May I ask if it would be possible to get a detailed explanation of applying this principle to a different type of data?
For example, I would like to see the US Sales totals, and have the ability to filter it to a US state without the ability to select a US territory (Guam, Puerto Rico, etc), but to have the US territory sales remain in the US national totals. How could I do this?
In this short post I cover two different techniques how to do this using a self-data blend and LOD expressions, respectively.
I’ve been trying to figure out how to write about this one, I think I finally have a simple enough scenario to describe: In my world of healthcare delivery, I have things like different payors where I want to know what % of the population is covered by a certain payor (like Medicare and Medicaid), and I don’t really need to show anything about the rest of the population other than have the raw numbers available in the computation. Using Superstore, we can do a equivalent modeling of that using Customer Segment as a stand-in for a set of possible distinct conditions for each patient (Customer). So I want to know what % of total Sales are in a given Segment, being able to filter for any set of Customer Segment(s) I want, and show the sum of the % of total Sales for only my filtered Customer Segments. Ideally ending up with something like this:
Read on for how this goes from relatively difficult in earlier versions to relatively simple in Tableau version 9.