How to create a Percentage Measure using Distinct Count?

I am trying to create a simple percentage measure and display in a table visualization. The measure is calculated by dividing two distinct counts - the denominator is a distinct count of clients within a time period, and the numerator is a distinct count of clients within the period who satisfy a condition. In other words, the numerator is just the denominator with a filter applied.


I can't figure out how to do this in Dundas, because the table visualization only allows 1 metric set - I am able to create the denominator simply by using a Distinct Count aggregator on my desired dimension. How can I create the numerator - by applying a filter to the denominator - while maintaining the distinct count?


I need this calculation to be done within the metric set (not in the data cube) because I want the distinct count to change based on the selected period within a time dimension.

Hello Dave,


You cannot have two dimension count measures on the same metric set so we will have to create two separate metric sets each with their own counts, and create a third metric set as your formula visualization:

  • Create a metric set displaying the distinct count of clients like you mentioned
  • Create another metric set exactly the same as the first metric set and have your filter applied here as a slicer
  • With these two metric sets displayed on a dashboard, create a formula visualization calculating the percentage:
    • Ex. $chart2.Metric Set 6.CustomerID$ / $chart1.Metric Set 6 - Copy.CustomerID$
  • Change the formatting of the newly created “formula measure” in the visualization to P0 (automatic percentage calculation and format)

Image title


  • When using a slicer in your two metric sets, the formula visualization will create a field where you can reference the metric sets slicer and have the data dynamically change as you change the filter.
  • You can add filter from the toolbar -> Filter -> Date/Time -> Range Date, and connect the filter to the slicer in your metric set.