RANKX with customizable variable

[cleytonp]

Hello everyone, I'm trying to create a dynamic ranking with variables.

I currently have a table with category and total and it has a measure that shows the ranking of each one.

However, I have an option to select values ​​in the case of categories, which must add up, for example if I select the category or multiple categories, the original rank will dynamically update according to the selected one.

here the file with data and measures i used.

pbix gdrive 

[cleytonp]

Thanks @lbendlin.
The case is to update this table with A + C + E + F which would give the total value of 16 and would be number 1 in the ranking, looking as follows.

How can do this?

In other words, the ranking remains the original and the selected categories become a variable adding the name of the category and the value it has in each one

[danextian · Accepted solution]

Hi @cleytonp 

In order for those combined category to appear based on the slicer selection, they need to physically exist first in your which means that there should be a column in your model showing all the possible combinations of those categories.  Example:

[lbendlin]

Adding the same data to the same visual twice is a design red flag.  Don't do it.  Use separate visuals (as you already did)

[cleytonp]

@lbendlin I don't understand, because the idea is to add it to the table where the visual already exists, that is, I keep the table and reorder it according to the selectable.

This is the final result that I expect in the case you mentioned with the selected options, that is, I add the sum of each selected category, this way it is top 1 in the ranking.

[cleytonp]

If I do not select any item, the table continues without changing the ranking, as follows.

However, if I select an item it must be iterated and added to its total and moved to a new position in the ranking, example image below.

[lbendlin]

I cannot assist you with this request, I don't think that is right. Maybe someone else can help you further.

[cleytonp]

Thanks for your help @lbendlin.

It really was as far as I got and I wasted a few days on this, I hope someone can help me.
grateful

[lbendlin]

You were close