BI Pixie Metrics
BI Pixie measures your BI audience with the metrics below. These tables list the main metrics in the BI Pixie Power BI app. You can find them on the report pages, some of them on more than one page, and as measures in the semantic model.
The metrics are grouped by the following categories:
- Activities
- User Engagement Metrics
- Report Metrics
- Page Metrics
- User Satisfaction Metrics
- Design Impact Metrics
- Performance Metrics
Activities
BI Pixie uses the activities below to tell active usage of Power BI reports from passive usage. For example, users that view reports without any active interactions are classified as passive users.
| Activity | Description | Activity classification |
|---|---|---|
| Page Views* | The number of times users view a report page. This activity counts as passive. Without any of the activities below, BI Pixie classifies users, reports and report sessions as passive. | Passive use of reports (Pixie Event type = Page) |
| Visual Interactions (Interactions in visuals) | The number of activities where users click on visuals that lead to interactions with other visuals and to recalculations of measures in your reports. | Active use of reports (Pixie Event type = Sense) |
| Slicer Clicks | The number of clicks in slicers that were automatically detected in the reports. Slicers include built-in or custom Slicer visual that are placed inside the report pages. | Active use of reports (Pixie Event type = Slicer) |
| Filter Clicks | The number of filter clicks (or other slicing events) that BI Pixie tracks when users filter on the columns you configured BI Pixie to track. | Active use of reports (Pixie Event type = Filter) |
| Bookmark Clicks | The number of times users click on bookmarks. BI Pixie collects also the bookmark names. | Active use of reports (Pixie Event type = Bookmark) |
| Link Clicks | The number of times users click on dynamic or static hyperlinks in the report pages. Hyperlinks are tracked in Table values and web actions in columns with conditional formatting, buttons, shapes and images (Matrix hyperlinks are supported as well). BI Pixie collects the URLs and the visual type and name from where the hyperlinks were clicked. | Active use of reports (Pixie Event type = Link) |
| Tooltip Opens | The number of times users hover on visuals elements that open report page tooltips. BI Pixie collects the name of the tooltip report pages and the selected column and value that were used to load the tooltip with specific details. | Active use of reports (Pixie Event type = Tooltip) |
| Drillthroughs | The number of times users perform drill through from selected visual elements to new report pages with details. BI Pixie collects the name of the drillthrough page and the selected column and value that were used to filter that page. | Active use of reports (Pixie Event type = Drillthru) |
User Engagement Metrics
| Metric | Description |
|---|---|
| Users | The count of unique users that are tracked. BI Pixie classifies each user as passive or engaged, so Users = Engaged Users + Passive Users. DAU, WAU and MAU count every user who opens a report, passive or engaged, as active. |
| Passive Users | The count of unique passive users. Passive users open reports and can even navigate through multiple pages of the reports, but they don't perform any activity. When users have no activity that is classified as active (as described in the Activities table above), they are counted as passive users by this measure. |
| Engaged Users | The count of unique engaged users. Engaged users view report pages and also perform at least one of the active activities in the Activities table above. The % Active Users metric is the share of engaged users among Users. |
| New Users | The count of unique users that are tracked for the first time in a given time period |
| Last-Time Users | The count of unique users that are tracked for the last time in a given time period |
| % Passive Users | The percentage of Passive Users relative to Users |
| Cumulative Users | The cumulative count of Users over time |
| Cumulative Lost Users | The cumulative count of Last-time users over time until the previous period. |
| % User Attrition | The percentage of Cumulative lost users relative to Cumulative users |
| DAU | Daily Active Users (including passive users who view report pages but don't interact with them, and including engaged users who interact with reports) |
| MAU | Monthly Active Users (including passive users who view report pages but don't interact with them, and engaged users who interact with reports) |
| WAU | Weekly Active Users (including passive users who view report pages but don't interact with them, and engaged users who interact with reports) |
| DAU/MAU | Daily active users divided by monthly active users. This ratio is an industry standard KPI to measure user stickiness. A ratio of 0.5 means that 50% of your monthly users are active on a daily basis. |
| DAU Excluding Passive | Daily Active Users (excluding passive users who view report pages but don't interact with them) |
| WAU Excluding Passive | Weekly Active Users (excluding passive users who view report pages but don't interact with them) |
| MAU Excluding Passive | Monthly Active Users (excluding passive users who view report pages but don't interact with them) |
| DAU/MAU Excluding Passive | Daily users who are active and engaged divided by monthly users who are active and engaged. Passive users are excluded from this metric. |
Report metrics
| Metric | Description |
|---|---|
| Tracked Reports | The number of reports you added Pixies to. Tracked reports = Opened Reports + Unused Reports. |
| Opened Reports | The number of reports with Pixies that have tracked activity (passive and active). Opened Reports = Interactive Reports + Passive reports. |
| Interactive Reports | The number of reports with Pixies that have tracked activities beyond page views. |
| Unused Reports | The number of reports with Pixies that have no tracked activity |
| Passive Reports | The number of reports with Pixies that have only page views but no other activity |
| % Passive Reports | The percentage of Passive Reports relative to Tracked Reports |
| % Unused Reports | The percentage of Unused Reports relative to Tracked Reports |
| Report Sessions* | The count of report sessions. A report session is a set of activities that are detected from the same user in a single report in a given time frame. BI Pixie aggregates events into a single session, as long as the user is not idle for a duration of 2 minutes. If a user is idle for more than 2 minutes, when they return to the report, BI Pixie tracks the new activity as a new session. You can change the 2 minutes timeout in Session timeout (min) parameter of the semantic model. |
| Passive Sessions* | The count of report sessions that include only page views (or navigation through pages). |
| Interactive Sessions* | The count of report sessions that includes interactions within pages (And not navigations through pages). |
| Avg Interactions* | The average number of interactions per report session (excluding page views or page navigations). |
| % Interactive Sessions* | The percentage of Interactive Sessions relative to Report Sessions |
| Avg Session Duration (Sec)* | The average duration of report sessions in seconds |
| Page Clicks | The count of times users opened report pages within report sessions. |
| Report Interactions | The count of times users interacted with reports through any of the activities in the table excluding page views or page navigation. |
*Note: In BI Pixie Power BI app, when you see this metric in a visual that aggregates data by page and not by report, a corresponding page-level metric is used for the calculation.
Page Metrics
Page metrics are page-level measurements, aggregated by pages and page sessions. BI Pixie counts page sessions on pages with the same name in two different reports as two separate sessions. A page session is a subset of a report session. BI Pixie breaks down report sessions into one or more page sessions.
| Metric | Description |
|---|---|
| Tracked Pages | The number of pages you added Pixies to. Tracked Pages = Opened Pages + Unused Pages. |
| Opened Pages | The number of pages with Pixies that have tracked activity (passive and active). Opened Pages = Interactive Pages + Passive Pages. |
| Interactive Pages | The number of pages with Pixies that have tracked activities beyond page views. |
| Unused Pages | The number of pages with Pixies that have no tracked activity |
| Passive Pages | The number of pages with Pixies that have only page views but no other activity |
| % Passive Pages | The percentage of Passive pages relative to Tracked Pages |
| % Unused Pages | The percentage of Unused pages relative to Tracked Pages |
| Page Sessions* | The count of page sessions. A page session is a subset of a report session. BI Pixie breaks down report sessions into page sessions when the user navigates between pages. |
| Passive Page Sessions* | The count of page sessions that include only page views. |
| Interactive Page Sessions* | The count of page sessions that includes interactions. |
| Avg Interactions* | The average number of interactions per page session (excluding page opens). |
| % Interactive Sessions* | The percentage of Interactive Sessions relative to Page Sessions |
| Avg Session Duration (Sec)* | The average duration of page sessions in seconds |
| Page Interactions | The count of times users interacted in the given page session through any of the activities in the Activities table above and beyond page views or page clicks. |
User Satisfaction Metrics
With BI Pixie you can collect user feedback from your BI audience and analyze their satisfaction. To collect feedback, BI Pixie adds a feedback control to your target reports, as described in Collect user feedback.
| Metric | Description |
|---|---|
| Feedback Clicks | The number of clicks in the Feedback control. |
| Positive Clicks | The number of clicks in the positive button of the Feedback control. |
| Negative Clicks | The number of clicks in the negative button of the Feedback control. |
| Satisfied Users | The number of users whose last click was positive in the selected reports |
| Dissatisfied Users | The number of users whose last click was negative in the selected reports |
| CSAT (Last Response) | The count of positive responses (last response per user and report) divided by all last feedback responses in the specified time period. |
| CSAT (Multiple Responses) | The count of all positive responses, including multiple responses per user, divided by all feedback responses in the specified time period. |
| Average Satisfaction | A number between -1 (negative) and 1 (positive) that is the average of all non-neutral feedback clicks. 0 means that the number of negative clicks equals to the number of positive clicks. |
| Respondents | The count of users who provided feedback, including neutral feedback. |
| % Feedback Responses | The percentage of feedback providers relative to all report consumers. |
Design Impact Metrics
You can analyze the effectiveness of your report design with three scores that estimate report complexity, usability and passivity.
Complexity score
An estimate of the report's complexity at the data visualization level as BI Pixie detected it during its last run. Higher score means that the reports may have excessive number of visuals or objects that are expensive to maintain.
Usability score
An estimate of the report's usability which is based on the detection of controls such as bookmarks, slicers, tooltips and drill-throughs. A higher score usually means a more user-friendly report.
Passivity score
An estimate of how much table and matrix visuals dominate the report, which tends to mean passive consumption. A higher score means people are more likely to consume the report passively, or to use it as a data export tool.
| Metric | Description |
|---|---|
| # of Visuals | The number of visuals found during the last BI Pixie run. |
| Visuals per Page | The average number of visuals (Best used at the report or project/workspace level). |
| # of Bookmarks | The number of bookmarks found during the last BI Pixie run at the report level. |
| # of Slicers | The number of slicers found during the last BI Pixie run at the page level. |
| # of Table Visuals | The number of Table and Matrix visuals found during the last BI Pixie run at the page level. |
| # of Columns in Table Visuals | The number of columns used in Table and Matrix visuals as found during the last BI Pixie run. |
| # of Objects | The number of buttons, images and shapes found during the last BI Pixie run. |
| # of Tooltip Pages | The number of Tooltip pages found during the last BI Pixie run. |
| # of Drillthru Pages | The number of Drillthrough pages found during the last BI Pixie run. |
| Complexity Score | An estimate of the report's complexity. Higher score means excessive visuals or objects. |
| Usability Score | An estimate of the report's usability based on controls detection. |
| Passivity Score | An estimate for the dominance of table and matrix visuals. |
Performance Metrics
You can measure the performance of your Power BI reports beside their usage by connecting BI Pixie to Azure Log Analytics.
| Metric | Description |
|---|---|
| Duration | Total duration of the execution in seconds. Provided as aggregations of 'durationMs' metric from Power BI in Azure Log Analytics. Aggregations: Sum, Average, P50, P75, P90. |
| Total CPU Time | Total CPU time of the request in seconds. Provided as aggregations of 'totalCpuTimeMs' metric. Aggregations: Sum, Average, P50, P75, P90. |
| Query Processing CPU Time | Total CPU time spent by tasks on Analysis Services query thread pool thread. Aggregations of 'queryProcessingCpuTimeMs'. |
| Vertipaq Job CPU Time | Total CPU time spent by Vertipaq engine in seconds. Aggregations of 'vertipaqJobCpuTimeMs'. |
| Execution Delay | Total time spent waiting for Analysis Services engine thread pool thread availability. Aggregations of 'executionDelayMs'. |
| Query Result Rows | Total number of rows returned as a result of the DAX query. Aggregations of 'queryResultRows'. |
| Performance Log Count | The number of load events per visual and other dimensions. Used as the weight in weighted average calculations. |
When a metric above uses the SUM function, BI Pixie adds up its aggregations with the DAX SUM function.
When a metric above uses AVG, P50, P75 or P90, BI Pixie combines its aggregations as a weighted average, with the count of visual load events as the weight.