Usage metrics tell you a report was opened. Not what happened next.
You opened the usage metrics report to answer a reasonable question: is this dashboard working? You got view counts and viewer names.
That tells you the report was opened. It does not tell you whether anyone used it. A user who opens a dashboard every morning, glances at it for five seconds, and moves on looks identical to one who filters, drills, and makes a decision.
Why this is hard today
Usage metrics were built to answer who opened what, and they do that well. They count views and viewers, and they keep a limited window of history.
The things that separate a working report from an ignored one all happen after the page loads: which visual someone clicked, what they filtered to, whether they drilled through, how long they stayed. None of that is a page view, so none of it is counted. Neither are embedded or publish-to-web reports, which usage metrics do not cover at all.
What BI Pixie shows you
BI Pixie adds invisible native Power BI elements, called Pixies, that detect interactions and send them as events. It runs alongside usage metrics rather than replacing them.
- Every interaction, not just the open
- Clicks on visuals, slicers, buttons, images, shapes, bookmarks, and hyperlinks in tables, plus filter changes, drill-through actions, and tooltip views.
- What data people selected
- The values users clicked and filtered to, so you can see which dimensions actually drive the questions being asked.
- Active versus passive users
- Interactions per session and session duration separate people working in a report from people merely opening it.
- Satisfaction alongside behaviour
- Feedback controls and embedded surveys let you pair what people did with what they thought.
- Anywhere the report runs
- The Power BI service, Desktop, mobile, Report Server, Power BI Embedded, and publish to web.
What it takes to set up
- 1 Connect your Power BI workspace and pick the reports you care about.
- 2 Add Pixies. The reports look and behave exactly as before.
- 3 Let people use them as normal.
- 4 Analyze engagement, adoption, and satisfaction in the BI Pixie Dashboard.
Worth knowing: No personally identifiable information is collected by default. User identity capture is an explicit setting you turn on in tracking setup if you want to analyze behaviour per person.