The idea of enabling a component that can orchestrate multiple copy job items on the platform is suggested. ... View more

When a category is selected in the Tree Map visual, the corresponding filter is applied to the Table visual. However, when a row is selected in the Table visual, the Tree Map selection gets cleared. Solution for this is that users hold the Ctrl key while selecting or deselecting rows or data points. This allows multiple selections without resetting the visuals. This interaction seems to be a bug to me. Even if it is the expected behaviour, what to do in case of mobile layout ? For mobile layout, instead of Ctrl key, what option can be used to view the expected behavior? ... View more

I would like to suggest an enhancement for Power BI Desktop: a more advanced search functionality. The idea is to allow users to search for a specific keyword or term and quickly identify which tables, columns, or even measures that term is being used in. This would greatly improve navigation and understanding of complex data models, as well as speed up the analysis and editing process. ... View more

📝 Description: In Microsoft Fabric Data Pipelines, the Switch activity requires a mandatory default branch, even when all valid cases are explicitly defined and handled. This limits flexibility and leads to unnecessary clutter or workarounds in pipeline design. 💡 Feature Request: Please add support for optional removal or disabling of the default branch in the Switch activity. 🔍 Why This Matters: 🔄 Cleaner Logic: In many scenarios, all possible input cases are explicitly handled (e.g., "csv", "json", "parquet"), and a default branch serves no meaningful purpose. 🛠️ Developer Control: Sometimes, it’s preferable to fail or stop execution if no case matches, rather than route execution through a forced dummy path. 🧼 Improved UX: The current behavior leads to unused or “empty” default branches, which clutters the pipeline canvas and adds confusion during maintenance. 📊 Consistency: Programming languages and orchestration tools often allow Switch/Case logic without requiring a default block. Aligning with this flexibility would be consistent and intuitive. ✅ Suggested Implementation: Add a toggle to disable the default path. Alternatively, provide an "on no match: terminate/error/continue" option. 🚀 Impact: This small change would: Simplify complex pipelines Improve usability Empower developers with more control over flow logic ... View more

My idea is for when you first turn on Fabric don't have a trigger that spams the user with thousands of "alerts" everyday. ... View more

Hi all, I’d like to suggest a new feature for the REST API: an endpoint to retrieve users’ favorite artifacts (e.g., reports, notebooks) on a per-user basis. Currently, I haven't found any way to get this data via the existing API or activity events. Having such an endpoint would be very helpful for tracking and personalizing user experiences. Thanks for your attention. Regards ... View more

Much like one can create several shortcuts in the interactive dialouge, need the same functionality and dialouge for parquet folders. the current process requires manual one-by-one loading with a long wait in between as the tables are being created...  ... View more

Hi, I've been using Power BI for the past two years and wanted to suggest an important improvement. My data is in Hebrew, which is a right-to-left language, but currently there’s no way to align values automatically in tables and matrices based on text direction. This makes RTL reports look less professional and harder to read. It would be great if you could add RTL value alignment support to the Power BI Theme JSON schema. Thanks a lot! Keren ... View more

Please add a close solution button to the Power BI file menu ... View more

There’s a known limitation when using calculation items in slicers. When a user selects a single calculation item, any unrelated measure on the page — like a title measure or a measure that returns text — ends up showing the result of that calculation item, instead of its intended output. The usual DAX functions like SELECTEDVALUE or VALUES don't work as expected with calculation groups. Also, excluding unrelated measures by adjusting the calculation items DAX's like using ISSELECTEDMEASURE & IF is not an scalable solution, and it makes the calculation item DAXs much more complex. One common workaround is to create a separate table with the same metric names and use that for slicers. For this purpose, a relationship is required between this proxy table and the Caclution Group table which is currently unavailable in Power BI. It'd be very useful if this issue was resolved sooner or at least the relationship of the proxy table and the calculation group table would become available. ... View more

**Description:** Currently, in Power BI Desktop, when a user attempts to clear a column or measure name and presses Enter, Power BI reverts to the default name (such as the original table column name or measure name). However, it is possible to use a zero width space (U+200B) as a column or measure name if it is entered manually, resulting in a column header that appears nameless in visuals. **Idea:** When a user removes the entire column or measure name and clicks Enter, Power BI should automatically insert a zero width space (U+200B) as the name, instead of reverting to the default name. This would allow for visually nameless columns or measures without requiring users to manually copy and insert the invisible character. **Benefits:** - Streamlines the process for users who want blank column headers, especially for cleaner visuals. - Reduces the need for manual workarounds or copying special characters. - Improves usability for both new and experienced users who need nameless headers for aesthetic or practical reasons. ... View more

It would be a nice feature to be able to add rules to existing service roles or to be able to create a new service role. Use case, I have a few developers that I don't want to be able to anything other than view and download the PBIX files in a production workspace. Adding more permissions creates potential issues with governance, but I would like for them to be able to download the PBIX files for developing or seeing how things work.  ... View more

connect to Lakehouse from the One Lake catalog, but you’re unable to see the reporting view to edit the model on the desktop. We can see only the model view and DAX view, but the model view is not accessible. ... View more

Simple report requested where a date frame can be entered and you can view all cases that were staged and by who that includes case numbers, date staged, employee name, QA'd, etc...  ... View more

Databricks now has a free tier/edition, so should Fabric...   Yes, I know Fabric has a free trial, but this eventually runs out. Databricks describe their free tier as for 'students, educators, hobbyists, and anyone interested in learning or experimenting with data and AI'. I've spoken to a number of people whose free trial has run out and now they can't afford to pay for a capacity despite only wanting to play/learn.   If databricks is offering this, then surely Microsoft (in order to compete) need to match this a similar offer in Fabric? ... View more

As a user, I would like the ability to bulk create multiple columns for a existing data table. Each column could have different calcuclations. To ideally have a script editor like in Power Query.  ... View more

When deploying things from stage 2 to stage 3 (Test to Live/Production)  the comparison and deployment works absolutely fine. However, when trying to deploy from stage 1 to stage 2 (Dev to Test environments)the comparison will not load and displays the error message in the below screenshot "Stage can't perform a comparison".  Additionally I am unable to select an item to promote from dev to test. The option to check the boxes is visable but will not engage.    ... View more

Often when designing a semantic model schema that uses bidirectional cross-filtering, the resulting model is ambiguous. Often, the accepted solution to this issue is to disable bidirectional filtering and make do with single direction cross-filtering. This is unfortunate, because bidirectional filtering is a powerful feature that can aid insights in reports. Another solution is to duplicate a certain dimension table (create a role playing dimension). This is also often undesireable, because two dimension tables have to be managed and often filtered in tandem.   It's not the case that bidirectional filtering always results in ambiguity, and it's not even the case that relationships currently dissallowed in semantic models due to ambiguity have to be ambiguous. Considering the path used in the relationship could, in many cases, resolve the ambiguity. Consider this model: Here it would be convenient and useful to be able to filter accounts based on selected invoices, and to filter accounts based on selected budgets. This would allow, for example, more direct comparison of budgets and invoices associated with a given account. Unfortunately this is not possible: setting the accounts -> budgets and accounts -> invoices relationships to cross-filter in both directions gives an error: Actually, even setting one relationship to cross-filter in both directions results in ambiguity. This is undesireable and is the reason that bidirectional cross filtering is discouraged.   It's easy to see why this happens : the Date dimension can filter accounts either through invoices or through budgets. The unfortunate part is that this effect is unrelated to the intention to filter the two fact tables from the accounts dimension; the Date dimension is not directly related to the accounts dimension.   It would be good if we had a feature in PBI to limit the filtering effect of a relationship to a certain group of tables (similar to the effect of a VLAN in a networking setup). For the above example, we could: Assign the Date -> invoices and Date -> budgets relationships to group A Assign the accounts -> invoices and accounts -> budgets relationships to group B Disallow filtering operations from a group A relationship to propagate along a group B, and vice versa This would allow/maintain the advantages of bi-directional filtering, while removing the perils of ambiguous relationships. Situations where all four tables are used in the same visual would still parse, because there is at most one relationship actively defined between each pair of tables.   I realise that this would not be a light feature to implement, but I think it has the potential to add a lot of value in terms of the insights available from models and the clarity of semantic models themselves. ... View more