Microsoft Fabric Community Conference 2025, March 31 - April 2, Las Vegas, Nevada. Use code FABINSIDER for a $400 discount.
Register nowGet inspired! Check out the entries from the Power BI DataViz World Championships preliminary rounds and give kudos to your favorites. View the vizzies.
I have a Semantic model that is built (in Desktop) using Dataflows to initially import the data. I don't have Fabric capacity, otherwise I would do all the modeling in Fabric. I have a number of transformations and data shaping in the Semantic Model PBIR.
In the Report editor for the Semantic Model file, I can't add my Fact values to a Card, or to Values on a Matrix, or even create a basic Bar chart with the Fact and a Single dimension. Dragging and dropping doesn't work. If I create a blank report and import the Semantic Model, I get the same experience.
If I connect a blank report to the Dataflows, and simply import the data (no ETL) then I can make a basic chart with the same underlying 'facts' and using dimension IDs (rather than the actual dimensions, since they don't exist without ETL).
I have a report I worked on in early June, that is connected to the same Semantic Model, and complex visuals (like above) are all working fine. I even opened it in PBI Desktop and refreshed to confirm.
I am so very confused about what is going on. Another user in our org is experiencing the same thing. I haven't been able to turn up any useful posts through Google. This seems like a bug, I'm missing something obvious, or somehow I'm still running into some 'imposed' limitation based on licesning or some such....
Thoughts? What am I missing?
Solved! Go to Solution.
Good day @rpiboy_1 ,
Do you, by any chance, have a calculation group in the semantic model? When a calculation group is present implicit measures are denied, i.e. you cannot drag a column onto a visual and you must create an explicit measure.
How this helps
Good day @rpiboy_1 ,
Do you, by any chance, have a calculation group in the semantic model? When a calculation group is present implicit measures are denied, i.e. you cannot drag a column onto a visual and you must create an explicit measure.
How this helps
I've the same problem like rpiboy1 - dataset are imported, transformed, aggregated and defined. Two pages are created (with facts as values). Then on a third page I need some aggregation to replay some data to match data of different rows for some calculation. Works fine (new data set, groupby, ...). Nevertheless up to this moment, my facts (imported and transformed as well as calculated columns) are no longer accepted as numbers - they still are defined as numbers but not accepted as values in y-axes in visuals. I also can't create measures ("no values found in ...") - so I could not create "explicit" measures . When looking at the data table they still are defined as numbers ... Redefining them as numbers doesn't solve the ignorance of the tool. Online I can't find any deployable solution ... any hints where to look at? Peter (actual desktop version of power bi, win 11)
I could solve my problem. It's connected with the import of empty cells in integer columns as far as I understand it. Nevertheless I'm a little bit irritated by the fact, that also other columns a redefined and that I don't find a simple way to redefine the correct data types.
I do!!! It was one of the last things I implemented getting ready for a presentation at a company retreat several weeks ago, ugh! I knew this would come down to something obscure and easy to forget about or not realize! SIGH!
So if I want to use Calculation groups, I have to write measures for all of my Facts... ugh. That leads to this more recent post I made, which is with Measures, how can I still provide end users freedom to determine the underlying aggregation (sum vs. avg, etc.).
I contacted support and we had a lengthy conversation that resulted in continued investigation on their part. The underlying problem seems to be that all of the Numerical fields are no longer being intereprted as a numerical data type, even though all of the numerical properties remain. Quite suspisciously the support person noted that the sigma symbols are all gone from all of the numerical fields, which typically graphically denotes that the field can be summed/averaged/min/max/etc. So the fields are all behaving like strings. Further investigation revealed that is somehow related to the specific Semantic model in the specific Power BI report, however, there no transforms to these fields in the report. When importing the dataflows into a new blank report, everything works as expected, but removing some of the data and re-importing it into Power BI does not solve the problem. Thus we don't have any idea as to the root cause that would suddenly shift the interpertation of the data type. We also tried in Power Query modifying the data type to no avail.
I should further note, I can add Measures to the aformentioned visuals, but not Facts. In the case of the Matrix, it will add the Fact as 'Columns' but if I try to drag it down to 'Values' it seems to 'try' in the UI (I get the little green highlight line where the Field would be inserted, but then nothing happens when I realese the mouse button.
Hi,@rpiboy_1 I am glad to hrelp you.
Unfortunately I tried to reproduce your problem by creating a dataflow, and making a report in desktop and publishing it to the service, and editing the report and its corresponding semantic model in the power BI service, and did not succeed in online your problem: the data in the fact table cannot be displayed in the Value field (visual)
Due to the environment, we are only able to try to reproduce the problem you are having on the forum by using the description of the problem you have given, to find out why the problem is occurring and to deduce a solution.
The problem you are experiencing is very strange, I think it might be related to the data types of the fields in the semantic model (this is just my guess) because you also mentioned that in June there existed the same semantic model that was used as well, but it worked without problems!
Here are some of my guesses:
Could you please check if the problematic semantic model is using some special data types, if there is any problem with the data source itself, and if you are having problems with Power BI Desktop or Power BI Service?
I recommend that you get more help from phone support, where you can connect remotely and talk to technical support one-on-one to find out what's wrong. You don't need to be limited by forum solutions: they don't really reproduce the problem you're having.
Here is my test:
In Power BI Service:
I did not find the issue you raised
I hope you can find a helpful solution soon!
I hope my suggestions give you good ideas, if you have any more questions, please clarify in a follow-up reply.
Best Regards,
Carson Jian,
If this post helps, then please consider Accept it as the solution to help the other members find it more quickly.
User | Count |
---|---|
89 | |
82 | |
47 | |
40 | |
35 |