RankX function across multiple Dimension Tables

// Create a measure that for each
// customer will tell you if
// they are in the top 3 cust for
// a product or not. Then use the measure
// as a filter for a table visual that will
// show you the existing combinations
// of Product and Customer and will display
// the [Total Sales] measure. Filter the
// visual by [Filtering Measure] = 1.

[Filtering Measure] =
var __oneProductAndCustomerVisible =
    HASONEVALUE( Customer[CustID] )
    && HASONEVALUE( Product[ProdID] )
var __result =
    if(
        __oneProductAndCustomerVisible,
        var __top3Customers =
            CALCULATE(
                // Bear in mind that if there
                // are ties with respect to
                // [Total Sales], all the qualified
                // customers will be returned, so
                // there might be more than 3.
                topn(3,
                    VALUES( Customer[CustId] ),
                    [Total Sales],
                    DESC
                ),
                ALL( Customer )
            )
        var __currentCustInTop3 =
            VALUES( Customer[CustID] )
            in __top3Customers
        return
            1 * __currentCustInTop3
    )
return
    __result