Hello
I have built a tabular model and am trying to develop a KPI (using SSDT). My problem is that when I drag the KPI that I have developed onto Excel (or import into PerformancePoint, or create a KPI from scratch in PerformancePoint based off a measure), all the dimension values are shown, and not just those dimension values that are linked to the KPI (via its measure).
For instance, consider the AdventureWorks data warehouse: the FactResellerSales fact has EmployeeKey and SalesAmount fields. Pretend that I only have 100 records in FactResellerSales, but have 1000 records in DimEmployee (i.e.: +-1000 employees). As such, not every record in DimEmployee will be referenced by FactResellerSales.
If I have a measure called Average Sales (Average Sales:= AVERAGE([SalesAmount])) and make this my KPI, when I drag DimEmployee.EmployeeNationalIDAlternateKey onto my pivot table in Excel (or my KPI in PerformancePoint), every employee record is shown, and not only those that are linked to by the fact table where the KPI resides. This results in many NULL values in the KPI. This doesn't happen when I drag only the measure (and not the KPI) to the pivot table - only those employee records that are linked to the measure are shown.
Is there a way to get around this? Adding EmployeeNationalIDAlternateKey as a new column to FactResellerSales fixes this if I then drag this field on instead of DimEmployee.EmployeeNationalIDAlternateKey, but it isn't ideal. The dimension that I am working with has 100k+ records, and only a small subset of these are relevant to a particular fact where the KPI has been created, which means that 100k+ lines are displayed, most of them null.
Thanks!
