Unity Catalog vs Microsoft Purview
Unity Catalog is the governance layer built into Databricks: it controls who can use which tables, files, models, and functions in a Databricks account, records lineage, and audits access. Microsoft Purview is Microsoft's family of data governance, data security, and compliance products, whose governance part (Data Map and Unified Catalog) scans data sources across an organization and adds business context such as governance domains, data products, and glossary terms.
The comparison typically comes up in companies that run Azure Databricks next to Power BI, Microsoft Fabric, and Microsoft 365. The short answer is that the two are rarely a choice of one or the other. They work at different levels, and Microsoft documents a connector that brings Unity Catalog metadata and lineage into Purview. The useful questions are what each one governs, where they overlap, and what sits beyond both.
Unity Catalog governs data and AI assets inside Databricks: fine-grained access control, lineage, auditing, and sharing, included with the platform at no separate license cost. Microsoft Purview spans data governance, data security (such as DLP and Information Protection), and compliance across a Microsoft tenant, and its Unified Catalog is billed per governed asset. Purview can scan Unity Catalog to pull in technical metadata, tags, and notebook lineage, and Microsoft documents running both together: Unity Catalog enforces access where the data lives, Purview adds the enterprise view. Unity Catalog stops at the Databricks boundary and Purview's depth varies by source, which is where a cross-platform catalog comes in.
What Is Unity Catalog?
Unity Catalog is Databricks' unified governance solution for data and AI. Every asset sits in a three-level namespace, catalog.schema.object, and the objects it governs include tables, views, volumes, functions, and models. Its main capabilities, as Databricks documents them, are:
- Access control through privileges, attribute-based policies, row filters and column masks, and workspace bindings.
- Lineage captured automatically, from source tables through to models and dashboards, and exposed in system tables.
- Auditing through an audit log system table that records data access and activity.
- Discovery, classification, and quality in Catalog Explorer, with automatic tagging of sensitive data and data quality monitoring.
- Sharing of live data and AI assets with other organizations and clouds.
Unity Catalog can also govern some data outside Databricks. Through Lakehouse Federation, external systems such as Snowflake, Google BigQuery, Amazon Redshift, SQL Server, or an AWS Glue metastore appear as read-only foreign catalogs, with Unity Catalog table-level access control on top. The governance still runs through Databricks: federated data is governed when it is queried from Databricks. An open-source implementation of Unity Catalog is available too.
What Is Microsoft Purview?
Microsoft Purview is not a single catalog but a set of solutions in three areas:
- Data governance: the Data Map, which scans and inventories data sources, and the Unified Catalog, which organizes them into governance domains and data products, adds glossary terms, critical data elements, data quality rules, and health management, and handles self-service access requests.
- Data security: Data Loss Prevention, Information Protection with sensitivity labels, Insider Risk Management, and Data Security Posture Management (in preview).
- Data compliance: Audit, eDiscovery, Compliance Manager, Records Management, and related tools.
The Data Map is not limited to Microsoft services. Microsoft lists sources such as Snowflake, Amazon S3, Google BigQuery, Oracle, SAP, Teradata, and Tableau alongside Azure services, Power BI, and Fabric, although the depth of support, such as classification, policies, and lineage, varies by source. For this comparison the relevant part is data governance, which is what competes with or complements Unity Catalog. The cost of that part is covered in Microsoft Purview pricing.
Unity Catalog vs Purview: Side-by-Side Comparison
| Dimension | Unity Catalog | Microsoft Purview (data governance) |
|---|---|---|
| Primary scope | Data and AI assets in a Databricks account | The data estate registered in the Purview Data Map, centered on the Microsoft tenant |
| Data outside the platform | Read-only foreign catalogs through Lakehouse Federation | Scanned sources across Azure, other clouds, databases, and apps, with capabilities that vary by source |
| Access enforcement | Enforced at query time inside Databricks: privileges, attribute-based policies, row filters, column masks | Access policies and self-service requests for data products; data-plane enforcement only on supported sources |
| Lineage | Captured automatically for Databricks workloads, down to columns | Collected from scanned sources and pipelines; for Unity Catalog it is read from Databricks system tables |
| Business context | Comments, tags, AI-generated descriptions, and metric views that define business metrics | Governance domains, data products, glossary terms, critical data elements, OKRs |
| Data quality | Data quality monitoring on Databricks tables | Data quality rules and scores across domains and products |
| Security and compliance suite | Not a DLP, eDiscovery, or compliance-management suite | DLP, Information Protection, Insider Risk, eDiscovery, Audit, Compliance Manager |
| Primary users | Data engineers, platform teams, Databricks users | Data governance teams, stewards, security and compliance teams |
| Pricing | Included with Databricks; underlying compute billed as usual | Unified Catalog billed per governed asset per day, plus processing units for data quality |
The table shows the pattern. Unity Catalog is deep and close to the data, so it is where access is actually enforced for Databricks workloads. Purview is broad and organizational, so it is where governance teams describe data in business terms and where security and compliance controls for the Microsoft estate live.
Can You Use Unity Catalog and Purview Together?
Yes, and Microsoft documents and teaches running them together. It provides an Azure Databricks Unity Catalog source for the Purview Data Map. It connects to a Databricks SQL warehouse, authenticates with a managed identity, a service principal, or a personal access token, and supports full and incremental scans of the whole metastore or of selected catalogs. What it brings into Purview:
- Technical metadata for the metastore, catalogs, schemas, tables, and views, including columns.
- Unity Catalog tags, which appear in Purview as asset properties.
- Lineage between tables, views, and columns, read from the Unity Catalog system tables
system.access.table_lineageandsystem.access.column_lineage.
The integration is read-only in one direction: Purview learns from Unity Catalog, and Unity Catalog keeps enforcing access. Microsoft's own documentation lists the limits worth planning around:
- Lineage covers relationships captured during Databricks notebook runs. In Purview, column-level lineage appears for notebooks run on clusters, not on SQL warehouses, and lineage for external tables is not supported.
- If an external service such as an Azure Data Factory pipeline triggers the notebook, Purview does not show that dependency.
- Lineage is only complete when every workspace and asset in a flow is scanned.
- Deleting an object in Databricks does not remove the asset from Purview on the next scan, and notebook names appear as numeric IDs.
In practice this means business definitions and data products maintained in Purview describe Databricks data, but the policies that decide who can query it are still set in Unity Catalog. Teams that run both need a clear rule for which tool owns which piece.
When to Use Unity Catalog, Purview, or Both
- Unity Catalog on its own fits teams whose data and analytics live mainly in Databricks, where the priority is secure, fine-grained access and lineage for engineers and analysts working in the platform.
- Purview on its own fits organizations centered on Microsoft 365, Fabric, Azure SQL, and Power BI, where data security, sensitivity labels, and compliance requirements drive the program and Databricks plays a small role or none.
- Both is a common answer for Azure Databricks customers that also run a Microsoft-wide governance or compliance program. Unity Catalog stays the enforcement point for Databricks, and Purview scans it to give governance teams one inventory with domains and data products.
- Unity Catalog plus a cross-platform catalog fits teams that need a business glossary, ownership, and lineage across Databricks and other platforms, and whose security and compliance controls are already handled elsewhere. Unity Catalog keeps enforcing access, and the catalog gives the business one view without per-asset billing. See Dawiso as a Microsoft Purview alternative.
The deciding factor is rarely a feature checklist. It is who owns governance: a platform team that lives in Databricks, or a central governance and compliance function that lives in the Microsoft portal.
Is Unity Catalog Free? How Pricing Compares
Databricks states that Unity Catalog is included with Databricks at no additional cost, with no separate license, per-seat, or add-on fee for the catalog. What does cost money is the compute behind features that run queries or jobs, such as SQL warehouses or data quality monitoring, which is billed through normal Databricks usage. See Databricks pricing for how that usage is metered. Free inside Databricks does not mean it covers the rest of the estate: Snowflake, Fabric, and source systems still need governance of their own.
Purview data governance is billed pay-as-you-go. The Unified Catalog charges per governed asset per day, and data quality and health checks are billed in Data Governance Processing Units. The cost therefore grows with the number of assets connected to governance concepts, not with the number of users. The current rates and a worked example are in Microsoft Purview pricing, and a wider comparison is in data catalog pricing.
See it in action
Dawiso as a Microsoft Purview alternative
Runs on Azure, catalogs Microsoft Fabric in depth, and is priced per user. See how Dawiso compares with Purview.
Where Both Stop
Each tool has a clear boundary. Unity Catalog governs what runs through Databricks. Purview governs the sources it scans, and Microsoft's own source list shows how much that varies: it lists no lineage support for Amazon Redshift, SAP BW, SAP HANA, Tableau, Qlik Sense, Salesforce, or MongoDB, for example. That leaves three needs that teams with several platforms usually meet with a cross-platform catalog:
- Definitions owned by the business, for every platform. A definition of "active customer" should be the same whether the data sits in Databricks, Snowflake, or a Fabric lakehouse, with a named owner who maintains it in one shared business glossary.
- End-to-end lineage. Questions like "which Power BI reports break if this source column changes?" need data lineage that crosses from source systems through pipelines and Databricks to the report, not lineage that stops at one platform's boundary or at a source the scanner does not trace.
- Context for AI. An assistant answering from company data needs definitions, owners, and trust signals for every source it reads from, which is what a metadata layer for AI provides.
That cross-platform layer is where Dawiso works. Dawiso connects to Databricks Unity Catalog, Power BI, Microsoft Fabric, Azure Synapse, Azure Data Factory, and Snowflake, brings their metadata and lineage into one catalog, and adds the business glossary and interactive lineage on top, so Unity Catalog keeps enforcing access in Databricks while the business works from one shared view. Dawiso does not replace Purview's data security and compliance tools such as DLP or sensitivity labels. It covers the catalog and governance side, and it is priced per user rather than per governed asset.
When Unity Catalog or Purview Is Not Enough
Some teams keep both tools and add a layer on top; others replace Purview's catalog with a cross-platform one. Either way, these are the common signals, with where to read more:
- Your data no longer lives only in Databricks. Snowflake, Fabric, or operational systems sit next to the lakehouse, and Unity Catalog covers only what runs through Databricks. See What Dawiso Adds to Databricks Unity Catalog and a context layer for Databricks, or for Snowflake.
- Business users cannot find or agree on definitions. Comments and tags on tables do not settle what "revenue" means across teams. See what a business glossary is and the Dawiso business glossary.
- You need to trace a number down to the report. Lineage that ends at the platform boundary cannot say which Power BI visual a column feeds. See Power BI visual-level lineage and interactive data lineage.
- Getting access to data still takes weeks. Permissions live in Unity Catalog, but the request, approval, and ownership steps happen elsewhere. See Why Data Product Access Still Takes Weeks.
- AI assistants and agents need governed context. Agents built across several platforms need definitions and owners from all of them. See the AI context layer, Dawiso MCP, and where to build the AI agents that talk to your data.
- Purview costs grow with the governed estate. Per-asset billing rises as more data is linked to governance. See Microsoft Purview pricing, Dawiso as a Microsoft Purview alternative, and the data catalog comparison guide.
See it in action
Data Catalog Comparison Guide 2026
Compare Dawiso with Collibra, Atlan, Alation, and Secoda on features, pricing, and time to value.
Conclusion
Unity Catalog and Microsoft Purview answer different questions. Unity Catalog decides who can use which data and models in Databricks and records what happened to them. Purview inventories a wider estate, describes it in business terms, and adds Microsoft's security and compliance controls. On Azure the two are often connected, with Purview scanning Unity Catalog and Unity Catalog staying in charge of access. What a company with several platforms still has to plan for is one business-owned view of its data across all of them, built either on top of both tools or in place of the Purview catalog.
Sources
- Databricks - What is Unity Catalog?
- Databricks - What is Lakehouse Federation?
- Databricks - How much does Unity Catalog cost, and is Unity Catalog free?
- Microsoft Learn - Learn about Microsoft Purview
- Microsoft Learn - Learn about Microsoft Purview Unified Catalog
- Microsoft Learn - Connect to and manage Azure Databricks Unity Catalog in Microsoft Purview
- Microsoft Learn - Data sources that connect to Microsoft Purview Data Map
- Microsoft Learn - Data governance using Unity Catalog and Microsoft Purview (training module)
- Databricks - Unity Catalog metric views
- Microsoft Learn - Microsoft Purview data governance billing
See it in action
Data & Analytics Catalog
Create a unified view of your data assets and gain insights faster with automated data discovery.