Filter with union instead of intersection Issue

I have two slicers in Power BI, and my goal is for them to behave as a union, not an intersection.

For example:

• Slicer A returns 10 people

• When I select a value in Slicer B, I want it to add to the result

• So the total should become 12 people, not less than 10

I understand that this behavior is not supported natively in Power BI, since slicers always intersect filters.

Because of that, I created a filter measure that simulates a union and then used this measure to filter visuals.

This approach:

• Works perfectly for card visuals

• Fails for table visuals

For tables, it becomes extremely slow and eventually fails. This makes sense because:

• The filter measure is evaluated row by row

• For each row, the measure re-evaluates all filter logic again

I tried two different approaches (see below). Both are logically correct and give the expected results, but both are very slow when applied to a table.

My question is:

Is there a better or more efficient approach to achieve a union-style slicer behavior in Power BI without the performance issues caused by row-level filter measures on tables?

Codes

Solution 1:

_SHOW_ROW_FILTER_UNION =
VAR Filter1Applied =
    ISFILTERED ( filter_1_table[source_type] )
        || ISFILTERED ( filter_1_table[level_1] )
        || ISFILTERED ( filter_1_table[level_2] )
        || ISFILTERED ( filter_1_table[level_3] )

VAR Filter2Applied =
    ISFILTERED ( filter_2_table[category] )
        || ISFILTERED ( filter_2_table[sub_category] )
        || ISFILTERED ( filter_2_table[sub_sub_category] )

VAR CurrentPerson =
    SELECTEDVALUE ( fact_table[person_id] )

RETURN
IF (
    NOT Filter1Applied && NOT Filter2Applied,
    1,

    IF (
        Filter1Applied && NOT Filter2Applied
            && CurrentPerson
                IN VALUES ( filter_1_table[person_id] ),
        1,

        IF (
            Filter2Applied && NOT Filter1Applied
                && CurrentPerson
                    IN VALUES ( filter_2_table[person_id] ),
            1,

            IF (
                Filter1Applied && Filter2Applied
                    && (
                        CurrentPerson
                            IN VALUES ( filter_1_table[person_id] )
                        || CurrentPerson
                            IN VALUES ( filter_2_table[person_id] )
                    ),
                1,
                0
            )
        )
    )
)

Solution 2:

_SHOW_ROW_FILTER_UNION =
VAR Filter1Applied =
    ISCROSSFILTERED ( filter_1_table )

VAR Filter2Applied =
    ISCROSSFILTERED ( filter_2_table )

VAR Filter1Persons =
    FILTER (
        DISTINCT ( filter_1_table[person_id] ),
        Filter1Applied
    )

VAR Filter2Persons =
    FILTER (
        DISTINCT ( filter_2_table[person_id] ),
        Filter2Applied
    )

VAR UnionPersons =
    DISTINCT (
        UNION (
            Filter1Persons,
            Filter2Persons
        )
    )

VAR CurrentPerson =
    SELECTEDVALUE ( fact_table[person_id] )

RETURN
IF (
    NOT Filter1Applied && NOT Filter2Applied,
    1,
    IF (
        CONTAINS ( UnionPersons, [person_id], CurrentPerson ),
        1,
        0
    )
)