AdWords is now Google Ads. Our new name reflects the full range of advertising options we offer across Search, Display, YouTube, and more. Learn more

Analytics
5.3K members online now
5.3K members online now
Ask questions about filter set-up and issues with using filters in Google Analytics reports
 
Guide Me
star_border
Reply

ga:sessions != ga:users for ga:userType==New Visitor

[ Edited ]
Visitor ✭ ✭ ✭
# 1
Visitor ✭ ✭ ✭

According to the documentation at https://developers.google.com/analytics/devguides/reporting/core/dimsmets :

 

Metric New Users
The number of sessions marked as a user's first sessions.

Dimension User Type
A boolean, either New Visitor or Returning Visitor, indicating if the users are new or returning.

Metric Users
The total number of users for the requested time period.

 

Since there should be only one first session for every user, this suggests that the number of sessions (ga:sessions) should be equal to the number of users (ga:users) for ga:userType==New Visitor or am I missing something? Even for small time periods ga:users is usually smaller than ga:sessions for the segment "New Users" (ga:userType==New Visitor). Is there a correct interpretation of this discrepancy?

1 Expert replyverified_user

ga:sessions != ga:users for ga:userType==New Visitor

Rising Star
# 2
Rising Star
 
Sessions should not equal users of any type except in a momentary time frame where all users are new. 
 
Users have sessions.  These sessions occur over time and the date range you select will impact your user count.    e.g. 10 users on Monday, 17 users on Tuesday, 12 users on Wednesday.   Users for Monday - Wednesday = 19.  This happens because many users came back each day.
 
All the while, each user (new or returning) could have come back several times a day. 
 
Lastly, if you consistently see two or more sessions a day from new users and this is unexpected, you could have a coding problem (missing GATC on some pages) or God forbid, you may be using utm parameters on links within your own site.  (This would cause 1 session to end and a new to begin).
 
Best,
 
Theo Bennett
Analytics Evangelist at MoreVisibility | Contact Me
Connect on LinkedIn

Re: ga:sessions != ga:users for ga:userType==New Visitor

Visitor ✭ ✭ ✭
# 3
Visitor ✭ ✭ ✭

Hi Theo,

 

Thanks for your reply.

 

Perhaps I have not stated the problem clearly, but this should not have anything to do with the coding and first sessions should in this case, by definition, be equal to new users. For full data samples this has proven true so far (ga:sessions==ga:newUsers for ga:userType==New Visitor).

 

The problem is that ga:users<ga:newUsers for ga:userType=New Visitor, which, as far as I can see, is either not properly documented or should not be the case. According to the definition, the first session of each user is equal to a "New User". Only distinct, known users are counted in ga:users, but if a user was already known a it would not be the first session and so it follows:

 

ga:newUsers==ga:sessions and ga:sessions==ga:users for ga:userType==New Visitor

 

While the first equation is true for the second equation it is unfortunately more something like ga:sessions>=ga:users>0.99 * ga:sessions for 500<ga:sessions<500'000 and I was wondering about the explanation for this occurrence.

 

Thanks again for the help,

Hansueli Fuchs