Hi @Donita Retaw ,
Based on the information in the provided docs, Purview’s sensitivity labeling reports/insights are filterable, but there isn’t any documented way to specifically break sensitivity label counts by sub-domain (i.e., treating sub-domains as a first-class dimension the way “label name, subscription name, source type” are).
What Purview reporting supports (per the docs)
In the Sensitivity labeling insights drilldown experience (Unified Catalog / classic sensitivity labels report), you can filter using items like:
- label name
- subscription name
- source type
…and then drill into Label by source for the selected label, which shows counts by source name, source type, subscription ID, etc.
So, Purview can filter at the reporting “grid” level across the dimensions it exposes (like label and source), but sub-domain-specific filtering/counting is not described.
What this likely means for sub-domain reporting
If your “main domain + multiple sub-domains” are represented in Purview only as part of a larger identifier (for example, included in a dataset name, a URL, or some field), then Purview reports won’t automatically aggregate on “sub-domain” unless that sub-domain is already part of the dimensions the reports filter on (such as a field Purview indexes into the reportable model).
When you may need Power BI (or another approach)
The provided documentation doesn’t explicitly say “you must use Power BI,” but it does indicate that Purview’s built-in reporting filters are limited to the dimensions presented in the reports (label name, subscription name, source type, etc.). If you need a side-by-side comparison per sub-domain, that’s typically where customers either:
- model the sub-domain as a distinct attribute that ends up in a reportable dimension, or
- use Power BI for more granular breakdowns beyond what the standard report dimensions provide.
Quick sanity checks that can impact what you see
If your report doesn’t show labels you expect (even before worrying about sub-domains), the docs call out common causes like:
- assets weren’t scanned successfully
- the data source doesn’t contain classified content
- label/policy configuration prerequisites aren’t met (and the label definition is properly associated and published)
- re-run a full scan if labels are missing
Reference
Hope this helps. If you have any follow-up questions, please let me know. I would be happy to help.