Add visual-level option to preserve scroll position (e.g., Keep scroll position on right) In Power BI (Fabric / Service), visuals that require horizontal or vertical scrolling currently reset their scroll offset to the default origin (left/top) every time a report is rendered. This occurs regardless of the scroll position at save time and cannot be controlled by report authors. As a result: Initial render frequently hides the most relevant data (e.g., latest periods in time-series visuals positioned at the far right) Users must manually scroll on every load or refresh Report authors rely on non-semantic workarounds (sorting, filtering, duplicate visuals) that alter visual meaning or increase maintenance overhead There is no visual-level or page-level API or formatting property to define or preserve the initial scroll position during the render lifecycle. Proposed Enhancement Introduce a visual-level formatting property that allows authors to explicitly define the initial scroll offset when a report page is rendered. Suggested configuration (Format pane → General / Advanced): InitialScrollPosition Auto (default – existing behavior) Start (left / top) End (right / bottom) PreserveLastSaved This property would be evaluated during: Initial report load Page navigation Visual re-render triggered by filter/context changes Expected Behavior When configured: The visual’s scroll container initializes at the configured offset without user interaction Behavior is deterministic and consistent across: Power BI Service Microsoft Fabric Power BI Apps Embedded scenarios The setting does not alter data semantics, sorting, or filtering logic Existing reports retain current behavior unless explicitly enabled Justification / Engineering Rationale Scroll offset is a presentation concern, not a data concern, and should be configurable independently of model logic Lack of control forces report authors to encode UX intent into data transformations, which is brittle and semantically incorrect Similar lifecycle control already exists for: Page state Filters Bookmarks (manual only) Visual layout and sizing This enhancement would close a functional gap at the visualization rendering layer. Common Scenarios Time-axis visuals where the most recent data appears at the end of the categorical or continuous axis Operational dashboards with daily/weekly monitoring patterns Executive reports where first render must surface “current state” without user guidance Large categorical visuals where scrolling is unavoidable Summary Providing a visual-level mechanism to control initial scroll position would significantly improve UX consistency, reduce report complexity, and align Power BI visual behavior with modern dashboard expectations—without impacting backward compatibility or data logic. ... View more

The API endpoint described in https://learn.microsoft.com/en-us/rest/api/power-bi/admin/reports-get-reports-as-admin is missing a few fields. A few are Contact list and Endorsement level. These should be made available to the API, because currently the API is very incomplete. ... View more

Real-Time Operation Agent Using Workspace Monitoring: Enhance with L2 Activity Data for Pipelines ... View more

Come on man, keyboard shortcuts still aren't available in PBI, why not and when is it going to happen?????? ... View more

DAX development in Power BI is inherently complex due to its dependency on evaluation context (row context, filter context), relationships, and data modeling structure. While experienced users can write optimized measures, a large portion of users struggle with translating business requirements into correct and performant DAX expressions. I propose a context-aware AI DAX engine embedded within the formula bar that goes beyond simple text-based suggestions. This system should leverage the underlying semantic model (tables, relationships, column data types, hierarchies, and measures) to generate accurate, optimized, and contextually valid DAX expressions. Unlike generic AI suggestions, this feature should: Understand data model relationships automatically Detect active/inactive relationships Interpret filter propagation and evaluation context Recommend best-practice DAX patterns (e.g., CALCULATE optimization, use of variables, iterator functions) The AI should allow users to input natural language queries such as: “Calculate year-over-year growth for sales by region” “Create a rolling 3-month average of revenue” “Generate a dynamic segmentation table based on customer revenue” The system should then: Generate the DAX expression Explain the logic step-by-step Suggest performance improvements Validate correctness against the data model  Key Advanced Features  1. Context-Aware DAX Generation Uses schema + relationships Avoids incorrect aggregations Handles filter context automatically 2. Performance Optimization Engine Suggests: Using VAR for readability Avoiding expensive iterators Replacing inefficient patterns 3. Debugging & Explanation Layer Explains: Row context vs filter context Why a measure returns a specific result Helps users learn DAX, not just use it 4. Auto-Pattern Recognition Detects use cases like: Time Intelligence Ranking (Top N) Cohort Analysis Suggests pre-built optimized templates 5. Interactive AI Prompt Panel in Formula Bar Inline AI assistant (not separate tool) Works while writing DAX Suggests corrections in real time  Advanced Use Cases  1. Time Intelligence (Complex) User Input: “Calculate year-to-date sales and compare with last year” AI Output: Generates TOTALYTD, SAMEPERIODLASTYEAR Handles date table automatically 2. Dynamic Segmentation User Input: “Group customers into High, Medium, Low based on revenue” AI: Creates calculated column or measure Uses SWITCH + thresholds 3. Advanced Filtering Logic User Input: “Show sales only for top 10 products within each category” AI: Uses RANKX + CALCULATE + FILTER properly 4. Rolling Calculations User Input: “3-month moving average of profit” AI: Builds window logic using DATESINPERIOD 5. Multi-Table Context Handling User Input: “Total sales per customer with region filter applied” AI: Handles relationships correctly Avoids common mistakes ... View more

Sometimes we may have several visuals on one page in a dashboard filtered by a slicer, if one of the visuals (size like 1/4 or 1/3 of the screen) on a selected date is blank, can we use conditions  or parameters in another visual to hold the space. (Also for using BI visuals in slides)    ==>    Thank you!  ... View more

A tree plot that has different formulas for different nodes by only pulling numerical values of the child nodes would be very cool to work with. ... View more

Current platform limitation: At this time, Microsoft Fabric does not provide a true fixed (non‑DST) UTC scheduling option. As a result: - Scheduler trigger times may shift during DST periods. - Time/Date extraction in notebooks continues to run internally in UTC (non-DST-aware). - A perceived one‑hour mismatch can occur between trigger time and UTC time referenced in a notebook   Suggestion: -Make it clear that UTC means UTC -Create option for DST adjusted variant of selected timezone.  ... View more

I have a pipeline that makes api calls to force fabric after a git checkin to update the fabric workspace but continue to run into issues when it's warehosue related. Please fix this. See below:   • The issue is that “deploy the Warehouse” and “clear Fabric Source Control sync state” are not the same operation. Your latest run proves this: - The pipeline did deploy the Warehouse. - sqlpackage published successfully. - The failure happened after that, when the pipeline rechecked Fabric git/status. So the broken part is not the Warehouse publish anymore. The broken part is Fabric Source Control reconciliation. What is happening: 1. The pipeline publishes the Warehouse schema directly to Gold_Management. 2. That changes the actual Warehouse. 3. But Fabric workspace Git metadata still says the workspace head is old: - workspaceHead = 827488413510f0a52a4458270c4 - remoteCommitHash = 989196d7abfcbbf86e1d3fea 4. The pipeline then asks, “can I finish the Git sync?” 5. The answer is still no, because app-only updateFromGit on a pending Warehouse item is the exact Fabric path that returns PrincipalTypeNotSupported.   So the issue is: - app registration can deploy the Warehouse - app registration cannot clear the Fabric workspace Git sync state for Warehouse items through updateFromGit That is why the run says: - publish succeeded - sync still pending - stop So if your expectation is “the pipeline should both deploy and mark Fabric Source Control fully synced,” that is the part Fabric is blocking under app-only auth. The current failure is not a code bug anymore. It is a platform limitation around Fabric workspace Git sync for Warehouse items under app-only auth.   Sources: - Fabric updateFromGit: https://learn.microsoft.com/en-us/rest/api/fabric/core/git/update-from-git - Fabric Warehouse service principal support: https://learn.microsoft.com/en-us/fabric/data-warehouse/service-principals - Fabric Warehouse deployment with SqlPackage: https://learn.microsoft.com/en-us/fabric/data-warehouse/development-deployment If you want, I can make the policy change now so pipeline 1363 succeeds after a successful Warehouse publish and only emits a warning about stale Fabric Source Control state. ... View more

I was thinking to start a project to build a Cross-Platform Mobile App (iOS & Android) using Flutter, dedicated specifically to the Microsoft Fabric community. The goal is to provide a central hub for tracking updates, community discussions, and learning resources on the go. ​I am currently in the infrastructure planning phase and designing the data schema. Below is the technical roadmap I’ll be following to bring this from an idea to the app stores.   Phase 1: Architecture & Data Schema ​Schema Design: Defining entities for Users, Community Posts, Knowledge Base articles, and Certification tracks. ​Backend Selection: Using Firebase for real-time synchronization and authentication, integrated with Azure Functions to fetch official Fabric updates. ​Phase 2: UI/UX & Brand Identity ​3D Integration: Leveraging my background in 3D motion graphics to create a high-end, cinematic Splash Screen and interactive UI elements using Rive for Flutter. ​Design System: Adopting Microsoft’s Fluent Design System principles to ensure the app feels like a native part of the Fabric ecosystem. ​Phase 3: Flutter Development (The Core) ​State Management: Implementing BLoC (Business Logic Component) to ensure the app is scalable and testable. ​Clean Architecture: Separating the app into three layers: ​Data: Repositories and Data Sources (API/Firebase). ​Domain: Use Cases and Entities. ​Presentation: UI Widgets and ViewModels. ​Networking: Using the dio package for robust API handling and interceptors. ​Phase 4: Testing & Quality Assurance ​Unit & Widget Testing: Ensuring every button and logic flow works across different screen sizes. ​Integration Testing: Testing the end-to-end flow from login to data retrieval. ​Phase 5: CI/CD & Production ​Deployment Pipeline: Setting up GitHub Actions for Continuous Integration and Continuous Deployment. ​App Store Optimization (ASO): Preparing the Google Play Store and Apple App Store listings with high-quality visuals. ​Monitoring: Using Firebase Crashlytics and Google Analytics to monitor performance and user engagement post-launch. ​I’d love to hear your thoughts! What features would make your life as a Fabric professional easier? ​Looking forward to building this together! ... View more

When a mirroing on sql server is created on Fabric. replication and cdc is automatically enabled on the source database. However when Mirroing is deleted on Fabric, it does not autoamtcially disable the replication and cdc on the source database. I had to manually disable them on the source database. It would be effective if this is taken care when the mirrored database is dropped in fabric. Before deleting it can prompt if cdc and replication on source system can be dropped as well.   ... View more

In T-SQL, the IDENTITY property is a powerful feature for automatically generating unique values, often used for primary keys. This capability is currently missing in Fabric Lakehouse, and enabling it would be highly beneficial. I understand that IDENTITY has been introduced for Fabric Data Warehouse tables; however, the seed value cannot currently be reset or modified. It would be great to see these enhancements included in the coming months.   ... View more

Hi,   I think it is ridiculous that this has been requested since at least 2020, still there is no possibility to customize the messages users see when opening a Power BI report that they do not have access to (only works for workspace apps so far). It doesn't seem like such a complicated feature to add relative to its importance. https://community.fabric.microsoft.com/t5/Fabric-Ideas/Custom-quot-Request-Access-quot-pop-up-button/idc-p/4451875  Pushing this once again..   Best Regards ... View more

We use fabric and have a workspace app containing fabric bi reports. We've noticed that the filters "remember" the user's last filters applied even if the BI reports are saved and published with a certain state. This is specifically a functionality of a workspace app and perhaps the browser's caching. Users have to click on the green "Reset filters" arrow icon to revert to the default state of the BI report. ... View more

It is possible for an SVG canvas background to have no issues when imported in Desktop, display in the service the same, but then will fail when exported to PDF. I've encountered this with attr() which I use to make customisable background templates, but in theory this could also apply to any new development to the web platform. I tested the latest version of my template in desktop and on the service, but did not think to test with exporting as I'd assumed the background appearance was fixed at that point. The export error only came to my attention when a different export error was queried. I suspect that different versions of WebView are at play, but whatever the cause the capabilities should be matched where appropriate. This is also affected by the recent decision to change the default canvas size. The entire point of using attr() in the first place in my template is so I can change the dimensions of the SVG template to match the canvas, then attr() can read this and automatically place certain elements. Since SVG doesn't natively let you place elements relative to the bottom or right of the canvas, CSS that lets you keep elements in those places needs to know the canvas size. Hence the attr() trick. But this isn't about one capability in particular - no doubt things will advance and future capabilities will emerge that will prove useful. But report authors should be confident anything they have tested in one place will work everywhere. ... View more

Problem In Power BI today, any custom visual that supports Advanced Edit Mode (for example Deneb, Charticulator, and other editor-style visuals) loses access to standard Focus Mode. Clicking the Focus Mode icon no longer performs the expected behaviour of zooming the visual to fill the canvas. Instead, it immediately opens the visual in Advanced Edit Mode. This behaviour is imposed by the Power BI host and cannot be controlled by the visual developer. As a result: Report consumers cannot use Focus Mode to inspect complex visuals Authors and viewers experience inconsistent UX compared to native visuals Visuals that are primarily analytical (not always being edited) pay a usability penalty simply because they support editing Why this matters Editor‑style custom visuals are increasingly used far beyond initial authoring: Deneb visuals are often parameter‑driven, interactive, and fully dynamic at run‑time Charticulator visuals are commonly consumed, not edited, by report viewers These visuals benefit more than native visuals from Focus Mode due to: Dense layouts Custom scales Fine typography Responsive rendering However, today users lose Focus Mode entirely for these visuals — even though an explicit Edit command already exists in the visual’s context menu (three dots). Requested change Please allow standard Focus Mode to remain available for custom visuals that support Advanced Edit Mode. Possible implementation options (any of these would work): Decouple Focus Mode from Advanced Edit Mode Introduce a capability flag that allows: Focus Mode = zoom only Edit Mode = editor lifecycle Allow visual developers to specify: “Advanced Edit via menu only” “Do not hijack Focus Mode” This would preserve: Expected Power BI UX Backwards compatibility Editor functionality for advanced visuals Who benefits Power BI report consumers (better analysis experience) Advanced report authors Custom visual developers Popular certified visuals such as: Deneb Charticulator Other authoring‑capable custom visuals This is a platform‑level improvement that enables more powerful visuals without reducing usability. @dm-p  ... View more

Currently, Power BI Desktop supports display units such as Millions and Billions, but it does not provide Indian numeric units like Crores, Lakhs, and Thousands. These units are widely used in India for financial, business, and government reporting. Adding them would make Power BI more user-friendly for Indian users, improve localization. This feature would enhance adoption and usability in the Indian market, aligning Power BI with regional reporting standards. ... View more

Within the PowerBI Subscriptions I would like the message to be both longer, as well as rich text.  I would be nice to send a link (Approval letter, etc.) within the email blurb that gets sent out.  OR to add an additional bit of rich text information that can be added to below the snip screen of the Subscription Email. Also, it would be nice if you can copy and paste a group of emails from an email or distribution list in Outlook. ... View more

It's incredible that we still don't have the option to control the favorites buttons to actually determine which favorites we want to display in the button visual.   We need an update to this component to improve our DataViz.   Let's say, for example, that I have 12 Indicators, but in the button I only want to select 6 to be displayed/selected.     ... View more