Help with Measure from one fact table used in conform dim with filter on another fact

[TinyTimothy2]

I have 2 Fact tables:

Fact_A

Fact_B

I have conforming dimensions

Dim_Date

Dim_User

I have measures that gives count of customers from Fact_A and Fact_B

[Total Customers w/ Fact A]= DISTINCTCOUNT(Fact_A[Cust_ID])

[Total Customers w/ Fact B]= DISTINCTCOUNT(Fact_B[Cust_ID])

And these are used in a matrix with Dim_User and another matrix with dim_Date

In the matrix with dim_User the users names are rows.

I set a filter using a slicer for Fact_A[Program X] = 'XYZ'

This limits the users to only users that have a record in Fact_A[Program x] = 'XYZ'

[Total Customers w/ Fact A] is correct

[Total Customers w/ Fact B] is not correct (it now show same values as [Total Customers w/ Fact A]

This is close. It show the correct Total Customers w/ Fact B. But the grand total is incorrect. It appears to be totaling all users that are filters in the matrix.

VAR _Result_Fact_B = Calculate (
DISTINCTCOUNT(Fact_B[Cust_ID])
, ALL(Fact_A[Program XYZ])
)

VAR _Result = IF ( ISBLANK ( [Total Sales] ) ,
BLANK() ,
_Result_Fact_B)
Return
_Result

Any ideas how to fix this.

Ultimately the 'process' would need to apply to any filter that is not conforming for FACT_A/FACT_B

[mlonsk]

Hi @TinyTimothy2

You seem to have a bidirectional filter set somewhere that propagates the slicer "Fact_A[Program X] = 'XYZ'" onto fact_b.

Is this the same model that you mentioned in #1311? If so, the bidirectional filter mentioned therein would explain this behavior.

Alternatively, could you share a demo model that shows the behavior you see?