How can I filter Hierarchy when using Full Null for measures

#1

Hello,


I have 4 diferent metrics arrange really close so they look like one table, but I have to have them in separate metrics because they have different fields to be filtered by. The thing is that I use the Full null option so all the rows are the same for all the metrics (They all have the one same row, Agents)

but when I filter by another field (Group), wich is a slicer in my metric, If I have the full Null option It shows me alla the rows even if they don't belong to that group, with null but it shows and I would like that I could only see the ones that belong to that group (even if they have null in their measure).


If I use the source null options it only show me the ones in that group but not the ones that have zero in that measure so.

If I dont filter by anything it's ok because it shows me all the agents with zero on the ones that have null.



This is what I get if I have all full nulls and I filter:


It shows me all of the agents even if they dont belong to the group Im filtering for, but only show me the values for those who belong to the group.

Image title
ivZ1XnTTMmv2j5XayeVW_full_null2.PNG904×447

This is a visual example of what happen if I have the first metric with Full nulls, and the next 2 with source nulls


Image title
7LWhayQSS5udar57OYbN_full_null.PNG895×420

so as you see, with source nulls is showing me only the ones that belong to that group but one the ones that have values. so the values are not matching with the agent to whom they belong.


Do you have any ideas of what could I do about this?


Thank you!!

#2

It looks like you are trying to find a way to keep the metric sets in sync which may doable but probably tricky. Since all of the metric sets are showing the agents on the rows you may be able to join your 4 data cubes using the agent field. You can join data cubes simply by dragging those onto a single data cube and defining a join transform between those. Will that option work for you?

#3

Hi,

I don't think so, because they are all in the same data cube, just different metrics because they are counts and have differents 'where' conditions and also diffferent fields to be filtered by..