Governed Tags enforce organizational metadata standards across Unity Catalog data assets. Account-level tags with built-in policies prevent inconsistent naming, unauthorized tag assignment, and incorrect tag values.
Source: official Databricks documentation · Databricks announcement
Key takeaways
- Governed tags enforce consistent data classification with restricted value sets across Unity Catalog.
- Account-level tag management creates common governance language across organizations.
- Tag permissions enable delegation: manage (create/update) vs. assign (apply tags) roles.
- Column-level tagging supports row and column masking policies for data protection.
- Certification tags mark assets as trusted or deprecated for discovery and compliance.
- Tag-based discovery enables filtering and searching data assets by classification.
- Tag inheritance simplifies governance across nested catalogs, schemas, and tables.
Chapters
- 0:00 — Introduction to Governed Tags and Account-Level Governance
- 0:32 — Use Cases: ABAC, Data Classification, Discovery, Certification
- 1:08 — Creating Governed Tags in Workspace Console
- 1:38 — Setting Permissions: Manage vs. Assign Roles
- 2:10 — Tagging Columns for Row/Column Masking Policies
- 2:40 — Tagging Tables for Discovery and Classification
- 3:11 — Certification: Mark Assets as Certified or Deprecated
- 3:59 — Conclusion: Search Filtering by Tags
Frequently asked questions
What are governed tags?
Governed tags are account-level managed tags with restricted values that enforce consistent data classification across Unity Catalog.
How do tag permissions work?
Manage permission: create/update tags; Assign permission: apply tags to assets. Separate roles enable governance delegation.
Can I use tags for masking?
Yes. Tag columns with sensitive data classifications, then build row and column masking policies based on those tags.
How do certification tags work?
Mark assets as Certified (trusted) or Deprecated for compliance, discovery, and governance workflows.
Full transcript
Auto-generated from the video and lightly edited for readability. Timestamps link to the moment in the video.
Show full transcript (11 segments)
0:04Hey folks, my name is Yousef Money and today we're going to talk about a way to govern the tax and it's basically called govern tax. Those are account level tags with built-in rules for consistency and control. When you create a govern tax, you can also define a tag policy and that policy enforces how the tags can be used. They can be applied to objects
0:26such as tables and cataloges but not compute resources like cluster and jobs which are which uses a separate tagging mechanism. Governs ensure that the tags are consistently applied and conform to the organization standards. They prevent inconsistent naming, unauthorized tag assignment or incorrect tag values. It will benefit people uh using attribute
0:48based access control, data classification, resource discovery and certification and deprecation classification. Now let's show you how you can govern them in the database workspace. Now that I am in the workspace, I need to click on catalog then manage govern tags and then I can create my govern tag. So let's create
1:10the first one collegef_mr and then the values can be name last name and city of course once you create the tag you need to govern the permissions you can select uh a group of people and you have two options either grant them the manage which allow the this group of
1:33people to modify delegate and apply the govern tag or just the assign which will allow them to use the tags you created. So for my case, I'm going to click on assign and then I'm going to create a new tag called depth MR to distinguish between assets for finance and marketing and HR. We're going to
1:57repeat the same uh uh the same steps. Select the group and which permission which is assign. And now I can click on catalog and then select my table. And over here I can I can for example select the full name which is my uh
2:19my table and I can select the PII MR and then I can can select the uh the value last name and this will benefit people using aback. So whenever I'm going to have this tag uh pi_mr, I can use a function to mask the full name for a group of people and maybe create an exception if I want still
2:42someone to view the value of this specific column. And I can also tag a table. So I can call it use for example use depth emr and this one maybe it's for marketing. It will simplify how people are are finding this table using the search bar by filtering by tags and specifying that I only want to
3:05find tables from the marketing team. And we also have a new feature called assign certification. By default, whenever you create a new data assets, the value is none. And then you can choose whether it's certified which means this table is meeting the organization standard and ready to use or deprecated which means that this table is going to be
3:27discontinued. So make sure to skip it or no longer use it. And once you select the certified you have uh you have a new emoji showing up on the top of the table. And you can also choose to duplicate it and you're going to have a different uh different sign. Of course, when you go to the search bar, once you create
3:49select the table, you can look only for certified assets. This will help the data team find the right assets. And that's it. Well, to you soon.
Official Databricks resources
- Unity Catalog documentation — docs
- Databricks Free Edition — product
- Open sourcing Unity Catalog (Databricks blog) — blog
Related
- Lakehouse RT, AI Gateway, Omnigent: Databricks Announcements
- Unity Catalog Managed Tables 2026: PM Insights at DAIS
- Catalog Commits with Unity Catalog: Multi-Table ACID Transactions
- Data governance with Unity Catalog on Databricks
- All Databricks video guides
- All tutorials & videos
Databricks Answers
Looking for a quick, source-backed answer? Browse Databricks Answers on Unity Catalog, governance & security — direct, documentation-validated answers to common Databricks questions, curated by Youssef Mrini, Senior Developer Advocate at Databricks.