TopN based on Measure instead of Column

[Anonymous]

Hi everyone, I am struggling with the TopN function and I hope you can help me out.

Data: event registration data, with calculated columns for CY (current year) and LY (Last year). I also created a measure to show the count difference between them, “CY-LY Reg Diff”

Goal: create a card that displays the region and store type with the largest deficits year-on-year (smallest value for “CY-LY Diff” measure)

Challenge: I have tried this formula, but it gives me “blank” as a region.

Largest Deficit Region = CALCULATE(SELECTEDVALUE('Sample Data'[Region]),TOPN(1,'Sample Data',[CY - LY Reg Diff],ASC))

Year	Store Type	Region	Registered	Last year	Current year
Year 1	Store Type A	West	1	1	0
Year 1	Store Type C	South	1	1	0
Year 1	Store Type C	Midwest	1	1	0
Year 1	Store Type B	West	1	1	0
Year 1	Store Type B	West	1	1	0
Year 1	Store Type B	West	1	1	0
Year 2	Store Type C	West	1	0	1
Year 2	Store Type B	Midwest	1	0	1
Year 2	Store Type B	South	1	0	1
Year 2	Store Type C	Midwest	1	0	1

Any help or tips would be greatly appreciated.

Thank you.

[Greg_Deckler]

Could you use ADDCOLUMNS to add your TOPN column and then take the MIN or MAX of it?

[Anonymous]

Hi, I must be doing something wrong here. I added ADDCOLUMNS to the expression, but now it looks like it's using it as multiple columns- I thought that I only added one. 😞 

= CALCULATE( SELECTEDVALUE('Sample Data'[Region]), ADDCOLUMNS('Sample Data',"Top N",TOPN(1,'Sample Data',[CY - LY Reg Diff],ASC)))

Error message: "The expression referrs to multiple columns. Multiple columns cannot be converted to a scalar value."

I must be missing a step or three. 

[Anonymous · Accepted solution]

This piece of code:

Largest Deficit Region = 
CALCULATE(
    SELECTEDVALUE('Sample Data'[Region]),
    TOPN(1,
         'Sample Data',
         [CY - LY Reg Diff],
         ASC
    )
)

returns BLANK because SELECTEDVALUE returns BLANK when there is more than 1 region returned.

Change the expression under CALCULATE to show the number of regions - distinctcount( 'sample data'[region] ) - and you'll see it's not 1.

Please read this: https://dax.guide/topn/

Best

D

[Anonymous]

Thank you- I was actually able to get it to work!