Mobile App Login vs non logged in

[gauthamraj]

Hello everyone, 
I am trying to understand how to analyze the login behavior. 
for example
if I use unique visitors as metric 
and have 2 segments with an event to differentiate if a user logged in or not 
then this user will fall into both the segments

segment 1- before login 

segment 2- after login

Is there any other way ?

[Jennifer_Dungan · Accepted solution]

That is the problem with Visitors.. visitor is a essentially a lifetime metric, but the logged in state fluctuates between hits...

If you only care about users who have ever logged in, and treat those as logged in, you could do a segment looking for visitors with a logged in hit... and for logged out, you could look for logged out, but add an exclusion for visitors who have ever logged in...

But since the login might have been 2 years ago, I am not sure that is really valid...

You could also potentially add a date qualifier to the above segments... of "today", or "last 7 days", etc.... 

That would make this a little more realistic... 

[Jennifer_Dungan]

Actually, you could probably restrict based on visit....

Something like:

and

 (basically visits where the visit has not logged in, AND excludes visits that were logged in within the visit)

These may need a little tweaking, I haven't had a chance to really play with these... but in theory this should get you close.