James, I'm sure Tableau will(should) agree that "struggling with this for two days" is not the user experience they were hoping for. While the pdf you posted does hint at your troubles, it is not sufficient to help us help you solve your problem. I hope you are not so frustrated that you won't dummy up a bit of data and post something that won't break your client's confidential data, but that will allow us to understand your data structure, and problem.
Sorry you're frustrated. Hoping we can help guide you in the right direction...
I just been trying to get a crash course and get a feeling for the application. Would have replied sooner and also did try to attach packaged workbook. Had some internet latency, Point to Point.
Anyway will attach next. Is there logic that can suppress labels on the others.
From what I have read you can't mix category detail with a summary of the same detail.
Top 10 Catsums by cat looks like this
If RUNNING_SUM(MIN([Number of Records])) <= 10 then 1 ELSE 0
Cat Sum Bins looks like this
IF [Top 10 CatSums by CAT] then max(CATSUM)
ELSE 'OTHER' END
END = 1
I combined a copy of CAT to to CAT SUM
seemed to get me closer
Heard a lot of comment about the
To bad my first graph is a bit of a work around.
Just figured out that I could Register and Log In.
Thanks for your response.
package WS has been added to original post.
Looking for any solution possible.
See the attached. This is actually a fairly advanced thing to do in Tableau, so it's no wonder you were running into issues. There are a couple of great threads on this from awhile back:
I used these concepts to build out what I think you're going for in the attached, they use nested table calculations with a couple of different addressing and partitioning options so pay attention to those settings.
top n or other jtd.twbx.zip 2.7 MB
Thanks! Just took an initial look but it looks good.
Message was edited by: James Pigott Jonathan, Can't thank you enough. I did come across the first post at least but didn't quite get to the answer. I looked at a ton of them. Thanks for putting it together for me. Just getting a handle on the basic structures and functionality.
I have a similar problem with the same data. I tried to drop down one level modeling everything as was done in the previous.
I reworked what I had built from scratch. I can see from the workout tab I am missing something in the calculations. I have examined everything I could but could not get the same result. The first version was a general category to a combined major category subcategory (CATSUM).
In this version I am trying do something like Major Category or Subcategory to vendor. The top records display the vendor name, the rest are classified as other.
From the workout sheet the following seems apparent in comparison to Jonathan's previous example.
The percent of total for bar is not summing to 100%.
The index/us amount within top N or other is not working correctly.
If it was good enough to go the graphs I am looking for 100% graphs composed of the vendor names and other.
Any help would be appreciated.
top n next level effort.twbx.zip 641.5 KB
Solved the primary problem now just having trouble with recognizing "not in top n" and using other rather than vendor name in workout tab.
The bar tab recognizes other but is not 100% graph.
top n next level effort3.twbx.zip 874.1 KB