Compare to reference group

Hey addicted87 

Here is a measure you could try. Using some test data at my end, it performed significantly better than the original measure:

number of suppliers in reference group =
VAR FilteredSupplierKey =
    VALUES ( supplier[column_supplier_id] )
VAR SimilarSupplierKey =
    CALCULATETABLE (
        VALUES ( supplier[column_supplier_id] );
        ALL ( supplier[column_supplier_id] );
        SUMMARIZE (
            supplier;
            supplier[column_customers];
            supplier[column_region];
            supplier[column_employees]
        )
    )
RETURN
    COUNTROWS ( EXCEPT ( SimilarSupplierKey; FilteredSupplierKey ) )

This should follow the same logic as your original measure when filtered on a single Supplier ID.

This measure also handles cases where you have filtered on multiple Supplier IDs, by counting the union of "similar" Suppliers.

Does this perform better with your data, and still produce correct results?

Regards

Owen