
How much does Dataedo cost?
Plans start at $18,000 a year for Essentials, the data catalog. Data Lineage costs $24,000 a year, and Data Quality, which adds data quality on top of lineage, costs $32,000. Each plan includes 3 editors, with unlimited viewers and connectors. See pricing.
Next steps
Moving off Purview? Book a free 30-minute Purview-to-Dataedo migration consultation and bring your current setup.
Considering Purview as your data governance tool but it's no longer a possibility? Explore Dataedo on a sample catalog: Try Dataedo or book a demo.