1.6 — Apply Sensitivity Labels and Endorse Items
DP-700 Domain 1 · Skill 1.6
What You Need to Know
Labels and endorsement are both governance metadata on Fabric items. They answer different questions. A sensitivity label says how confidential the content is and can protect it. An endorsement says how trustworthy it is.
Sensitivity labels (Microsoft Purview Information Protection)
| Fact | Detail |
|---|---|
| Where labels come from | Defined in Microsoft Purview. A Fabric admin enables Allow users to apply sensitivity labels for content |
| Who can apply one | Any user with edit permission on the item (plus a Pro or PPU license) who is allowed to use that label |
| Where to apply | From the sensitivity indicator in the item header, or in the item's settings |
| Downstream inheritance | A label on a lakehouse or semantic model can flow to the items built on it, such as models, reports and dashboards |
| Export protection | A protected label travels with the data exported to Excel, PowerPoint or PDF |
| Policies | Default labels, mandatory labeling and data loss prevention (DLP) policies can enforce labeling |
When an item is first deployed through a deployment pipeline, its label is copied with it. Later deployments copy the label only in specific cases, for example a protected source label over an unlabeled target.
Endorsement
| Badge | Meaning | Who can apply it | Which items |
|---|---|---|---|
| Promoted | The owner thinks the item is ready to share and reuse | Anyone with write permission on the item | All Fabric and Power BI items except dashboards |
| Certified | An authorized reviewer confirms it meets the organization's quality standards | Only users or groups a Fabric admin names (this can be delegated per domain) | All items except dashboards |
| Master data | The item holds core, authoritative data (a single source of truth, such as a customer or product list) | Only users a Fabric admin names | Only items that contain data, such as lakehouses, warehouses and semantic models |
- Certification and master data must be enabled by a Fabric admin before anyone can use them.
- Anyone can request certification. Only the named reviewers can grant it.
- An item has one endorsement badge at a time.
- Endorsed items are listed first in the OneLake catalog and in search, so people find trusted data sooner.
Matching scenarios to badges
| Scenario | Badge |
|---|---|
| A lakehouse that holds the official customer records everyone should use | Master data |
| A semantic model a reviewer checked against company standards | Certified |
| A report its author thinks is ready for colleagues to reuse | Promoted |
| A dashboard for executives | None (dashboards can't be endorsed) |
Key Concept: Labels protect (how sensitive), endorsement guides (how trustworthy). Dashboards are the exception to endorsement. Only data items can be master data.
Exam Traps
- Trying to endorse a Power BI dashboard
- Marking a report as master data (master data is only for items that contain data)
- Letting any item owner certify (only Fabric-admin-approved reviewers can)
- Expecting an endorsement badge to grant access or protect data (it's discovery metadata only)
Check your understanding
Practice Scenario
The data governance team wants the lakehouse that holds the company's official product catalog flagged in the OneLake catalog as the authoritative source for product data. Which endorsement should it get, and who can apply it?
Log in to save your progress.
Sources
- Endorsement overview, Microsoft Learn
- Endorse Fabric and Power BI items, Microsoft Learn
- Apply sensitivity labels to Fabric items, Microsoft Learn
- Information protection in Microsoft Fabric, Microsoft Learn