How to have filters with OR condition in Power Bi desktop

[TusharGaurav]

Hi Experts,

I need to use a filtercondition where I want to show data as below:

Product.ProductName ="Apple"

or

Geography.Region = "US".

I need to create a filter for the same in power bi,but I cant create a filter with "OR".

Can you please suggest, how I can overcome this challenge.

Thanks and Regards,

Tushar Gaurav

[DataInsights · Accepted solution]

@TusharGaurav,

Try this measure:

OR Filter =
VAR vFilterTable =
    FILTER (
        CROSSJOIN ( ALL ( 'Product'[ProductName] ), ALL ( Geography[Region] ) ),
        'Product'[ProductName] = "Apple"
            || Geography[Region] = "US"
    )
VAR vResult =
    CALCULATE ( SUM ( FactTable[Amount] ), KEEPFILTERS ( vFilterTable ) )
RETURN
    vResult

https://www.sqlbi.com/articles/specifying-multiple-filter-conditions-in-calculate/ 

[TusharGaurav]

Hi All,

Thanks for your reply.

Your response is really helpfull. 🙂

[Anonymous · Accepted solution]

Hi @TusharGaurav 

The solution @DataInsights  offered is excellent,  and you can also consider the following solution.

If you want to control the filter, you can create a measure.

Measure =
IF (
    OR (
        SELECTEDVALUE ( 'Product'[ProductName] ) = "Apple",
        SELECTEDVALUE ( 'Geography'[Region] ) = "US"
    ),
    1,
    0
)

Then put the measure to the visual filter, and set the condition that the measure is equal to 1.

Best Regards!

Yolo Zhu

If this post helps, then please consider Accept it as the solution to help the other members find it more quickly.

[DataInsights · Accepted solution]

@TusharGaurav,

Try this measure:

OR Filter =
VAR vFilterTable =
    FILTER (
        CROSSJOIN ( ALL ( 'Product'[ProductName] ), ALL ( Geography[Region] ) ),
        'Product'[ProductName] = "Apple"
            || Geography[Region] = "US"
    )
VAR vResult =
    CALCULATE ( SUM ( FactTable[Amount] ), KEEPFILTERS ( vFilterTable ) )
RETURN
    vResult

https://www.sqlbi.com/articles/specifying-multiple-filter-conditions-in-calculate/