Issue Currently, when fields or measures are added to the X-axis or Y-axis in Power BI visuals, they are automatically included in the tooltip. This can become frustrating for report authors, particularly in cases where additional measures are used strictly for visual structuring or analytical purposes, and are not meant to be shown to end users in the tooltip. Example   The automatic inclusion of these fields can lead to cluttered and confusing tooltips, reducing the overall clarity and effectiveness of the visual. As a workaround, authors are often forced to disable tooltips entirely, which negatively impacts accessibility and removes useful contextual information from the report. Possible Function To enhance the clarity and usability of tooltips, it would be highly beneficial to introduce a feature that allows report authors to toggle the visibility of specific axis fields within the tooltip. Benefits This enhancement would enable cleaner and more focused tooltips, improve the end-user experience, and give developers greater control over how information is presented in a report. ... View more

Dear Microsoft Power BI Team, I would like to propose several enhancements to the PowerBI.com monitoring page to improve visibility, monitoring, and operational efficiency for semantic model refreshes and deployments. Current Challenges: Limited Visibility: The latest status of all semantic model refreshes and deployments is not easily discernible. The current page only shows a history of succeeded and failed refreshes, which makes it difficult to quickly understand the overall health of an instance. Restrictive Permissions: Viewing semantic model refresh statuses currently requires a high-level Contributor role. This restricts DEV/QA teams from monitoring refresh health in non-production environments unless they are granted permissions that may be unnecessarily broad. Proposed Solutions: Comprehensive Monitoring Dashboard: Introduce a new Monitoring view that lists all semantic models in a workspace or tenant, with filters for environment or model name. For each model, display: The latest refresh status (success or failure) A quick visual summary (e.g., green/red dots) representing the status of the last 4–5 refreshes to help identify systemic issues The latest deployment status Role-Based Access Improvements & New Role: Allow users with lower, read-only (R/O) roles to view this Monitoring dashboard. This enables DEV and QA teams to monitor refresh health without requiring Contributor permissions. Introduce a new role: Semantic Model Refresh Orchestrator. This role would have a single permission: the ability to trigger semantic model refreshes. This empowers DEV/QA teams to rerun failed refreshes in lower environments and allows Production Support to rerun refreshes in production, without needing full Contributor or Admin rights. Action buttons (e.g., "Rerun Semantic Model Refresh") should be available only to users with Contributor or Semantic Model Refresh Orchestrator permissions. Improved Issue Detection: The dashboard should visually summarize recent refreshes (e.g., green/red dots for last 4–5 refreshes), making it easier to spot recurring or systemic problems with specific semantic models. Alert Subscriptions for Lower Roles: Enable users with access to the Monitoring dashboard (including R/O and Orchestrator roles) to set up email alerts for failed semantic model refreshes, configurable per model via checkboxes. This allows teams to proactively monitor and respond to issues. Benefits: Improved Operational Efficiency: Teams can quickly assess the health of all semantic models and respond to issues without unnecessary permission elevation. Better Issue Detection: Visual history of recent refreshes makes it easier to spot recurring problems. Granular Access Control: The new role and dashboard sharing options align permissions with actual job responsibilities, reducing risk. Proactive Monitoring: More users can subscribe to alerts, ensuring faster response to failures. Thank you for considering these suggestions to make Power BI’s monitoring more robust and accessible for all teams involved in the data lifecycle. Best regards, ... View more

Can we get a fix for the Bug where when using 2 bookmarks in the same report page, they both simultaniously work without the bug/glitch?    Current Behavior: The report page contains two bookmark navigators: Online/Offline toggle (primary selection) Sales/Visits toggle (secondary selection) Expected Behavior: When selecting "Offline" in the primary navigator, the "Sales" view should automatically display as the default visual (as configured in the bookmarks). The secondary navigator ("Sales/Visits") should reset to show "Sales" as selected   Issue Identified: If a user: Selects "Offline" (shows Sales view correctly) Then switches to "Visits" view Changes to "Online" Returns to "Offline" The visual correctly displays "Sales" (as per default bookmark settings) However, the "Sales/Visits" navigator incorrectly shows "Visits" as selected (retaining the previous selection state)   Visual Reference: The screenshot highlights this discrepancy in pink - the displayed visual (Sales) doesn't match the bookmark navigator's selected state (Visits).     ... View more

Power BI Desktop doesnt natively support a tenant switch liek Power BI service. This would be a much needed addition when working with mutiple organization. ... View more

I would like to propose an enhancement for the Date Range Picker visual in Power BI. Specifically, I believe it would be very useful if users could define default values for the "From" and "To" dates when the report is loaded. My ideal default behavior would be for both the "From" and "To" fields to automatically select the current date. This would provide immediate context for daily data and reduce the need for repetitive manual selection. Of course, users should still be able to adjust the range as needed afterward. This functionality would significantly improve usability, especially in dashboards that are accessed daily or in operational reports that rely on current-day data. ... View more

1st- I would request "Crore" and "Lakhs" units in the value for Indian users, We are able to achieve it using dax, but it gets convert into string type, result of that column is not able to sort in Asc or desc order, also cant use on graphs. 2nd- we also want Indian comma formatting for values, as the service does not support it. ... View more

After public publish, beeing able to download data and a picture of object would be great (Y) ... View more

Having the option for justifying text like in word would be great (Y) ... View more

df = spark.read.format("jdbc").option("url", jdbc_url).option("dbtable", f"({query}) as tmp").options(**jdbc_conn_properties).load() df.printSchema()   When checking how Azure SQL Database types are read in Fabric using JDBC INT, BIGINT and SMALLINT are transformed into logically sized types. TINYINT is read as an integer by default, which is not efficient in storage use.     ... View more

By applying a fully consumption-based pricing model to the OneLake API, access to OneLake will be possible even when the Fabric capacity is paused. ... View more

I would like to suggest a Global Workspace feature in Power BI that enables centralized report design management with automatic propagation to multiple Sub-Workspaces, each configured with its own client-specific data source. This would remove the need to duplicate and maintain separate report versions for each client or data environment. The core design elements—such as visuals, layout, bookmarks, and themes—would remain in the Global Workspace, and any updates made there would reflect across all linked Sub-Workspaces without affecting their individual data connections. This feature would greatly improve scalability, reduce manual maintenance, and ensure consistency in multi-client or multi-tenant deployments. ... View more

Hi Fabric Team,   I’m writing to report a critical issue we've encountered while using Microsoft Fabric for our ETL and ELT workloads. We rely heavily on both Warehouse and Lakehouse components. As expected, when configuring a copy activity, we can use SQL queries if the source is a Warehouse. However, we discovered that Fabric currently does not prevent queries from referencing Lakehouse tables in a Warehouse-scoped query. This unexpected behavior led to a serious issue in one of our production pipelines—records from a daily load were not copied to Oracle as expected. After an in-depth investigation, we traced the problem to a query that mistakenly referenced a Lakehouse table, even though the copy activity source was configured as a Warehouse. Fabric did not flag this mismatch, and the copy activity silently failed to extract the intended data. This kind of inconsistency is very hard to detect and can have severe consequences in production environments. 🛠️ Suggested Improvement: Implement a validation step during query submission in copy activities to ensure that all referenced tables belong to the selected source type (i.e., Warehouse). This would prevent these subtle but impactful issues. Thank you for your attention to this matter.   Kind regards, Yazdan Aghajanian Data Engineer ... View more

There is title (or subtitle) padding on shapes or buttons, only Text padding. If the title (or subtitle) is left or right aligned, it's too close to the border/edge and there is no way to control that! ... View more

Having the report as different json files is great, but it's tricky to modify the files given the complex structure that they may have. Even trying to compile the structure from the public schemas is tricky. It would be great if microsoft published a c# class that can deserialize all the different json files of the PBIR format thank you! ... View more

With PBIR is now possible to programatically modify the visual.json files to speed up the report design process. However, having to close and open the file to see the changes is time consuming and may even defeat the purpose of programatically modfying the visual.json files. Automatic update like when a measure is created by an external tool would we awesome, but at the very least there should be some way of reloading the report layer without closing everything and reopening the file.   thank you!! ... View more

Super simple request - would like to see the ability for a table visual in PowerBI to be able to have the background color modified to make it easier when a row is selected. Often users select a particular row to drive interrelated visuals and then they lose track of the row they selected as the backbround color of grey (blah) is easy to scroll by.   ... View more

In our investigation of the new AI data agents, a common question that our users want to ask is something along the lines of "What was the total sales for such and such product this year?" However, the agent would consistently use the year 2023 as the current year. Out of curiousity, we asked it "What is the current year?", and again it answered "The current year is 2023." If we ask it about the current time, it doesn't even know how to answer. It's odd that the current date and time are not included by default. ... View more

I don't know if this is the right place to make suggestions, but I think that would be amazing if we could create tables or metrics from visual tools, such as tables or cards, without being necessary write the code for it, whereas we already made a visual for itself. ... View more

We currently use Azure DevOps pipeline to manage our CI/CD, much like https://learn.microsoft.com/en-us/fabric/cicd/deployment-pipelines/pipeline-automation-fabric . However, because not all Fabric REST Api's support Service Principal authentication, our CI/CD Pipeline has to use a service user account (with disabled MFA) to authenticate our requests to e.g. trigger a deployment https://learn.microsoft.com/en-us/rest/api/fabric/core/deployment-pipelines/deploy-stage-content?tabs=HTTP . Unfortunately the next version Azure Connect-AzAccount will no longer support MFA from July 2025 - see Upcoming breaking changes in Azure PowerShell | Microsoft Learn . The impact of multifactor authentication on Azure PowerShell in automation scenarios | Microsoft Learn is that this change will prevent it's use in automation tasks, and so users need to switch to using either a service principal, managed identity, or a federated identity. Currently Service Principal authentication is not supported for the key Fabric REST Api calls needed for our CI/CD pipeline, for example deploying to stages, scheduling pipelines, invoking notebooks.   Please advise how we can prevent our automation pipelines breaking later this year?   We would expect service principals to have to be supported at some point in the future for deployment pipeline and for running notebooks and pipelines, but because I haven't been able to find this on e.g. What's new and planned for Fabric Data Engineering in Microsoft Fabric - Microsoft Fabric | Microsoft Learn we now need urgently need clarity on the timescale. ... View more