Our organization is currently running into a dilemma with how we're using PowerBI. The inability to directly query other PowerBI datasets and queries essentially prevents users from instancing datasets into multiple reports.
We are in a situation where we've defined one large dataset of ~40 queries as our base dataset, let's call this Dataset A. This dataset holds pretty much everything we care to know about our project, and thus can inform nearly every report we'd wish to spin up from it. So we've created Dataset A and have published it up to the web with a report, let's call this Report A.
There are multiple reports we wish to spin off from Dataset A. Each report is for vastly different audiences, and each is fairly robust. So to avoid having Report A be 50+ tabs (and tanking its usability), we broke this out into ~3 different reports. This splitting of the reports is where we start to run into issues.
So, with Dataset A published to our PowerBI web workspace, we created Report B from it. Again, Report A and B are from the same Dataset A, they are simply for vastly different audiences, thus the visuals are structured quite differently. When creating Report B, there are certain things which we simply cannot do on an editing level in the web view, with a common standout being the inability to adjust layering (ex. Bring to Front, Send to Back, etc) while editing in PowerBI web. To get around this, we would then download the report and edit it locally with PowerBI desktop. However, once Report B is re-published to web (with the edits that can only be made on PowerBI desktop), it also re-publishes another dataset since both Report A & B have unique names.
Now after this editing & republishing, there are two identical datasets existing in the PowerBI web workspace, Dataset A and Dataset B. These sets are identical to another, however I can no longer control both reports with just Dataset A now. Report B is now looking to Dataset B, so any changes that may need to be made on Dataset A now need to be brought over to Dataset B, essentially doubling the work required for query upkeep/maintenance.
The ability to directly query published datasets from another datasets would void this doubling of efforts. If Dataset B could simply query the queries within Dataset A, it would not functionally matter that a separate dataset is created once Report B is re-published to web from the desktop. Our organization uses comparable tools that have this ability to query queries (such as Excel). We are currently weighing whether we should go back to using Excel to inform our data-driven decisions simply because this ability to instance queries to multiple reports has saves so much time when it comes to maintaining these queries across multiple reports.
Any help or suggested workarounds here would be greatly appreciated.
