Passing filters from one page to another
Hi, I have a employee number slicer in one tab that i want to use to filter data on another page. The first tab has a table with employee number with repeat data and the second table has another table with employee data but not repeated. The behaviour i want is for the user to select a population on the first page and then goto the second page to see the filtered population there. i can't put the slicer from the first page on the second page as the data comes from different tables. Can anyone advise me as to the best way to achieve this ? many thanks
Incremental refresh with late-arriving data: how do you refresh older partitions safely?
Hi all, I'd like some input from people running incremental refresh on larger semantic models in the Service, specifically around data that arrives (or gets corrected) after its period has already moved out of the refresh window. Setup Import semantic model on Fabric capacity, fact table ~400M rows Incremental refresh policy: archive = 5 years refresh = last 10 days (by TransactionDate) detect data changes = off Source = SQL (warehouse), scheduled refresh 4x per day The problem Every few weeks the source team backfills or corrects transactions that are 1-3 months old. Because those days are outside the 10-day window, the model never picks them up, and month-end numbers in the report drift from the source until someone notices. Options I'm weighing: 1. widen the refresh window to ~90 days -> simple, but every scheduled refresh gets much heavier 2. turn on "detect data changes" with a LastModified column -> only changed partitions reprocess, but needs a reliable audit column 3. leave the policy as-is and refresh specific partitions on demand via XMLA / enhanced refresh API when a backfill happens 4. a weekly full refresh with applyRefreshPolicy = false -> brute force, long-running For option 3, I was thinking of something like this, triggered by the source team's backfill job: POST .../datasets/{datasetId}/refreshes { "type": "full", "commitMode": "transactional", "applyRefreshPolicy": false, "objects": [ { "table": "FactSales", "partition": "2026Q3" } ] } Questions 1. Which approach are you using in practice, and how has it held up? 2. With detect data changes, does a LastModified that only moves forward per row work well when rows get updated multiple times, or have you run into partitions being missed? 3. For targeted partition refreshes, how do you find the right partition names reliably (they change granularity as they age from day to month to quarter to year)? Do you query the partitions first, or derive them from the date? 4. Any gotchas with running an on-demand partition refresh while a scheduled refresh is in progress? Thanks, I appreciate any real-world experience here.Handling late and out-of-order events in Eventstream windows and Eventhouse aggregations?
I'm coming from Apache Flink, where late and out-of-order events are handled with event-time watermarks and allowed lateness. I'm trying to understand the equivalent approach in Fabric Real-Time Intelligence. Scenario: IoT devices send telemetry with an event timestamp. Because of network delays, some events arrive several minutes late, and occasionally out of order. I need accurate 5-minute aggregates (average and max per device) for dashboards and Activator rules. Questions: Eventstream: when using windowed aggregations (tumbling or hopping), are windows based on event time or arrival time? Is there a setting for out-of-order tolerance or late-arrival handling, and what happens to events that arrive after a window has closed: are they dropped, or do they produce a corrected result? Eventhouse: if I aggregate in Eventhouse instead, for example with a materialized view using bin(EventTime, 5m), are late events included automatically when they arrive? Any caveats with materialized views and late data? Which layer is better for aggregations when correctness under late data matters more than lowest latency? Activator: if a rule has already fired on an aggregate that later gets corrected by late data, is there a pattern for handling that, or should rules only run on finalized windows?
Fix the RLS issue for inactive relationship and find the missing country Name
Hi All, I am currently working on a report to identify missing Country Names based on the number of Contracts and Vehicle Plate Numbers. The report is subject to Row-Level Security (RLS) implemented on the Country dimension. While troubleshooting the issue, I attempted to activate an inactive relationship using the USERELATIONSHIP() function. However, the user encountered the following error: Underlying Error: QueryUserError More details: The USERELATIONSHIP() and CROSSFILTER() functions are not available in this context. As an alternative approach, I implemented the TREATAS() function to replicate the relationship behavior. However, the end user continues to experience the same issue. The relevant table relationships are as follows: Vehicle[FK_Subsidiary_Code] → Subsidiary[PK_Subsidiary_Code] Contract[FK_Subsidiary_Code] → Subsidiary[PK_Subsidiary_Code] Contract[FK_Vehicle_Code] → Vehicle[PK_Vehicle_Code] Could anyone advise on a possible solution or recommend an alternative approach for handling this scenario within the RLS context? Both Measure are not working TREATAS and USERELATIONSHIP Test = CALCULATE( DISTINCTCOUNT(Contract[PK_Contract_Code]), ALL('Vehicle'), TREATAS (VALUES (Vehicle[FK_Subsidiary_Code]),Subsidiary[Subsidiary_Code])) Test = CALCULATE( DISTINCTCOUNT(Contract[PK_Contract_Code]), USERELATIONSHIP(Contract[FK_Subsidiary_Code],Subsidiary[Subsidiary_Code])) I Can't change the relationship ThanksPBIP + Git with several report developers: how do you avoid merge conflicts?
Hi everyone, We're moving a team of four report developers from shared PBIX files on SharePoint to Power BI Projects (PBIP) with TMDL and PBIR in Git. Single-developer work is great: readable diffs, proper history, PR reviews. But when two people touch the same project in parallel, things get messy, and I'd like to know how other teams have organized this. Repo layout today /Sales.SemanticModel/definition/ tables/*.tmdl <- one file per table, nice relationships.tmdl <- everyone edits this one model.tmdl /Sales.Report/definition/ pages/<pageId>/page.json pages/<pageId>/visuals/<visualId>/visual.json report.json <- also touched by almost every change What we're running into - two branches add measures to the same table -> conflict in <table>.tmdl - adding a relationship on both branches -> conflict in relationships.tmdl - report.json / page ordering changes even when someone only edits one visual - after resolving a conflict by hand, Desktop sometimes refuses to open the project until the JSON/TMDL is fixed again Questions 1. Do you split ownership, e.g. one person owns the semantic model and others only touch the report, or a thin report per developer connected live to a shared model? 2. Do you keep measures in a dedicated "_Measures" table (or several, per subject area) specifically to reduce conflicts in fact-table TMDL files? 3. Any conventions for branches and PR size? For example, short-lived feature branches merged daily vs. one branch per report page. 4. Are you using any tooling to validate a merged project before someone opens it in Desktop (Tabular Editor CLI, a schema check in CI, etc.)? 5. For PBIR specifically: is there a way to stop unrelated changes (page order, report-level settings) from showing up in every commit? Curious to hear what has actually worked in a team setting. Thanks!Subscriptions now coming "from" subscription owner
According to this article https://learn.microsoft.com/en-us/power-bi/collaborate-share/service-troubleshoot-subscribe Microsoft has made the unilateral decision (and apparently without a proactive announcement) to have email subscriptions come from the creator/owner of the subscription instead of a generic Power BI email address, if the creator/owner uses Exchange Online. This is causing all kinds of headaches in my organization, because I personally set up many subscriptions on behalf of others, and now not only are they being received with a different sender, all of the individual emails are showing up in my Sent Items folder. The emails also now bypass out email security systems that all external emails go through. Has anybody figured out a way to reverse the changes or otherwise perform a workaround (setting the owner of a subscription to a generic or service user, etc.) ? Thanks David
Monthly average line chart — Some months legitimately have no data
I have incident data from Jan 2010 - July 2026 (17 years, but 2026 is incomplete; some records not entered yet). The data is stored in a MS list and used in Power BI Service. I want a line chart that shows the average monthly incidents to look at seasonal trends. I was vaguely aware of the "date trap" with averages on incomplete years, when it comes to the months remaining in 2026: Aug-Dec. But I hadn't considered the "date trap" for other so-called incomplete years. For example, there are no incidents/rows for April 2013. The data isn't missing; there just weren't any incidents in that month. Power BI has taken the total incidents for all Aprils in the data, dividing by 16 years = 2.1 April average incidents. But when checking in Excel, I took the total incidents for all calendar Aprils, dividing by 17 years = 1.9 April average incidents. I think what I did in Excel is correct for April (a month that had past years with no incidents). But perhaps Power BI is right for the latter half of 2026. Regardless, I'm more concerned about the April scenario and less concerned about the Aug-Dec 2026 scenario. In Power BI Service, how can I create a line chart of the monthly average and correctly handle past months that legitimately had no incidents?Select Entire Hierarchies in Power BI Personalize Visuals
Feature Request: Select Entire Hierarchies in Power BI Personalize Visuals Hi Power BI Team, I'd like to suggest an enhancement to Personalize Visuals that allows users to select an entire hierarchy from a semantic model instead of adding each level individually. For example, my semantic model contains this hierarchy: Responsible Division Hierarchy Responsible Division Responsible Cost Center Currently, when users open Personalize this visual, Power BI correctly displays the hierarchy in the field picker. However, selecting the parent hierarchy does not add the complete hierarchy to the visual. Users must expand it and add each level individually in the correct order. Expected Behavior Users should be able to select Responsible Division Hierarchy and have Power BI automatically add the complete hierarchy to the visual, preserving the order defined in the semantic model. Individual hierarchy levels should remain selectable for users who only want a specific field. Why This Matters Personalize Visuals is designed to enable self-service reporting without requiring report editing permissions. Requiring users to manually reconstruct existing hierarchies creates unnecessary friction and assumes they understand the underlying data model. Since the hierarchy is already defined in the semantic model and recognized by the field picker, Power BI should leverage that metadata. Suggested Enhancement Allow users to select either: The parent hierarchy to add the complete hierarchy in its predefined order. An individual hierarchy level to add only that field. This would make Personalize Visuals more intuitive for business users while preserving flexibility for advanced users. Thanks for considering this enhancement!
Matrix Table where measures are not aggregated by month
I was wondering if this would be possible to accomplish in Power BI. I want to be able to have a slicer that the user filter maybe on 3 months 3 years and be able to see the sold by month and weight for the months selected, also adding measures. Measure could be at the end not necessary in the middle as view below. In the past I have built a matrix measure to order the columns, but usually the months are static and there is not date filter. What I am looking to build. I DO NOT want to look like this where the measure/calculations are aggregated by month. Any thought or way to go about this would be greatly appreciated.
