Community Spring Cleaning week is here! Join your fellow Maveryx in digging through your old posts and marking comments on them as solved. Learn more here!

Alteryx Designer Desktop Discussions

Find answers, ask questions, and share expertise about Alteryx Designer Desktop and Intelligence Suite.
SOLVED

Split data by given level of detail

onanm
6 - Meteoroid

Hi,

 

I am trying to split my data into two groups as 'best option' and 'least option' by given level of detail. I am planning to apply some calculations/transformation to 'best option' data table, then simply union with 'least option'. Below is a simple table that illustrates what I am trying to do. In this table, 'Actual' scenario has priority over 'Budget' scenario given same company and date. In other words, for a given company and date, if I have both 'Actual' and 'Budget' scenario, choose 'Actual' scenario. I have highlighted the cases below. But I do not want to remove least option of 'Budget'. I just want to keep as separate data table to be later added to analysis

 

Thanks a lot for your help.

 

CompanyDateScenario
CompanyA31.12.2018Actual
CompanyA31.12.2018Budget
CompanyA31.03.2019Budget
CompanyB30.06.2017Budget
CompanyB31.12.2017Actual
CompanyB31.12.2017Budget
CompanyB31.03.2018Budget
CompanyC30.04.2019Budget
CompanyD30.06.2019Actual
CompanyD31.12.2019Budget
CompanyD30.06.2020Actual
2 REPLIES 2
AngelosPachis
16 - Nebula

Hi @onanm ,

 

What you can do is to use a summarize tool to count the number of scenarios per company and date. Then with a filter tool, you can isolate the Budget scenarios for those companies that had more that one scenarios on a certain date 

 

So out of the T output anchor you will have

 

AngelosPachis_0-1616878139564.png

 

And out of the F output anchor you will have the rest of your records.

 

AngelosPachis_1-1616878174153.png

 

You can go on and process them separately, before bringing them back together with a union tool.

 

Hope that helps,

 

Angelos

onanm
6 - Meteoroid

Perfect Angelos, thanks a lot!

Labels