Super interestingly, this brings us full circle back to our conversation a few days ago about the use of slicers built on disconnected tables, connected virtually to the relevant fact table(s) via TREATAS.
We can use that exact same technique here - Ad Type I is a normal slicer based on Ad Type, while Ad Type 2 is based on our disconnected table. The left (blue) side of the comparison uses “normal” measures since the Ad Type 1 slicer comes from a field connected within our data model, while the right (red) side adds an additional TREATAS statement to each measure to virtually connect the Ad Type 2 slicer to the Sales table:
Total Sales = SUM( Sales[Line Sales] )
Total Sales Disconnected =
CALCULATE(
[Total Sales],
TREATAS(
VALUES( 'Disconnected Ad Type'[Ad Type] ),
Sales[Ad Type]
)
)
Here’s what it looks like all put together:
See, I told you those techniques would come in handy…
Hope this is helpful. Full solution file posted below.
- Brian
eDNA Forum – Comparative Analysis Disconnected Slicer solution.pbix (525.7 KB)