Curr Month cume / Prev Months Total

[slians]

Hi there,

I've been racking my brains over this for some time and I can't figure out what's happening. I'm fairly new at PP so please bear with me.

The basic problem I have is I'm trying to get the percentage of returning users from the previous month as a percentage. My dataset is as follows:

Date | Month | User | Day of Month | Stage


With stage being whether a customer is returning or new,
and the pivot I'm working towards looks like this:

Month
Day Jan-12 | Feb-12 | Mar-12
1
2
3
...

my running total users calc is:

CALCULATE(
COUNTROWS(DISTINCT([users])),
FILTER(ALL([day_of_month]),
[day_of_month]<=MAX([day_of_month])
),
[stage]="Returning"
)

and this works perfectly. The issue I have is calculating the total number of users in the previous month. I currently have:
CALCULATE(
COUNTROWS(DISTINCT([users])),
ALL([day_of_month]),
DATEADD([month],-1,MONTH)
)

And this works up to a point - for some reason the last figure that displays in the pivot is the 29th for all months. If I remove Feb from the dataset and go back to after Feb-11 then 30 is the max I get, everything below is blank.

Any ideas why this is happening, and what I can do to get around it? Thanks in advance.

 

[ruve1k]

It would be very helpful if you provided some sample data to work with that illustrates the problem.