Prerequisites
To populate the Metadata Insights page, you must first process workbooks using the Database output type. This exports the necessary metadata to your configured MySQL, MS SQL, or PostgreSQL database.
Read-Only Users — Ask an Admin or Power User to configure a metadata source and run an export.
Global Filters
At the top of the page, a filter bar scopes the data shown across all tabs. You first choose a tool scope — Tableau or Power BI — and then the cascade for that tool appears.
| Tool scope | Cascade |
|---|---|
| Tableau | Site(s) → Project(s). Select at least one site before projects become available. |
| Power BI | Tenant → Workspace → Report. Each level activates once the one above it has a selection. |
Click Submit to apply the selected filters and refresh the data in all tabs.
Tab 1: Source Data Connection Overview
Analyze data source usage and identify consolidation opportunities. This tab is the cross-tool one: it has no Tableau/Power BI split, because its counts are drawn from the physical layer both tools share. The usage rankings are ranked by how many Tableau workbooks and Power BI reports connect to each source.
Summary KPI Cards
| Card | Description |
|---|---|
| Data Connections | Total number of data connections |
| Databases | Total number of unique source databases |
| Tables | Total number of unique source tables |
| Fields | Total number of unique source fields |
Each KPI card includes an Export button to download the underlying data to Excel.
Charts & Tables
| Component | Type | Description |
|---|---|---|
| Top 5 Data Connections | Table | Lists the 5 most frequently used data connections, showing Connection Name, Type, and Usage Count |
| Most Used Tables | Bar chart | Visualizes which tables are shared most across your content. Hovering shows the table name, database, and connection name. |
Tab 2: Calculated Fields Insights
Assess the volume and distribution of custom calculated logic, per tool.
Tableau
| Component | Type | Description |
|---|---|---|
| Calculated Fields | KPI card | Total count of calculated fields across all data sources |
| Calculated Fields by Data Source | Paginated table | Breaks down the number of calculated fields per data source. Supports server-side pagination. |
| Measures vs Dimensions | Donut chart | Shows the split of calculated fields classified as Measures versus Dimensions |
Power BI
| Component | Type | Description |
|---|---|---|
| Measures | KPI card | Total count of DAX measures across all semantic models |
| Calculated Columns | KPI card | Total count of DAX calculated columns across all semantic models |
| DAX Objects by Model | Paginated table | Breaks down measures and calculated columns per semantic model |
| Measures vs Calculated Columns | Donut chart | Shows the split between the two kinds of DAX object |
Tab 3: Dashboard Composition Stats
A quick inventory of your presentation layer, per tool.
| Tableau card | Power BI card |
|---|---|
| Workbooks | Reports |
| Dashboards | Pages |
| Worksheets | Visuals |
| Storyboards | Bookmarks |
Each card includes an Export button.
Tab 4: Optimization Indicators
Highlights potential performance risks and cleanup targets, per tool.
Tableau
| Component | Type | Description |
|---|---|---|
| Unused Fields vs Total Fields | Donut chart | Shows the proportion of fields defined but never used — candidates for cleanup |
| Unused Calculated Fields vs Total Calculated Fields | Donut chart | Highlights custom calculations that are unused — candidates for removal |
| Largest Workbooks by Component Count | Paginated table | Ranks workbooks by total component count (Dashboards + Worksheets + Storyboards) with per-workbook breakdown |
| Largest Workbooks by Data Source Count | Paginated table | Ranks workbooks by the number of distinct data sources — useful for identifying overly complex workbooks |
| Data Sources with Multiple Joins | Paginated table | Lists data sources that use multiple joins, flagging complex data models that may impact query performance |
| % of Workbooks Using Published Data Sources | Donut chart | Compares workbooks using governed (Published) data sources vs. Embedded data sources, tracking adoption of centrally managed data |
Power BI
| Component | Type | Description |
|---|---|---|
| Unused Columns vs Total Columns | Donut chart | Shows the proportion of model columns defined but never used |
| Unused Measures vs Total Measures | Donut chart | Highlights measures that are unused — candidates for removal |
| Unused Calculated Columns vs Total | Donut chart | Highlights calculated columns that are unused |
| Bidirectional Relationships | Paginated table | Model relationships that filter in both directions |
| Many-to-Many Relationships | Paginated table | Model relationships with many-to-many cardinality |
| Storage Mode Mix | Donut chart | Data sources split by how the model reads them: Import holds a copy inside the model; DirectQuery leaves the data at the source and queries it each time a report runs; Dual does both. Mixed means one source is read both ways by different tables. |
| Relationship Composition | Donut chart | Model relationships split by Active and Inactive. Inactive ones are defined but stay dormant until a specific measure switches them on for its own calculation. |
| Largest Reports by Component Count | Paginated table | Ranks reports by total component count |
All paginated tables in this tab support pagination for large datasets.
Exporting Data
Every KPI card, chart, and table includes an Export button. Clicking it downloads the underlying data as an Excel file for offline analysis or reporting. Exports respect the currently applied tool scope and its filters.
Next Steps
- Power these dashboards with metadata: Processing & Scheduling
- Run impact analysis: Lineage & Impact Analysis
- Connect a Power BI tenant: Power BI Service