Magdalena, you've stumbled onto one of those times when it would be great if we could use Measure Values & Measure Name in a calculation like this:
CASE [Measure Name]
WHEN '% of Marked Up along Table(Across)' THEN SUM([Marked Up])
WHEN '% of Total Profit along Table(Across)' THEN SUM([Profit])
WHEN '% of Total Sales along Table(Across)' THEN SUM([Sales])
Putting this on the Label shelf would give you what you want. But alas, we don't have access to these variables. So the best I could come up with was to do them as a series of Mark Annotations with the line and line end set to none. Tedious, but doable. (See attached). FYI, to get the sums to be available in the annotation, you'll need them on the Level of Detail shelf.
Zen Master Joe may have a better solution, he usually does.
Magdalena, the good news is V-8 will allow you to easily put multiple labels on charts so your problem should be solved Q1-2013. Hope this helps,
PS: That case example was intended for the Devs that have been known to lurk around these parts.
Magdalena.twbx.zip 1.1 MB
You are facing this limitation because of your data structure, if you have the capability to transform your data prior to Tableau, then this limit goes away.
The Measure Names/Measure Values pills are performing a unpivot of your data in Tableau, and has limits, but if you perform an unpivot of your data prior to Tableau, then you can get the result and interaction that you want.
For example, if your data source looks like this:
Group value 1 value 2 value 3 a 1 2 3 b 4 5 6 c 7 8 9 d 10 11 12
reshaping it to look like:
Group Measure Value a value 1 1 a value 2 2 a value 3 3 b value 1 4 b value 2 5 b value 3 6 c value 1 7 c value 2 8 c value 3 9 d value 1 10 d value 2 11 d value 3 12
would enable the analysis that you want.
See http://kb.tableausoftware.com/articles/knowledgebase/preparing-excel-files-analysis for more details and http://www.tableausoftware.com/public/blog/2012/08/columns-and-rows-any-difference-1628 for a great explanation.
There are other routes as well if that is not an option, but transforming your data prior to Tableau is ideal in this case. Another route would be to use a static scaffold data source (helpful if you have large data and consistent contents), and data blend your data onto this, another option is to use a dashboard, one sheet for each aggregated continuous pill.
See the attached for some examples.
There are other potential routes depending on your constraints or situation.
Shawn, I am not sure I see how multiple pills on the label in v8 would help in this situation, each mark would have 3 numbers, but I did not use v8 for long, I may be missing an option.
reshape.twbx.zip 48.9 KB
Magdalena, make sure you scroll all the way to the bottom of that first link Joe provided. It links to the Andy's Excel Reshaper, which is the ultimate solution to your issue.