Hi Dany,
As far as I know, it is not possible to cherry-pick the selections that should be applied, it is all of them of none for a component. This would mean that you should adjust the formula to so, for instance by using PU-fuctions with a CONSTANT statement (see CONSTANT (celonis.com)) or using a static variable (see https://docs.celonis.com/en/analysis-variables.html#UUID-1cc56c73-87c7-49e3-d99f-439d39ceae31_id_AnalysisVariables-Createastaticvaluevariable).
Hopefully this helps!
Thanks Jan-Peter for your answer.
It is quite a tricky operation, as I have tried using COINSTANT, DOMAIN_TABLE and GLOBAL associated with PU-functions. The difficulty resides on the fact that I still need the KPIs to be responsive to filter on other Columns of the data model, but not on the time column. While using CONSTANT, the value won't be responsive to filters at all. It is the same situation if I use DOMAIN_TABLE on a column having only on distinct element. It is a solution, but I was wondering whether I could have a better one, or if someone have experienced already this issue.
Rregards,
Dany
Thanks Jan-Peter for your answer.
It is quite a tricky operation, as I have tried using COINSTANT, DOMAIN_TABLE and GLOBAL associated with PU-functions. The difficulty resides on the fact that I still need the KPIs to be responsive to filter on other Columns of the data model, but not on the time column. While using CONSTANT, the value won't be responsive to filters at all. It is the same situation if I use DOMAIN_TABLE on a column having only on distinct element. It is a solution, but I was wondering whether I could have a better one, or if someone have experienced already this issue.
Rregards,
Dany
I can imagine that it is suboptimal. One last thing you can check if the function FILTER_TO_NULL to apply all other selections, and make a distinct part in your formula for the time filtering (for instance using variables) instead of global selections. However, this is still not a quick solution to implement. Unfortunately, I'm not aware of the existence of the ideal solution.
Best regards,
Jan-peter