Cohorts report dax

[Ritaf1983]

Hi,

In my table, I have the usage of accounts by service level and week.

IdWeekServiceOpen DateIs__Active

1

a	0	aaa	01/01/2023	0
a	0	bbb	01/01/2023	1
a	0	ccc	01/01/2023	1
a	1	aaa	01/01/2023	1
a	1	bbb	01/01/2023	0
a	1	ccc	01/01/2023	1
a	2	aaa	01/01/2023	0
b	0	aaa	01/01/2023	0
b	0	bbb	01/01/2023	1
b	0	ccc	01/01/2023	1
b	1	aaa	01/01/2023	0
c	0	aaa	08/01/2023	0
c	0	bbb	08/01/2023	1
c	0	ccc	08/01/2023	1
c	1	aaa	08/01/2023	0
d	0	aaa	10/01/2023	1
d	0	bbb	10/01/2023	0
d	0	ccc	10/01/2023	0
d	1	aaa	10/01/2023	1
d	1	bbb	10/01/2023	1
d	1	ccc	10/01/2023	0
e	0	aaa	17/01/2023	0
e	0	bbb	17/01/2023	1
e	0	ccc	17/01/2023	1

I need to do a cohort analysis.

The analysis involves calculating the number of active and inactive accounts on a weekly basis with a view that includes both the level of all services and the ability to filter by service.

I need assistance with two challenges in the calculation of inactive accounts:

1. Counting inactive accounts by week, so that an account is only counted if it was inactive in all services when no service selected on the slicer 

2. After accounts are closed, they disappear from the tracking in the following weeks, so I need a way to count them as inactive after they have been "deleted" (last week they appear which I have ).

The solution needs to be with DAX (the original report has millions of rows)
Link to the sample pbix  file with the model: https://drive.google.com/file/d/1Bb4BoLm3ZWvJXbAGYvgMozs67GI3uPTC/view?usp=sharing