Introduction
Data governance is a critical part of managing an analytics platform. As organizations adopt Microsoft Fabric, the volume of data assets grows quickly. Lakehouses, warehouses, semantic models, and reports multiply across workspaces. Without governance, it becomes difficult to distinguish authoritative data from exploratory work, and sensitive information can reach consumers who shouldn't have access.
Suppose you work at a retail analytics organization where multiple teams create analytics assets in Microsoft Fabric. The sales team builds semantic models from lakehouse data, the finance team maintains warehouse tables for quarterly reporting, and a customer insights team experiments with new data sources. Recently, a report built from an uncertified dataset led to incorrect quarterly projections shared with leadership. At the same time, an AI agent surfaced confidential salary data in a natural language query because no sensitivity labels existed on the source.
In this module, you explore how to classify data with sensitivity labels, endorse assets to signal trust, document data for discoverability, and govern your data estate for AI consumption. You learn how to use the OneLake catalog to manage governance at scale across your Fabric environment.