Analytics
4.6K members online now
Understand information in your reports and troubleshoot reporting issues such as self-referrals, (not set) data, and inaccurate information
 
Guide Me
star_border
Reply

User count increases when I shorten the date range. How is this possible?

Follower ✭ ☆ ☆
# 1
Follower ✭ ☆ ☆

I'm trying to look at the # of new and returning visitors during the month of June to a filtered view.

 

When I do not apply any segments and am just looking at all users for June 1 - June 30, I see 29,984 users. I then apply the advanced segments of New Users and Returning Users, while keeping All Users as well. For this 30 day date range, I get sampled data, even at the highest precision.

 

So I decreased my date range to 29 days (June 1-June 29).  I no longer get the sampled data notification, but now for All Users, the user count is showing 30,394. 

 

How is it possible that in unsampled data, I'm seeing a higher user count for all users in a 29 day window than I am in a 30 day window? It doesn't seem to make any sense.

 

Thanks

1 Expert replyverified_user

Re: User count increases when I shorten the date range. How is this possible?

Follower ✭ ✭ ☆
# 2
Follower ✭ ✭ ☆
Hi Leah,

are you using the free version of Google Analytics, or the Analytics 360 version of the tool?
Is it possibile that you are exceeding the quotas for the free version of Google Analytics?
If you send to GA more than 10 million hits per month, then this behaviour could be plausible.

Best regards,

Elena

Re: User count increases when I shorten the date range. How is this possible?

Top Contributor
# 3
Top Contributor
Keep in mind that a New User can return in the reporting period and appear as a Returning User, making it look like there is one more than there actually was.
Mike Sullivan, Google Analytics Top Contributor
Was my response helpful? If yes, please mark it as the ‘Best Answer.’ Learn how here.

Re: User count increases when I shorten the date range. How is this possible?

Follower ✭ ☆ ☆
# 4
Follower ✭ ☆ ☆
Thanks Mike. I'm not adding up New and Returning. I'm leaving three segments: the default All Users, New Users, and Returning Users. It's the default All Users number that increases when I shorten the date range to 29 days and there is no label or indication that the data is sampled.

Re: User count increases when I shorten the date range. How is this possible?

[ Edited ]
Follower ✭ ✭ ☆
# 5
Follower ✭ ✭ ☆

Hi Leah,

I've been reading again your questions, and I noticed that you say "For this 30 day date range, I get sampled data", but that you don't get sampled data anymore if you select 29 days instead.
In this case, is perfectly normal having two different numbers,
Since the "30 days" numbers is calculated basing on a sample of the data, the number of users could be slightly wrong; while the not sampled data (29 days) is correct.

If the number of sessions in the property over the given date range exceeds 250k sessions (25M for Premium)1, Analytics will employ a sampling algorithm which uses a sample set proportional to the distribution of sessions by day for the selected date range.

source: https://support.google.com/analytics/answer/2637192?hl=en

Hope this helps!

Best,

Elena

Re: User count increases when I shorten the date range. How is this possible?

Follower ✭ ☆ ☆
# 6
Follower ✭ ☆ ☆
Thanks Elena. Allow me to clarify:

The first number for the 30 day date range was unsampled for All Users since I had not yet applied any advanced segments. This showed 29,984 users for the 30 day period.

After applying the advanced segments for New and Returning Users, the data was sampled, even at the highest precision. So I decreased the date range to 29 days with the three segments applied (All users, Returning Users, New Users). After decreasing the date range, the data was no longer sampled, yet the All Users count was higher than original All Users count for the 30-day range, without any advance segments applied and without any sampling.

Summary:
1. First Data Set
Range = June 1-June 30
Segments = None
Sampling = False
All Users = 29,984


2. Second Data Set
Range = June 1-June 30
Segments = All Users, New Users, Returning Users
Sampling = True, based on 98% of sessions
All Users = 31, 456
*Not confused about this result because of sampling

3. Third Data Set
Range = June 1-June 29
Segments = All Users, New Users, Returning Users
Sampling = False
All Users = 30,394
Marked as Best Answer.
Solution
Accepted by topic author Leah P
July 2016

Re: User count increases when I shorten the date range. How is this possible?

Top Contributor
# 7
Top Contributor
Leah, this is one of those mysteries of GA...in the document linked below, it mentions that the default numbers shown in the reports are precalcuated. As soon as you apply a segment, you cause GA to go back to the source data and recalculate the numbers. Others have found that the numbers can be a little different.
https://support.google.com/analytics/answer/2992042?hl=en
Mike Sullivan, Google Analytics Top Contributor
Was my response helpful? If yes, please mark it as the ‘Best Answer.’ Learn how here.

Re: User count increases when I shorten the date range. How is this possible?

Follower ✭ ☆ ☆
# 8
Follower ✭ ☆ ☆
Thanks Mike!