Effective July 2024, the Power BI Admin portal Usage metrics dashboard will be removed. Comparable insights are now supported out-of-the-box through the Admin monitoring workspace (preview).
The Admin monitoring workspace provides several Power BI reports and semantic models, including the Feature Usage and Adoption report which focuses on Fabric tenant inventory and audit activity monitoring. Over the recent months, the team has made incremental improvements to the reporting provided out of the box, including adding inventory-level analysis within the Feature Usage and Adoption report. In this post, we’ll focus on the Feature Usage and Adoption semantic model which unlocks many curated insights similar to the legacy Usage metrics dashboard in the Admin Portal.
To learn more about the Admin monitoring workspace, please read the documentation page.
The legacy dashboard provides some baseline inventory and usage stats for several Power BI item types: Reports, Semantic models (formerly called Datasets), and Dashboards. We have split the legacy Admin portal Usage metrics dashboard elements into four sections.
| Section # | Insight Category | Metric | Data Source |
| 1 | Inventory | Count of items by Item type | Metadata |
| 2 | Authoring | Count of authored items by User Id | Metadata |
| 3 | Usage | Unique user count of consumed Dashboards | Audit |
| 4 | Usage | Unique user count of consumed content This is via the Get Data process so it includes not just Fabric items like semantic models, but many other data sources like content packs, files and databases. | Audit |
Please note that there was once a separate section of the Admin Portal usage metrics dashboard for “Groups”. This section is no longer supported, because we do not aggregate authoring and usage of content by Groups, whereby “Group” we mean “distribution groups and mail-enabled security”. Instead, we only show inventory and audit logs by each unique User principal name (UPN).
The Inventory page inside the Feature Usage and Adoption report provides an overview of all Fabric items. At the bottom section of the report page, we have the decomposition tree visual.
To learn more about using the decomposition tree visual, please read the documentation page. The decomposition tree visual lets users visualize and aggregate data across multiple dimensions. In this case, we will use the decomposition tree visual to effectively summarize inventory by item types and authors.
To simplify the comparison, we have cross-filtered the Inventory page by Item type = {Dataset, Report, Dashboard} in order to provide similar inventory insights to the Admin Portal Usage metrics dashboard.
One quick way to get specific information about Fabric content usage is through the ‘Explore your data’ feature in the Power BI service. The exploration can also be saved later as a report.
To learn more about using Explore your data in the Power BI service, please read the documentation page.
Following along with our previous example, if we are interested in overall dashboard consumption, we may have some questions in mind:
- Which are the most consumed dashboards (by activity counts)?
- Which are the most popular dashboards (by unique users)?
To start with, let’s add the “Item type” dimension as a filter, and only include Dashboards.
Then, let’s add the “Operation” dimension as a filter, and only include Dashboard consumption events (exclude authoring activities). Then, we are ready to add the reporting dimension “Item name”, and the corresponding measures “Active users” and “Activities”.
This provides similar consumption insights to the Admin portal Usage metrics dashboard (section 3 in Figure 1), but also explains how many consumption activities actually occur for the most popular dashboards.
We can take this analysis a bit further and look at overall Fabric consumption activities.
This quick exploration can provide some of the insights originally provided by the Admin portal Usage metrics dashboard (section 4 in Figure 1).
Please note, currently we only limit Consumed Packages to Fabric items such as semantic models. To learn more about inventory lineage and data sources, please refer to Fabric Metadata scanning overview.
This example comprises one of the multiple scenarios which organizations can now unlock as part of the out of the box reports and semantic models within the Admin monitoring workspace (preview). The experience is specifically tailored to enable a flexible data model, a holistic “tenant-level” view of inventory, usage, and other auditing scenarios. As we incrementally add content, the workspace automatically updates.
During the preview period, we’ll be listening to feedback from customers and incorporating it into platform evolution and updates to the workspace and related content.
See hereto learn about Admin monitoring workspace andhereto learn about the Feature Usage and Adoption semantic model /report.
