Forum Discussion
RANKX in Card
- 6 years ago
In case there´s someone else with the same problem. There´s something wrong when you try to rank decimals. To solve the problem we had to change the measure in the overall rank to integer.
Thank you Anonymous. I tried this but it doesn´t work. I believe there´s something else. I tried the same solution that you gave me in another dataset and it works perfect, but in this one it doesn´t. The only difference that I can see is that there are negative values in the Sales mesasure in this one, could that be the problem?
I don't think it is because of the negative sales figures. To test this quickly is to wrap the ABS function on Sales and see if this works as expected. If not, any chance you could attached the PBIX file in here with removing the sensitive bits. I believe it may be because of the filters/slicers that are passing through which can be viewed by clicking on Edit Interactions and hovering on the card visual.
Thanks
Ashish
- JOKA7 years ago
Advocate I
Hi Anonymous ! I still can´t solve this problem. I tried your recommendation in another dataset and it works perfect. But in this one it doesn´t.
I am sharing the link to the .pbix file: https://drive.google.com/file/d/155T88JrH1mciasgj3Dcx_7ZHwZwxV842/view?usp=sharing
Hope you could help me. I would be very grateful!
- JOKA6 years ago
Advocate I
In case there´s someone else with the same problem. There´s something wrong when you try to rank decimals. To solve the problem we had to change the measure in the overall rank to integer.
- melmwood3 years agoRegular Visitor
Hours. Tons of hours. That's what it took to finally find this post. And guess what - after three years, MSFT still hasn't fixed this comical issue. Had to set Sales $ column in "advanced editor" to be an integer like you said (Int64.Type), and magically my RankX calc that was ranking brands would finally give the right rank at all times when isolated to a single brand in a card. It's like the BI and Excel teams have never talked.