Earn the coveted Fabric Analytics Engineer certification. 100% off your exam for a limited time only!
I would like to calculate my margin of the current year and devide it with the amount of customers without a date filter.
I have tried:
Margin = (Calculate(SUM(Table[Margin]);Table[Date]=YEAR(TODAY()))/(DISTINCTCOUNT(Table[Customer]))
However this turns out with "(BLANK)"
I have tried:
Margin = SUM(Table[Margin]/CALCULATE(DISTINCTCOUNT(Table[Cusomter];ALL(Table[Cusomter]))
And use visual level relative date filter: Is in this Year
However it still does not counts all the customers from before 2018.
What do I have to change in this formulas?
Regards,
Guido
Solved! Go to Solution.
Ok, for that, this works
Margin divided by total count of customers = DIVIDE ( SUM ( MarginTable[MarginSum] ), CALCULATE ( DISTINCTCOUNT ( MarginTable[Customer] ), ALL ( MarginTable[Date] ) ) )
The denomitor is not affected by the date columns now.
See the dummy table I created and the result table
Hi,
Does this work?
Margin = SUM(Table[Margin])/CALCULATE(DISTINCTCOUNT(Table[Cusomter]);ALL(Calendar))
Calendar is a Table with running dates. In that Table, create Year and Month columns. Create a relatioship between the Date column of the base data table to the Date column of the Calendar Table. In the filter section, drag year from the Calendar Table.
Try this
Measure = DIVIDE ( SUM ( 'Table'[Margin] ); DISTINCTCOUNT ( 'Table'[Customer] ) )
Then drag the date field into the canvas and create a date slicer like so
You have your dates and the average margin per customer.
It still does not return the preferred outcome.
When I add a slicer or filter the visual with a date column it will have affect on both the Sum of margin and the count of customers.
However, the date must only apply on the margin and not on the customer count.
Ok, for that, this works
Margin divided by total count of customers = DIVIDE ( SUM ( MarginTable[MarginSum] ), CALCULATE ( DISTINCTCOUNT ( MarginTable[Customer] ), ALL ( MarginTable[Date] ) ) )
The denomitor is not affected by the date columns now.
See the dummy table I created and the result table
User | Count |
---|---|
139 | |
113 | |
104 | |
77 | |
63 |
User | Count |
---|---|
135 | |
126 | |
110 | |
70 | |
61 |