Forum Discussion
Count unique reference numbers based on date slicer
- 6 years ago
Hi Anonymous ,
First off, thanks for providing clear tables with expectations and what you currently have. That helps us recreate your situation and help you figure out a solution 🙂
Then to your problem. The "issue" with calculated columns is that they are not re-evaluated upon slicing or filtering. They are evaluated upon refresh of the dataset and never thereafter. Measures are (by nature) re-evaluated by every change in filter context. However, Measures also (by nature) return a single value and therefor can not be used in a slicer. Fortunately for us, they are allowed to be used in a Visual Filter.
I created a measure like this:
CountMeasure = VAR _curCustomer = SELECTEDVALUE('Table'[URN]) RETURN CALCULATE(COUNTROWS('Table'), FILTER(ALLSELECTED('Table'), [URN] = _curCustomer))Then I created a Table visual with column URN, Date (as Date) and Count (column). I then dragged the above measure to the Filter pane and I can dynamically filter the table visual by using the FIlter pane and the date slicer. For example, I want to only show customers that have a count of less then 3 for a certain date (in this case, for whole 2019). I expect 2 rows for customer 1 and 2 rows for customer 2 (it has 2 total rows for that year so my Filter pane will let it show us):
You can find my PBIX here, please ignore any other possible tables, they are for other questions.
Let me know if this helps you!
Kind regards
Djerro123
-------------------------------
If this answered your question, please mark it as the Solution. This also helps others to find what they are looking for.
Keep those thumbs up coming! 🙂
Hi Anonymous ,
First off, thanks for providing clear tables with expectations and what you currently have. That helps us recreate your situation and help you figure out a solution 🙂
Then to your problem. The "issue" with calculated columns is that they are not re-evaluated upon slicing or filtering. They are evaluated upon refresh of the dataset and never thereafter. Measures are (by nature) re-evaluated by every change in filter context. However, Measures also (by nature) return a single value and therefor can not be used in a slicer. Fortunately for us, they are allowed to be used in a Visual Filter.
I created a measure like this:
CountMeasure =
VAR _curCustomer = SELECTEDVALUE('Table'[URN])
RETURN
CALCULATE(COUNTROWS('Table'), FILTER(ALLSELECTED('Table'), [URN] = _curCustomer)) Then I created a Table visual with column URN, Date (as Date) and Count (column). I then dragged the above measure to the Filter pane and I can dynamically filter the table visual by using the FIlter pane and the date slicer. For example, I want to only show customers that have a count of less then 3 for a certain date (in this case, for whole 2019). I expect 2 rows for customer 1 and 2 rows for customer 2 (it has 2 total rows for that year so my Filter pane will let it show us):
You can find my PBIX here, please ignore any other possible tables, they are for other questions.
Let me know if this helps you!
Kind regards
Djerro123
-------------------------------
If this answered your question, please mark it as the Solution. This also helps others to find what they are looking for.
Keep those thumbs up coming! 🙂