Issue with SUMX including null values in formula

[Anonymous]

I have the following dataset that includes a date, a store ID, and a profit value for that month. One site has not compiled it's report yet for the time frame:

Date	ID	Profit
10/1/2019	1	-25
10/1/2019	2	24
10/1/2019	3	0
10/1/2019	4
10/1/2019	5	-1
11/1/2019	1	34
11/1/2019	2	14
11/1/2019	3	3
11/1/2019	4
11/1/2019	5	-5

My goal is to countthe number of sites in a given time period that are profitable. Profitable in this case is anything that is greater than or equal to 0. My current measure for this is as follows:

Positive Values = SUMX(

VALUES(Table[ID]),

IF(CALCULATE(SUM(Table[Profit])) >= 0, 1 , BLANK()))

Here's the expected values I should get with the measure given each time frame on a Date Slicer:

10/1/2019: 2

11/1/2019: 3

10/1/2019 - 11/1/2019: 3

However, the SUM formula is including the null values, giving me values of 3, 4, and 4, respectively. I'm not sure where to put the FILTER(ISBLANK(Table[Profit]) = FALSE) conditional in the measure to account for this. Anyone have any suggestions?

[Anonymous · Accepted solution]

Positive Values = SUMX(
SUMMARIZE(

    FILTER(

        Table,

        NOT(ISBLANK(Table[Profit]))

    ),

    Table[ID]

),
IF(CALCULATE(SUM(Table[Profit])) >= 0, 1 , BLANK()))

[Anonymous]

Thank you very much!