Hi, is there a better way of doing this?
Lets say I have to identify the current application stage of candidates in our system. To do so, I'll first flag candidates with an IF-statement that evaluates 1 on their most current status (based on date and status name). However, the calculation is incomplete because the total does not sum correctly. Therefore, I add an additional step to iterate with SumX.
My question is, can I boil this into one tidy DAX calc rather than two? I've tried but haven't been successful. To be clear, I want an end table that lists ALL the status available, and sums the current status.
PBI file: Example File
Solved! Go to Solution.
It's fairly trivial to combine the measures you have into one by expanding [Flag] inside [Total] but I'm guessing you're looking for something cleaner.
I propose the following:
StatusCount =
VAR Summary =
SUMMARIZE ( ALL ( Data ), Data[Applicant ID], "MaxDate", MAX ( Data[Date] ) )
RETURN
CALCULATE (
COUNTROWS ( Data ),
TREATAS ( Summary, Data[Applicant ID], Data[Date] )
)
First, we compute a Summary table to get the maximal date for each distinct applicant. Note that I've used ALL since, otherwise, we'd only get the rows corresponding to the current Application Status filter.
Bearing this in mind, inside of COUNTROWS, Data is a subtable already filtered on Application Status but we wish to filter it further by only considering the most recent status.
To do this, I use TREATAS to treat the [Applicant ID] and [MaxDate] columns of Summarize as the only set pairs allowable for Data[Applicant ID] and Data[Date].
This simple-looking measure hides quite a bit of subtlety. I recommend reading this article to get a better handle on this sort of pattern:
https://www.sqlbi.com/articles/propagate-filters-using-treatas-in-dax/
It's fairly trivial to combine the measures you have into one by expanding [Flag] inside [Total] but I'm guessing you're looking for something cleaner.
I propose the following:
StatusCount =
VAR Summary =
SUMMARIZE ( ALL ( Data ), Data[Applicant ID], "MaxDate", MAX ( Data[Date] ) )
RETURN
CALCULATE (
COUNTROWS ( Data ),
TREATAS ( Summary, Data[Applicant ID], Data[Date] )
)
First, we compute a Summary table to get the maximal date for each distinct applicant. Note that I've used ALL since, otherwise, we'd only get the rows corresponding to the current Application Status filter.
Bearing this in mind, inside of COUNTROWS, Data is a subtable already filtered on Application Status but we wish to filter it further by only considering the most recent status.
To do this, I use TREATAS to treat the [Applicant ID] and [MaxDate] columns of Summarize as the only set pairs allowable for Data[Applicant ID] and Data[Date].
This simple-looking measure hides quite a bit of subtlety. I recommend reading this article to get a better handle on this sort of pattern:
https://www.sqlbi.com/articles/propagate-filters-using-treatas-in-dax/
Hey @AlexisOlson , thanks for the response.
Let me clarify, how would we output a measure that displays all the available application status and a count of the current status of each application? So when I apply application status row context, the status without any "active" applications does not drop out.
So your concern is that it returns a blank instead of zero for Screen?
You could just add " + 0" to the end of the measure I suggested. More discussion of 0 vs blank here:
https://www.sqlbi.com/articles/how-to-return-0-instead-of-blank-in-dax/
Wow, I never thought of just adding 0 to the end.
Thanks!
User | Count |
---|---|
78 | |
43 | |
36 | |
14 | |
13 |
User | Count |
---|---|
86 | |
31 | |
27 | |
18 | |
13 |