BI OS is the analytics layer over everything else. Because modules share a data layer, cross-department questions are answerable without a warehouse.
Building views
- ›Dashboards: assembled from widgets, per team or per person.
- ›AI chart builder: describe the chart you want and get it.
- ›Natural-language queries: ask a question in plain language and get a chart or table.
- ›Reports: scheduled or on-demand, with export.
Analytics depth
Correlation analysis, attribution, segments, executive briefings, and annotations for explaining what happened on a given date.
Governance
BI is where data governance matters most: data masking, export controls, an audit log of who queried what, row-level scoping by role, and a data catalog with lineage and a glossary. Configure these before opening BI to a wide audience.
Data prep
Transforms, a query builder, a semantic layer for consistent definitions, and incremental refresh. The semantic layer is worth the setup: it is how "revenue" means the same thing in every dashboard.
Important: Analytics inherits permissions. A dashboard cannot show someone data they could not see in the source module, which means dashboards are safe to share but may look different to different viewers.