Hi Team, We are encountering an AccessDenied error when trying to refresh the Power BI report in the Power BI Service (workspace/online). However, when we refresh the same report locally using Power BI Desktop, it works without any issues. Could you please help us investigate and resolve this issue? Please let us know if you need any additional details from our side. Data source error: AccessDenied. The exception was raised by the IDbCommand interface Thank you for your support. Best regards, Sunit Kumar ... View more

When converting from a local semantic model to a live connection, it leaves local semantic model folder intact. This can be removed using a workaround but it would be much more convenient if this could be done within the app.   The workaround is here; Solved: Re: Remove local model after switch to direct quer... - Microsoft Fabric Community Specifically michelsilva's response.   When developing, it is useful to use a local model and switch to a published "live connection" later. Various posts say switching back is not possible but the workaround shows there is no reason for this to be the case.   ... View more

Below is an explanation of my IQR Clustering Analysis in Excel hotspot model analysis applied to over 190,000 records of 311 Toronto Customer Initiated Service Request Data from January 1 to April 30, 2026. Model Input:   The input data for this model analysis was an Excel worksheet (or it could be queried from or located in some other place like a database) with 26 columns and 533 rows of data.   Column A of the Excel worksheet lists each of the 533 reasons (Service Request Types) why a resident of Toronto has contacted 311 Toronto to complain or make a request over the four month period studied.   Columns B-Z column headings are the names of each one of the 25 City of Toronto wards.   Each cell contained a raw count of the number of 311 Toronto Customer Initiated Service Requests made by residents of a City of Toronto ward about a specific Service Request Type.   Model Output:   The output data for this model analysis is shown in an Excel worksheet that you can email me to obtain: it still has 26 columns and 533 rows of data (buy I do also have output without the raw counts overwritten).   Instead of raw counts for the cell values, one of the following 5 text labels are used: 'High Outlier', 'High', 'Normal', 'Low', 'Low Outlier'.   The 'High Outlier' values in cells indicate that, based on distribution of raw counts of service requests (across Service Request Type) received by 311 Toronto within a given ward, complaints for a specific Service Request Type in a ward have been found to be significantly, statistically high based on the IQR Rule theory of the box & whisker plot.   Model Conclusion:   Based on the attached Excel output sample, my analysis revealed that not all City of Toronto Wards are the same in terms of the time, energy and resources that the City must spend fixing the various service requests logged by residents.    Animal Noise complaints are significant issue in only 3 of the 25 City of Toronto Wards for the first four months of 2026: Parkdale-High Park (04), Toronto Centre (13) and Toronto-Danforth (14). It is for this reason, that their cell values are shaded yellow and have the words 'High Outlier' written in red.   Residential / Yard Waste / Missed proved a significant issue for only 2 Wards and their cell values are shaded yellow with the words 'High Outlier' written in red.   The Hotspot Analysis Spreadsheet Template used to automate this analysis works on over 8,000 columns and as many rows of data as your version of Microsoft Excel accepts. The process will run as fast as your computer chip can handle. And, this IQR analysis process is an easier to apply and interpret segmentation process than either K-means Clustering or Hierarchical Clustering. ... View more

It would be good to have a recycle bin in Power BI. Today, we can recover the workspace from Admin center but their is no way to recover the reports which are deleted.   Expected Behavior Recyle bin should hold all the deletion which has happened. A restore option should then be avaiable. ... View more

PROBLEM STATEMENT The PBI Audit Log doesn't provide a way to link App-Owns-Data activity to a user since it is "user-agnostic". Everything shows under the app's identity rather than logging a property that distinguishes User A from User B activity. As it stands today, App-Owns-Data logs basically show "we know SOMETHING happened but can't tell WHO produced it"; that doesn't offer much value.   IMPACT SUMMARY This poses signficant challenges to adoption/implementation when needing to track activity for audit/security/compliance reasons. This forces customers to create custom solutions to export data rather than leverage PBI's built-in Export feature, which is a shame.   SOLUTION By tagging App-Owns-Data activity with the EmbedTokenID that produced it, this would make it possible to correlate log activity. This empowers orgs to then ensure EmbedTokenIDs are unique to users and capture the IDs so they can link a user to the logs.   RELATED IDEAS Add Export Triggered Event Handler Include User Identity in Power BI Embedded Logs ... View more

Please add support for additional authentication methods for the Azure Databricks connector in Fabric/Data Factory Copy Data and Copy Job activities. Current documentation shows that Copy supports Personal Access Token authentication, whereas other experiences such as Dataflow Gen2 support additional authentication methods including Microsoft Entra ID.  PAT-only authentication is not sufficient for enterprise production scenarios. PATs introduce challenges around security governance, token rotation, ownership, expiry, and operational support. Databricks also recommends OAuth where possible instead of PATs because OAuth provides stronger security.  In our current Copy Data / Copy Job implementation, we are only able to connect to serverless Databricks SQL/warehouse-style endpoints, and we are not able to use the other Databricks cluster types required for our workloads. This limits flexibility for existing Databricks architectures where teams need to connect to specific non-serverless or interactive compute resources. Please consider adding support for enterprise authentication options such as: Microsoft Entra ID authentication Service principal authentication System-assigned managed identity User-assigned managed identity This would reduce dependency on PATs, improve security and governance, simplify credential lifecycle management, and provide more flexibility when connecting Fabric/Data Factory Copy activities to Azure Databricks compute. ... View more

We have data agents working on top of date that comes from medallion layer, shortcuts and other fabric sources. every time we try to share an agent, we start jumping through which access is the user missing loop. is in one of the medallion layers, is it in one of the lakehouses, is it in a specsific table or something else. the agent itself just returns generic responses (sometimes with actual production layer suggestion like: i can't help you right now, try asking specifically about an asset like ZCR443322 - which actaully exists somewhere in the data). this is counter-productive and very frustrating consideting that the agent itself is semi-intelligent. the idea is that the agent, instead of providing error codes that no-one understands or spewing some AI-SLOP type generic answers not connected to the data, will respond with the missing access rights that user needs in order for it work (like when trying to access a sharepoint site with no access etc.)  ... View more

The new "Grow to fit" table layout option is amazing and long overdue. I would like to suggest an improvement. Please can we have an additional option that grows to fit with equal sized columns widths in the same way that you can "Distribute Columns" in word. I appreciate that there is a custom width option however to have it automatically do it would make the end result super clean automatically ... View more

Hi Power BI Embedded is currently incompatible with web applications that enforce the Cross-Origin-Embedder-Policy (COEP) security header   Since Power BI Embedded does not currently support this scenario, applications are forced to disable COEP globally, introducing architectural and security limitations   Please provide official support for COEP-compatible Power BI Embedded scenarios or document the limitations and recommended workarounds Our applications need to enable the following HTTP response header: <add name="Cross-Origin-Embedder-Policy" value="credentialless" /> Thanks ... View more

Hey team, Is there an ETA for supporting Fabric Capacity Contributor assignment via API? The Capacity Contributor permission is critical for organizations that want to give end users the ability to create workspaces and assign Fabric capacity — without granting full Capacity Admin permissions. This is an important permission boundary that many customers rely on for least-privilege access models. However, based on my current understanding, the API only supports assigning Admin permissions. Assigning the Contributor role is currently only possible through the Fabric Admin Console, which introduces two significant pain points: Operational overhead — It requires manual effort from Fabric admin teams for every new Fabric capacity. Limited automation — It blocks the ability to set up and manage Fabric environments programmatically at scale. Supporting Capacity Contributor assignment via API would unlock fully automated Fabric provisioning workflows and reduce the administrative burden on platform teams. Would love to hear if this is on the roadmap! ... View more

We have DEV -> UAT -> PROD workspaces with deployment pipelines moving content between these.  We also use variables to allow us to dynamically change connections destinations and other names that changes between these, but there are some things that can't use variables, which requires pipeline rules.    The problem with these rules are that only the person who creates them are able to see the content of it, and change it. It is a very immature system as it is hard to maintain and organize properly. If these deployment rules were stored in a variable object instead, it would be easier to maintain and more flexible. ... View more

The current dashboard subscription has only options for Hourly, Daily, Weekly and monthly. However, it would be helpful if we have option for fortnightly too. There are lots of organisation that have payment cycles fortnightly in NZ. Please have this option added in the repeat category.  ... View more

I suggest an improved Google Analytics connector that allow us to connect more than one GA account to the same model. ... View more

Organizations with large and dynamic user populations require scalable Row-Level Security (RLS) management. The current approach using USERPRINCIPALNAME() and user-to-data mapping tables becomes difficult to maintain as users frequently join, leave, or change responsibilities. Proposed Enhancement Make the USERMEMBEROF() DAX function generally available across Power BI Premium and Microsoft Fabric capacities, allowing RLS rules to evaluate Azure AD (Microsoft Entra ID) group membership directly within DAX. Business Value Using USERMEMBEROF() would enable organizations to: Manage security through Azure AD / Entra ID groups instead of individual user accounts. Reduce administrative overhead for large deployments. Align Power BI/Fabric security with existing enterprise identity governance practices. Simplify RLS model design and maintenance. Improve scalability for organizations with thousands of users and frequent access changes. Current Limitation Today, USERNAME() and USERPRINCIPALNAME() return individual user identities rather than group membership. While AD groups can be assigned to Power BI roles, dynamic filtering based on group membership is not available without maintaining custom user mapping tables. This creates significant maintenance challenges for enterprise-scale implementations. Example Scenario A security table could contain: AD GroupRegion Sales_NorthAmerica North America Sales_Europe Europe Finance_Global All Finance Data With USERMEMBEROF(), RLS could dynamically grant access based on the user's Azure AD group membership, eliminating the need to store and maintain individual user records. Requested Outcome General Availability of USERMEMBEROF() for Power BI and Microsoft Fabric tenants. Support within Import, Direct Lake, and DirectQuery semantic models. Clear documentation and administration controls for enterprise adoption. This capability would significantly improve security management, governance, and scalability for organizations adopting Microsoft Fabric and Power BI at enterprise scale. ... View more

We would like to request a native Power BI / Microsoft Fabric connector for Archer GRC.   Archer is widely used by enterprises for Governance, Risk, and Compliance (GRC), including risk management, audit tracking, compliance reporting, and third-party risk monitoring. Currently, extracting data from Archer for analytics is heavily Reliant on custom APIs a native connector would unlock significant value for organizations by enabling: - Real-time risk dashboards in Power BI - Executive reporting across risk domains - Integration with Microsoft Fabric for centralized governance analytics - Advanced analytics and AI scenarios (Copilot, anomaly detection, forecasting)   This connector would benefit: - Risk & compliance teams - Internal audit teams - Security & governance stakeholders Key features requested: - Access to core Archer modules (risk, audit, compliance, issues) - Support for scheduled refresh and large datasets - Compatibility with Fabric data pipelines and DirectQuery (if feasible) Given the increasing importance of enterprise risk visibility and Microsoft's strong focus on Fabric, this connector would significantly enhance integration between GRC platforms and the Microsoft analytics ecosystem. ... View more

My PowerBI use is almost always to monitor time series. Here, sorting graph data as 'x-axis ascending' is the way to do it. Power BI, by default, sorts graphs 'y-axis descending', which is probably more useful for ranking porposes. It would be great if this default setting was switched or custumizable.    Thank you for considering this. ... View more

In Power BI, Waterfall charts visually represent cumulative changes, but there is limited ability to explicitly display running totals at each step (as available in Excel). Requested improvements: Option to display running total values on each column Ability to toggle cumulative labels on/off Improved control of intermediate and subtotal bars This would improve financial reporting, variance analysis, and reconciliation use cases where users need to clearly see the evolving balance at each step. ... View more

Currently, SAP Connectors in Microsoft Fabric do not support integration with Azure Key Vault (AKV). However, in many enterprise environments, it is a mandatory requirement to manage credentials centrally using Azure Key Vault for secure access to databases and external systems. SAP Business Warehouse Server SAP Business Warehouse Message Server SAP HANA       ... View more

Given that the existing reports need to be rebuilt from scratch, it would be greatly appreciated if you could suggest an approach and design to ensure the smooth migration of the public materials listed below.       Get data from Excel workbook files - Power BI | Microsoft Learn ---------- Is there a workaround to keep my existing reports updating? No. There's no workaround to keep reports created with the legacy experience updating past July 31, 2026. You need to recreate these reports using the current Excel and CSV connectors. ---------- ... View more