Analytics
1.8K 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

Weird spike in analytics

Visitor ✭ ✭ ✭
# 1
Visitor ✭ ✭ ✭

Hello,

 

We have a weird spike in traffic from an old email campaign. Not sure why traffic is still coming through. Can anyone explain?

Screenshot below.

 

Screen Shot 2017-02-20 at 3.29.44 PM.png

 

Thank you

1 Expert replyverified_user

Weird spike in analytics

Top Contributor
# 2
Top Contributor

GA uses last non-direct attribution, so if these visits are from someone that usually visits direct, all of their visits are attributed to the previous referral (your email).

Mike Sullivan, Google Analytics Top Contributor
Was my response helpful? If yes, please mark it as the ‘Best Answer.’ Learn how here.

Weird spike in analytics

Visitor ✭ ✭ ✭
# 3
Visitor ✭ ✭ ✭

So you are saying if 10 people opened that email the day we sent, clicked on the link and then all following direct sessions from those people would come in through email?

 

That seems very odd plus, given the amount of sessions we are getting, seems like up to 30-40 people are doing the same.

Weird spike in analytics

Top Contributor
# 4
Top Contributor

Yes, that explanation may not be what you are seeing, but it does work that way.

 

I say this because I have seen my own (many, many sessions) website visits being recorded as email since I once upon a time clicked a link to test it....GA has remembered ever since.

 

I have never seen anyone study the impact, but I wonder what happens on mobile devices when you open many tabs and turn off/on your phone, 'refreshing' the browser tabs each time...I know the mobile browsers are now prefetching and other magic. Maybe that plays a part? Just guessing out loud. 

 

Check the Secondary Dimension in your reports - see if you can track it down to many, many sessions from one or two cities (users) or if it is many.

Mike Sullivan, Google Analytics Top Contributor
Was my response helpful? If yes, please mark it as the ‘Best Answer.’ Learn how here.

Weird spike in analytics

Visitor ✭ ✭ ✭
# 5
Visitor ✭ ✭ ✭

Good idea.

Thank you.

Weird spike in analytics

Visitor ✭ ✭ ✭
# 6
Visitor ✭ ✭ ✭

So 61% of the users are from mobile (in last year) and 52% of those are using Safari (in-app). That tell you anything?

 

In regards to city, it is widely spread out.

Weird spike in analytics

Top Contributor
# 7
Top Contributor

Sounds like users bookmarked the link from the email...that happens to contain the URL parameters.  BUT you have 93% new sessions...and that means they are not return visitors. Bizarre...

 

Note that your report covers a long time period of well in excess of the threshold that sampling kicks in at (500,000 sessions). Your report is highly sampled, so any specific traffic metrics can be wildly wrong. Try very short time frames to see if the numbers are consistent, or if this is the sampling at work.

Mike Sullivan, Google Analytics Top Contributor
Was my response helpful? If yes, please mark it as the ‘Best Answer.’ Learn how here.

Weird spike in analytics

Visitor ✭ ✭ ✭
# 8
Visitor ✭ ✭ ✭

For just February 2017, it's 51% mobile users, but much less people using Safari in-app.

 

In regards to bookmarks, that's a lot of people bookmarking. Hard to believe.

 

Looking at this again, I am seeing other past email links come up. Perhaps you are right that anyone on mobile who has clicked an email link, anytime they come back, GA is reporting it as email traffic.

 

Sounds about right?

Weird spike in analytics

Top Contributor
# 9
Top Contributor

what about the other 49%? Still a little weird.  Might be a function of browser auto-complete. Overall, it looks like valid traffic from real users that once upon a time read your email and clicked the link. Nothing further to learn by digging deeper...

Mike Sullivan, Google Analytics Top Contributor
Was my response helpful? If yes, please mark it as the ‘Best Answer.’ Learn how here.

Weird spike in analytics

Visitor ✭ ✭ ✭
# 10
Visitor ✭ ✭ ✭

Thank you Mike.