Data cataloging is the discipline of creating an organized, searchable inventory of all data assets within an organization. Without a catalog, data consumers spend significant time hunting for the right datasets, duplicating existing work, or unknowingly using stale or non-compliant data. In enterprise environments where hundreds of teams produce thousands of artifacts, a well-maintained catalog becomes the foundation of effective data management.

Microsoft Fabric addresses this challenge through a dual-catalog model that operates at two complementary scopes. OneLake Catalog is Fabric's built-in catalog — deeply integrated into the Fabric experience, optimized for day-to-day discovery and governance of OneLake assets. Microsoft Purview is the enterprise-wide governance platform that spans the entire data estate: Microsoft 365, Azure services, Fabric, on-premises systems, and even third-party clouds like AWS and Snowflake.

These two catalogs are not competitors — they are layers in a unified governance architecture. OneLake Catalog provides the fast, contextual experience that data engineers and analysts need within Fabric. Purview provides the cross-estate visibility, compliance controls, and regulatory-grade governance that IT and compliance teams require. Together, they ensure that data is discoverable, trustworthy, and protected at every level.

The key pillars of effective data cataloging in Fabric are:

OneLake Catalog is the native data catalog experience within Microsoft Fabric. It provides a centralized interface for discovering, governing, and securing all data assets stored in OneLake — including lakehouses, warehouses, semantic models, pipelines, notebooks, and reports. Unlike external catalog solutions that require separate configuration and synchronization, OneLake Catalog is deeply embedded in the Fabric platform and reflects the current state of your data estate in real time.

The catalog is organized around three primary tabs, each addressing a distinct aspect of data management:

Explore Tab

The Explore tab is the primary discovery interface for data consumers. It allows users to browse all accessible data assets across workspaces, domains, and item types. Users can filter by metadata attributes, sensitivity labels, endorsement status, owner, and freshness. The search experience supports natural language queries and surfaces results based on relevance, usage patterns, and the user's permissions.

Key capabilities include cross-workspace browsing with domain-based organization, metadata-driven filtering (item type, owner, sensitivity label, endorsement), lineage visualization showing upstream and downstream dependencies, and permission-aware results that respect workspace and item-level security. Users see only what they have access to — the catalog never exposes assets beyond a user's permissions.

Govern Tab

The Govern tab provides governance insights and recommended actions for catalog administrators and data stewards. It surfaces governance reports that highlight gaps in metadata quality, ownership assignments, sensitivity labeling, and endorsement status. The actionable governance experience presents specific recommendations — such as "42 datasets lack descriptions" or "15 items have no assigned owner" — and enables bulk remediation directly from the interface.

Policy compliance monitoring tracks whether assets meet organizational governance requirements. Administrators can see which domains are well-governed and which need attention, track trends over time, and identify areas where governance debt is accumulating. The Govern tab transforms governance from a periodic audit exercise into a continuous, observable process.

Secure Tab

The Secure tab provides a unified view for managing security across OneLake assets. It consolidates workspace roles, item permissions, row-level security, and sensitivity labels into a single pane. Administrators can audit who has access to specific assets, review permission inheritance, and identify over-permissioned users or groups.

This tab is particularly valuable for access reviews and compliance audits. Rather than navigating to each workspace individually, security teams can assess the access landscape from a centralized view, identify anomalies, and take corrective action.

Catalog API

The OneLake Catalog exposes a REST API for programmatic metadata discovery. This enables automation scenarios such as custom governance dashboards, integration with external ITSM tools, automated metadata validation pipelines, and bulk operations on catalog assets. The API supports querying across workspaces, retrieving lineage information, and reading governance metadata.

GET https://api.fabric.microsoft.com/v1/admin/items
    ?workspaceId={workspace-id}
    &type=Lakehouse

Authorization: Bearer {token}
Content-Type: application/json

The catalog also integrates with Microsoft Teams (sharing and discovering assets in context), Excel (connecting to endorsed datasets), and Copilot (natural language questions about your data estate).

Microsoft Purview is the enterprise-wide data governance platform that provides a unified view of your entire data estate — spanning Microsoft 365, Azure services, Microsoft Fabric, on-premises systems, and third-party clouds including AWS, Google Cloud, Snowflake, and Databricks. While OneLake Catalog focuses on the Fabric experience, Purview operates at the organizational level, delivering the cross-estate visibility and compliance controls that large enterprises require.

Purview's Unified Data Catalog serves as the authoritative inventory of all data assets across the organization. It automatically discovers and classifies data wherever it resides, maintains a living data map, and enables governance teams to apply consistent policies across heterogeneous environments. For organizations using Microsoft Fabric, Purview extends governance beyond OneLake to encompass the full data supply chain.

The platform provides AI-aware security capabilities that understand how data flows through AI systems, ensuring that sensitive information is protected even when consumed by machine learning models, copilots, and automated processes. Cross-cloud coverage means that governance policies defined in Purview apply regardless of where data is stored or processed.

OneLake Catalog and Microsoft Purview serve different but complementary roles in the governance architecture. The following comparison highlights where each excels and helps you understand when to use which tool for specific governance tasks.

The key takeaway is that these tools operate at different scales. OneLake Catalog is your "inner loop" — fast, contextual, and optimized for the Fabric user experience. Purview is your "outer loop" — comprehensive, policy-driven, and designed for enterprise-wide governance and compliance.

Who Uses What? Persona Guide

Different roles interact with OneLake Catalog and Purview in different ways. Use this guide to understand which tool each persona should focus on and why.

OneLake Catalog and Microsoft Purview are designed to work as an integrated governance stack. Fabric assets discovered and managed through OneLake Catalog are automatically surfaced in Purview's Unified Data Catalog, creating a seamless bridge between Fabric-local governance and enterprise-wide compliance.

Metadata Synchronization

When data assets are created or modified in Fabric, OneLake Catalog captures their metadata — item type, workspace, owner, description, sensitivity label, endorsement status, and lineage relationships. This metadata is automatically synchronized to Microsoft Purview, where it joins the broader data map alongside assets from Azure SQL, Synapse, Databricks, S3, and other sources. Changes flow bidirectionally: sensitivity labels applied in Purview propagate back into Fabric, and metadata updates in Fabric are reflected in Purview.

Federated Governance Model

The integration enables a federated governance approach. OneLake Catalog handles local management — domain owners govern their assets within Fabric, apply endorsements, manage descriptions, and control workspace access. Purview provides global compliance — enterprise policies, cross-tenant controls, regulatory frameworks, and organization-wide sensitivity classifications are defined centrally and enforced everywhere.

This model respects the principle of subsidiarity: decisions are made at the most local level possible, with central oversight ensuring consistency and compliance. Domain teams have autonomy within guardrails established by the governance office.

Sensitivity Labels & DLP

Sensitivity labels defined in Microsoft Purview are enforced within Fabric. When a label is applied to a lakehouse or semantic model, it automatically propagates to downstream artifacts — reports, exports, and shared links all inherit the protection. Data loss prevention (DLP) policies configured at the enterprise level in Purview are visible and enforced within Fabric, preventing unauthorized sharing of sensitive data.

End-to-End Lineage

Lineage tracing spans the entire data lifecycle. Purview captures lineage from external source systems, through Fabric pipelines and transformations, all the way to Power BI reports consumed by business users. This cross-system lineage enables impact analysis (what happens if this source schema changes?), root-cause investigation (where did this data quality issue originate?), and compliance auditing (can we prove the provenance of this regulatory report?).

Data Quality & Governance Policies

Data quality rules and governance policies defined at the enterprise level in Purview are visible within Fabric's OneLake Catalog. Data stewards can see which assets meet quality thresholds, which are flagged for remediation, and how their domain compares to organizational standards — all without leaving the Fabric experience.

Integration Flow

┌─────────────────────────────────────────────────────────────────────┐
│                    DATA GOVERNANCE FLOW                              │
├─────────────────────────────────────────────────────────────────────┤
│                                                                     │
│  Data Sources ──→ OneLake ──→ OneLake Catalog ──→ Purview Unified  │
│  (SQL, APIs,      (Storage)   (Fabric-local       Catalog           │
│   Files, S3)                   governance)        (Enterprise-wide) │
│                                                                     │
│  ┌──────────────────────────────────────────────────────────────┐  │
│  │  POLICIES FLOW BACK DOWN:                                     │  │
│  │  Purview → Sensitivity Labels → OneLake Catalog → Fabric     │  │
│  │  Purview → DLP Policies → OneLake → Reports & Exports       │  │
│  │  Purview → Data Quality Rules → Visible in Fabric UI         │  │
│  └──────────────────────────────────────────────────────────────┘  │
│                                                                     │
└─────────────────────────────────────────────────────────────────────┘

This bidirectional integration means that governance is not a one-time setup but a living process. As data assets evolve, the catalog stays current. As policies are updated centrally, enforcement propagates automatically. The result is a governance posture that scales with the organization without requiring manual synchronization.

Implementing data cataloging and governance is a journey, not a one-time project. The following phased checklist provides a structured path from initial setup to mature, organization-wide governance. Each phase builds on the previous one — complete the foundation before advancing to classification and policy enforcement.

Phase 1: Foundation

• Enable OneLake Catalog for your Fabric tenant
• Define and configure business domains (align with organizational structure)
• Assign domain owners with clear accountability for governance within their scope
• Set baseline metadata requirements — every asset must have an owner, description, and domain assignment
• Establish naming conventions for workspaces, lakehouses, and data items
• Document governance roles and responsibilities (data steward, domain owner, platform admin)

Phase 2: Classification & Labeling

• Connect Microsoft Purview to your Fabric tenant
• Configure sensitivity labels aligned with your organization's data classification policy
• Set up automated classification rules for PII, financial data, and regulated information
• Define business glossary terms for key metrics, dimensions, and business concepts
• Link glossary terms to technical assets in both OneLake Catalog and Purview
• Train data producers on labeling responsibilities and classification criteria

Phase 3: Governance & Policy

• Establish endorsement criteria — define what "Promoted" and "Certified" mean in your organization
• Configure DLP policies to prevent unauthorized sharing of sensitive data
• Set up governance monitoring dashboards tracking ownership, labeling, and metadata coverage
• Define data quality rules for critical datasets (freshness thresholds, completeness checks)
• Implement access review processes — quarterly reviews of who can access what
• Create a governance council with representatives from each business domain

Phase 4: Scale & Optimize

• Roll out governance domain by domain, starting with highest-value areas
• Enable Catalog API integrations for custom governance automation
• Connect external data sources to Purview for cross-estate lineage and discovery
• Measure and optimize catalog health KPIs (discovery time, metadata coverage, compliance rate)
• Implement Git-based version control for all Fabric artifacts
• Establish continuous improvement cycles — review governance effectiveness quarterly
• Explore advanced scenarios: data contracts, data products marketplace, automated quality gates