Month by Month Data Comparison.

[#1]

Hi, I'm trying to figure out a solution for the issue I'm going to explain.

I have a log file which reports, day by day, all the customers accesses to a specific platforms.

What I'm not able to define is a way to determine how can I find out users with no accesses this month and that have made at least an access during the previous one.

Do I have to implement logics into the load scripts or is it possible to manage it with, let's say, graph expressions?

Thanks for the precious support.

Luigi Zanchi

[chematos]

The expression to use in a table or graphic would be something like this:

Only({1<Month={'$(=Month-1)'}> - $<Flag_Access={1} >} >} Users)

I should now more details about your model and your requirements to make the correct solution you need

[#3]

Hi Josè and many thanks for your feedback.

Few words about the data model, in order to have a consistent focus on what the concern is all about.

I have a table named ACCESS which displays a row every time a Customer_ID connects to the platform, lets say an iPad.

This row shows the timestamp of the access so that I'm able to determine, for instance, the month of the access.

What I would like to do is an istogram which shows for every month how many customers connected during that month have made a connection (at least) the previous one.

I do manage for determining it, but only if I select 1 month, or week: the statement is as the following one:

=count (distinct 

{

<A_Customer_ID = P({$<AccessWeek = {$(#=Only(AccessWeek) -1 )}>}  A_Customer_ID )> *

<A_Customer_ID = P({$<AccessWeek = {$(#=Only(AccessWeek)    )}>}  A_Customer_ID )>

}

A_Customer_ID )

But this works only if a week is selected; the graph of course doesn't show bars if all the weeks are selected.

I'm struggling with it a lot....

Hop this can improve you understanding and many thanks again.

Luigi