If Statement Total does not add up

[lionpbiz]

Hi,

I know there have been lots of questions about this topic already but I just can't figure out how to apply the solutions to my own problem.

I have this IF statement here:

calc = 
IF(
        MAX(DateTable[Exp_Dates]) >= MAX(A[Closed_Date])
        , [USD_AP] - [USD_FC]
        , [USD_AP]
    )

When now building a table I get this:

But what I want is the column sum of the calculated values displayed, so 155MM + .591MM. How do I do that?

Thanks!

[DataInsights]

@lionpbiz,

Try this measure. It iterates DateTable, so totals will be the sum of parts. Not sure if you need MAX for A[Closed_Date] since this may depend on your business logic.

calc =
SUMX (
    DateTable,
    IF ( DateTable[Exp_Dates] >= A[Closed_Date], [USD_AP] - [USD_FC], [USD_AP] )
)

Excellent article on totals:

https://www.sqlbi.com/articles/obtaining-accurate-totals-in-dax/ 

[lionpbiz]

Thanks for the suggestion, but unfortunately this provides completely different results for me altogether.

[DataInsights]

@lionpbiz,

Are you able to provide sample data (table format or pbix link), screenshots of your model, and business logic?

[lionpbiz]

So I have two measures which work fine: USD_AP and USD_FC

Then I have my condition where basically when 2023, then use USD_AP and when 2024 then use USD_AP-USD_FC.

But for some reason once I do this IF statement the sum gets completely messed up eventhough the individual underlying measures work perfectly.

[DataInsights · Accepted solution]

@lionpbiz,

Here's an example you can adapt to your model. It evaluates the IF statement on a yearly basis. If you need it to evaluate at the day level, change "Year" to "Date".

calc = 
SUMX (
    VALUES ( DateTable[Year] ),
    IF ( DateTable[Year] >= 2024, [USD_AP] - [USD_FC], [USD_AP] )
)

Replace the hardcoded "2024" with the correct expression. I'm not sure what your table "A" is and how it's related to your model. The measure USD_AP has an incorrect total in your screenshot.

[BiMica]

Worked for me. Spent a lot of time on this, thanks!