Skip to main content
purview to fabricpurview data governance fabriconelake catalogpurview data governance retirementmicrosoft fabricmicrosoft purview

Purview to Fabric: Mapping Data Governance to the OneLake Catalog

Purview to Fabric means moving data governance from Microsoft Purview to the OneLake catalog in Microsoft Fabric, one of the paths Microsoft offers Purview Data Governance customers before end of support on November 12, 2027. It works well for data that lives in Fabric. Microsoft's own mapping, though, shows that much of what Purview customers built, such as the business glossary, data products and data quality rules, has only a partial Fabric equivalent or none.

So before mapping objects, decide what the catalog is for. A data catalog pays off when the business actually uses it: to find data, agree on definitions, see who owns what and request access. A catalog chosen because it comes with the platform can cover the technical side and still leave that work undone. This page walks through Microsoft's mapping object by object, the gaps, the limits of Fabric Policies and the five-step migration Microsoft recommends, and ends with how to choose a catalog people will use. The wider decision is covered in Microsoft Purview Data Governance is retiring, and the catalog itself in What is the OneLake catalog?

TL;DR

The OneLake catalog is one of the paths Microsoft offers Purview Data Governance customers, and it fits data that lives in Fabric. In Microsoft's transition guidance, discovery, domains, lineage and governance health move to the OneLake catalog for Fabric items. The glossary and critical data elements map only to tags, descriptions and endorsement, data products need their contract and lifecycle defined elsewhere, data quality rules have no rules engine, and external metadata scanning has no equivalent: "data connectivity is not enterprise metadata scanning". Fabric Policies are in preview with three policy types. Labels, DLP and audit stay in Purview. Microsoft recommends moving in waves, not copying every Purview object. For a governance program the business uses every day, or one that will grow beyond Fabric, plan for a dedicated data catalog next to Fabric.

What moves from Purview to Fabric?

Microsoft's announcement calls Fabric its data governance solution and the OneLake catalog its foundation, included in Fabric capacity. The retirement covers the Unified Catalog, Data Map, Workflows and classic Azure Purview, per the November 2026 partner notice. Microsoft's architects then published transition guidance on November 13, 2026, with a rule of thumb: the OneLake catalog to discover and govern Fabric data, OneLake security and Fabric controls to enforce access, and Purview to classify, protect, monitor, investigate and prove compliance.

The OneLake catalog has three experiences, documented on Microsoft Learn: Explore to find Fabric items, Govern for governance insights, recommended actions and policies, and Secure for workspace roles and OneLake security roles. Explore covers Fabric items, which in Microsoft's words is not automatically an inventory of every external source.

Purview to Fabric mapping table

The table follows Microsoft's transition guidance. The status column is our reading of it: covered where Fabric has a documented target for Fabric items, partial where Microsoft's guidance says part of the outcome needs another method, not covered where it names no Fabric target.

Purview capabilityFabric target todayStatusMicrosoft's guidance
Catalog search and browseOneLake catalog Explore, Catalog Search REST APICovered for Fabric itemsMove Fabric discovery now, keep separate discovery for sources outside Fabric
Collections, governance domains, ownershipFabric domains and subdomains, domain admins, item ownersPartialKeep a responsibility matrix for stewardship duties with no dedicated role
Glossary, critical data elements, custom contextTags, descriptions, endorsementPartialExport and preserve terms and ownership; do not claim full glossary or CDE parity
Data productsDomains, endorsement, sharing, discoverable semantic modelsPartialDefine the product contract, owner, health, lifecycle and access process outside the catalog where needed
Lineage and impact analysisFabric lineage and impact analysis for supported itemsPartialKeep another method for cross-system, unsupported or granular lineage
Governance healthGovernance insights and recommended actionsCovered for Fabric itemsKeep business KPIs and issue closure in the owning systems
Data qualityFreshness and curation insights, workload toolsNot covered as rulesGovernance health is not a rules, profiling and scoring engine
Access requests and workflowsItem sharing, semantic model access requestsPartialKeep request, approval, expiry, revocation and evidence where Fabric does not cover them end to end
Classification and sensitive data protectionPurview sensitivity labels and DLP on supported Fabric itemsStays in PurviewValidate item types, inheritance, export behavior and licensing
Access policies and enforcementFabric Policies (preview), workspace, item, OneLake and source permissionsPartialDo not assume universal policy coverage
External metadata scanningShortcuts, mirroring, external sharing, scanner and admin APIsNot coveredData connectivity is not enterprise metadata scanning; define a bridge or partner approach
APIs and integrationsCatalog Search REST API, Fabric metadata and scanner APIsPartialDo not assume compatibility with existing Purview API integrations
Audit, risk and compliancePurview Audit, DLP, Insider Risk Management, Fabric evidenceStays in PurviewCorrelate evidence across identity, data, action and retention
AI and agent governanceOneLake controls, Purview AI controls, Fabric semantic capabilitiesPartialPilot one end-to-end AI path and verify grounding, permissions and evidence
Purview to Fabric at a Glance PURVIEW TO FABRIC AT A GLANCE COVERED For Fabric items Search and discovery Governance health PARTIAL Part of the outcome Domains and ownership Glossary and CDEs Data products Lineage Access requests Policies, APIs, AI NOT COVERED No Fabric target named Data quality rules External metadata scanning STAYS IN PURVIEW Security and compliance Sensitivity labels Data loss prevention Audit, Insider Risk Our grouping of Microsoft's transition guidance, November 2026
Click to enlarge

What has no Fabric equivalent?

Four gaps decide most migration plans, and Microsoft names them itself: external metadata scanning, glossary and critical data element depth, data quality workflows and broad policy coverage.

  • Sources outside Fabric. Purview's Data Map scanned Snowflake, Oracle, SAP, Amazon S3 and many more. Fabric reaches them through shortcuts, mirroring and connectors, which move or expose data, not metadata about it. A Databricks or SAP system that is not brought into OneLake is not governed in the OneLake catalog.
  • Business glossary and critical data elements. Terms become tags, descriptions and endorsement on Fabric items. Owners, synonyms, hierarchy and custom attributes have no target, see Purview glossary migration.
  • Data quality rules. Govern reports freshness and curation, and Microsoft says it is not a rules, profiling and scoring engine, so Purview's rules and scores need another home.
  • Broad policy coverage. Fabric Policies are in preview with three policy types, and Microsoft's advice is not to assume universal policy coverage. Enforcement happens per data path, through workspace, item, OneLake and source permissions.

Workflows and stewardship come up in most plans too. Approvals and access request flows are covered only where Fabric handles them end to end, and the data steward role has no Fabric equivalent, see Purview role migration.

Microsoft's Govern documentation adds limitations worth checking before you plan: insights do not cover subitems such as tables, cross-tenant scenarios and guest users are not supported, and Govern is not available when Private Link is enabled.

Fabric Policies limits

Fabric Policies replace part of what Purview access policies did. They are managed from the Policies center in Govern, logged to audit, available through REST, Git and CI/CD, and included with the Fabric license. The transition guidance lists their limits in preview:

  • The only effect is Allow, so a policy works as an allow list.
  • Three policies are supported: external data sharing and workspace settings editing at tenant scope, item creation at capacity scope.
  • One active policy per scope, with up to 100 rules at tenant scope and 50 at capacity scope.
  • Changes take effect within about 15 minutes.
  • Not supported in the West Europe, North Europe or West US capacity regions.

The region limit matters for European customers in particular. Check where your capacities run before you count on Fabric Policies.

See it in action

Dawiso as a Microsoft Purview alternative

A data catalog across Microsoft Fabric and the platforms next to it, with a business glossary the business actually uses.

How to migrate from Purview to Fabric in waves

Microsoft's guidance is to move in waves and not to copy every Purview object. Its five steps:

  1. Inventory. Registered sources, scans and runtimes, collections and domains, owners and stewards, glossary terms and critical data elements, custom attributes, data products, lineage, quality rules, workflows, access policies, APIs, reports, licenses and evidence retention.
  2. Classify every requirement. Move now to a documented Fabric capability, coexist temporarily, bridge with a process or a partner, or redesign.
  3. Design the target operating model. Domains and subdomains, workspace and capacity layout, delegated admins, item owners, certifiers, tag taxonomy, OneLake roles, Fabric Policies and the handoff to Purview for security and compliance.
  4. Pilot one end-to-end scenario. One domain, one critical asset, one producer, one consumer, one denied identity and one analytics or AI use case. Prove that access is blocked where it should be, not only that it works.
  5. Migrate in waves and verify. Move supported metadata and ownership, train users, run old and new processes in parallel only as long as needed, and retire each Purview dependency once documented acceptance criteria pass.

The hand-built content still has to be exported first, because Microsoft names no migration tool. What can be exported, and how, is in Is there a Microsoft Purview migration tool?

What stays in Purview after the move to Fabric

Microsoft Purview stays as the data security and compliance product. For Fabric, the transition guidance lists sensitivity labels on Fabric items and DLP on lakehouses, warehouses, databases and semantic models, both requiring Purview licensing, plus Fabric activities in the Purview audit log, Insider Risk Management indicators for Power BI and lakehouse activity, and AI governance for supported Fabric Copilots and agents. None of this needs to be migrated. The security team keeps working where it works today.

When Fabric is not enough

For a small estate that lives in Fabric and Power BI, with a governance program run by a few data engineers, the OneLake catalog can carry much of the load, and it costs nothing beyond Fabric capacity compared with Purview's per-asset billing. Fabric is not enough when the data also lives in Databricks, Snowflake, SAP or on-premises systems, or when the governance program runs on a business glossary, data products with owners and data quality rules. For sources outside Fabric, Microsoft's own guidance calls for a bridge or partner approach.

How to choose a catalog people will use

Choose the catalog for the way your organization will use it, not for the platform it comes with. A governance program at scale lives or dies on adoption by the business, and the move away from Purview is the moment to get that right, not to repeat a migration in two years.

QuestionOneLake catalog aloneA data catalog such as Dawiso
Who uses it every day?Fabric users and data engineersBusiness users, stewards and data teams, searching in their own words
Where do definitions live?Tags and descriptions on Fabric itemsA business glossary with owners, synonyms and approval workflows
What does it cover?Fabric items, and data shortcut or mirrored into OneLakeFabric together with Databricks, Snowflake, SAP and other connected sources
What happens as you grow?New platforms outside Fabric need another tool laterNew platforms join the same catalog through connectors
Who owns the data product?Contract, lifecycle and access defined outside the catalogData products with an owner, a lifecycle and access requests in one place

If you run governance at scale, or plan to, the safer choice is a catalog that does not tie your definitions to one platform's roadmap. Purview customers have already moved once, from Azure Data Catalog, and are moving again. Definitions, owners and approvals that live in a platform-independent catalog do not have to move the next time a vendor changes its strategy.

Dawiso works as that layer. It catalogs Microsoft Fabric and Power BI together with Databricks, Snowflake and other platforms, adds a business glossary, data products and end-to-end lineage on top, and leaves labels and DLP in Purview. It is available on Microsoft Marketplace and priced per user. See the Microsoft Purview alternative page or book a demo.

Sources

See it in action

Moving off Purview Data Governance?

Dawiso catalogs Microsoft Fabric together with Databricks, Snowflake, SAP and more, with a business glossary on top.

A cookie a day keeps bad UX away.

We use cookies to personalize content, ads and to analyze our traffic. We also share information about your use of our site with our advertising and analytics partners who may combine it with other information that you've provided to them or that they've collected from your use of their services. By clicking "Accept All", you allow us to use cookies for analytics and ads via Google Tag Manager. You can also customize cookies.

Customize Consent Preferences

We use cookies to personalize content, ads and to analyze our traffic. We also share information about your use of our site with our advertising and analytics partners. Privacy Policy

Necessary cookies allow core website functionality such as user login and account management. The website cannot be used properly without strictly necessary cookies.

Functionality cookies are used to remember visitor information on the website, eg. language, timezone, enhanced content.

Analytics cookies are used to see how visitors use the website, eg. analytics cookies. Those cookies cannot be used to directly identify a certain visitor.

We use Microsoft Clarity to see how you use our website (including heatmaps and session replays) so we can improve it. By using our site, you agree that we and Microsoft can collect and use this data. See our Privacy Policy for details.

Advertisement cookies are used to identify visitors between different websites, eg. content partners, banner networks. Those cookies may be used by companies to build a profile of visitor interests or show relevant ads on other websites.