Forum Discussion

vinaypugalia's avatar
vinaypugalia
Resolver I
7 years ago
Solved

Power BI Matrix: Filter rows where all columns evaluate to zero

I have a matrix, which has around 6-7 fields plotted as rows (for grouping, expand collapse). The matrix has 10+ columns.

In value, I have a single measure, which returns value based on the column it is plotted for.

ie.

CALCULATE (
        SWITCH (
             SelectedValue,
            "Column1", [measure1],
            "Column2", [measure2],
            "Column3", [measure3],
            "Column4", [measure4],    
            .
            .
            .
            "column20", [measure20]
        )
)

Everything works perfectly here. Now we have a measure, which sums all measures used in the switch case above. This measure is used as a visual level filter for the matrix, so we can remove rows where all values are zero.

This works as well, however, it takes a heavy toll on performance. Thus, I am looking for a more efficient and less performance-intensive approach to filter rows in a matrix where all values are zero.

Any help would be appreciated, thank you.

  • Anonymous's avatar
    Anonymous
    7 years ago

    Hi vinaypugalia ,

    I'd like to suggest you unpivot these 10 column and convert them as attribute and value on query editor side.
    Then you can write measure formula with simple conditions to check all records based on its attribute and correspond values.

    Unpivot columns (Power Query)

    Regards,

    Xiaoxin Sheng

     

1 Reply

  • Anonymous's avatar
    Anonymous
    Not applicable

    Hi vinaypugalia ,

    I'd like to suggest you unpivot these 10 column and convert them as attribute and value on query editor side.
    Then you can write measure formula with simple conditions to check all records based on its attribute and correspond values.

    Unpivot columns (Power Query)

    Regards,

    Xiaoxin Sheng