Drill through overwrite calculate filter

[ahmadhatahet]

 Hello Community.

I made a cumulative measure using calculate, all calendar and max date, very easy basic DAX.

When I drill through from page to page in Power BI desktop report this cumulative measure doesn't work, more specific All Calendar doesn't work, because I tried using All Data and it works (it gives me the sum of all rows in the table).

The source report has month and year filters, which the target report inherits from the source.

My point of view is the Calendar table filter the Data table because of the relationship,

and the Cumulative function has ALL(Calendar[Date]) filter but the drill through overwrite the filter inside calculate.

Is this what happening?

How to solve this?

[ahmadhatahet · Accepted solution]

I just went to the model and I don't know how but the cumulative works as it should.

A great mystery to me.

Thank you guys, really appreciate the efforts and the time 😊

[ahmadhatahet]

@v-deddai1-msft 

@amitchandak 

Hello again,

I am very thankful to your interest in my issue, and I have a question regarding this.

Is it a matter on how Power BI prioritize the filters?

[ahmadhatahet]

I just tested drilling down from an account, who has a balance at the end of selected month and year, but has no transaction in the same period.

Normal sum = blank [Sure there are no transaction in the selected period]

Cumulative = blank [which is my issue]

All = gives me a total [which is all transaction in the selected period only all data, not all dates and all data, therefore the ALL(Calendar) is not working after drilling]

Once I clear Month or Year filter the data comes as anticipated.

Hop this test be useful in any way.

[amitchandak]

@ahmadhatahet , share Normal sum and Cumulative  formula

[ahmadhatahet]

Thank you for your reply.

[v-deddai1-msft]

Hi @ahmadhatahet ,

What is the relationship between your fact table and calendar table ?

If this post helps, then please consider Accept it as the solution to help the other members find it more quickly.

Best Regards,

Dedmon Dai

[ahmadhatahet]

Thank you for your reply.

[ahmadhatahet]

Calendar > Data (1 > *) [Date column as a key]