Jump to content

  • Log in with Facebook Log in with Twitter Log In with Google      Sign In   
  • Create Account

Subscribe to HRA Now!

 



Are you a Google Analytics enthusiast?

Share and download Custom Google Analytics Reports, dashboards and advanced segments--for FREE! 

 



 

 www.CustomReportSharing.com 

From the folks who brought you High Rankings!



Photo

Confusing Data In Google Analytics


  • Please log in to reply
6 replies to this topic

#1 qwerty

qwerty

    HR 10

  • Moderator
  • 8,625 posts
  • Location:Somerville, MA

Posted 07 February 2010 - 05:09 PM

I'm using GA to study a client's traffic for the period of November '09 through January '10. Specifically, I was checking on which non-paid keyword phrases brought users to a particular page during that period, so I went to Content > Top Content, scrolled down the list of pages and ran a search on a string from the URL of the page in question. The URL came up in the search results and I clicked on it, bringing me to a Content Detail Page. From there, I clicked on Entrance Keywords, and on the resulting page I filtered the data to non-paid.

The data showed that the top keyword phrase bringing people to this page brought in 197 page views and 134 unique page views. That figure really stood out, because the next most popular keyword in the table only brought in 34 unique page views, and also because the page really isn't very strongly optimized for the keyword.

This made me kind of curious, so I checked the page's current ranking for the keyword, and the results weren't very impressive: 44 on Google, 35 on Yahoo and 58 on Bing.

The thought occurred to me that, since I'm looking at data for a period of three months, maybe the page used to be better optimized for the phrase, and all that traffic came before some change was made.

So I went to Traffic Sources > Keywords, filtered the data to non-paid and ran a search for the keyword, expecting to see a graph of those 134 uniques over the three-month period, with the vast majority of them taking place in the early part of the period.

Instead, I got a graph showing me a measly 9 visits.

So where did the 134 unique page views go? Am I confusing the difference between page views and unique page views? The table showing the 9 visits for the keyword does indicate an unusually high Pages/Visit figure of 21.9 (average for the site is 4), but I thought that a unique page view didn't count multiple visits to a given page during a single session, so the difference between the 197 page views and the 134 unique page views was the number of times someone came to the page more than once in a session.

I could understand if there were more than 134 visits for the keyword, since it's possible different pages ranked for the keyword, but less than 134? And so many less? How does that work?

#2 Jill

Jill

    Recovering SEO

  • Admin
  • 32,987 posts

Posted 07 February 2010 - 06:25 PM

If I'm following you, and I'm not sure that I am, it's possible that what you're seeing is visitors to the page in question and the keyword that got them to the SITE.

I believe that doesn't mean that they actually LANDED on that page, just got to that page eventually after coming in from the keyword that you saw.

Does that make sense for your situation?

#3 qwerty

qwerty

    HR 10

  • Moderator
  • 8,625 posts
  • Location:Somerville, MA

Posted 07 February 2010 - 06:39 PM

I considered that, but wouldn't that mean that the keyword led to more visits rather than less?

For example, let's say a search for [wicked cool stuff] brought 30 people to the site, but that traffic was split between three pages: wicked-stuff.htm (6 visits), cool-stuff.htm (4 visits) and wicked-cool-stuff.htm (20 visits).

Then when I checked entrance keywords for wicked-cool-stuff.htm I should see 20 views from searches on [wicked cool stuff], and when I went to just check that keyword (this is for the whole site, not just one page) in Traffic Sources, I should see all 30 visits instead of just the 20 that went to the one page.

But I'm seeing the opposite. Whether more than one page got traffic from that search or just the one page got it all, how can the number of searches for the site be less than the number of searches leading to the one page?

#4 Jill

Jill

    Recovering SEO

  • Admin
  • 32,987 posts

Posted 07 February 2010 - 10:27 PM

Well the other thing is that lots of times Google Anal. (giggle.gif) is just borked.

I had a specific date period set the other day, and then did the same date period another day and saw different numbers. And then it went back again to the old ones.

Other than that...I got nuthin...

#5 Worldwide

Worldwide

    HR 1

  • Members
  • Pip
  • 1 posts

Posted 08 February 2010 - 07:10 AM

QUOTE(Jill @ Feb 7 2010, 10:27 PM) View Post
Well the other thing is that lots of times Google Anal. (giggle.gif) is just borked.

I had a specific date period set the other day, and then did the same date period another day and saw different numbers. And then it went back again to the old ones.

Other than that...I got nuthin...


I have a similar question,
I have noticed several times that my twitter visits were not tracked , so i started using google url tool to track my data on GA, and i got few visits.

So to cross check my results with GA i did a check on GA , i used google URL tool or tracking twitter visits with a unique campaign name to it , and then added that url to bit.ly shrunked it ,and then checked the visits on bit.ly after tweeting it , in the result of bitly i got 50 visits, so checked stats in GA if same visits were tracked in GA, Surprisingly the data was very less it was somewhere near 15- 20.
I then thought that GA might take some time to track the data . so i chec kthe data for it the next day , the campaign shows the same result ,,

I really dont understand where have the other visits gone ..? how do i track this data in GA for twitter ? Does Ga gives the perfect Analysis data ?

#6 Jill

Jill

    Recovering SEO

  • Admin
  • 32,987 posts

Posted 08 February 2010 - 09:18 AM

Your bit.ly may be counting some bots, which google analytics won't.

But you're doing a good job in adding the campaign tracking codes before you shorten them. It's the only way to get some measure of your social media success...at least as it applies to clickthroughs.

#7 qwerty

qwerty

    HR 10

  • Moderator
  • 8,625 posts
  • Location:Somerville, MA

Posted 08 February 2010 - 10:17 AM

QUOTE
Does Ga gives the perfect Analysis data ?

I don't think any of the applications are perfect, but the only inconsistencies I've seen in the past have come up when I've tried to compare data in two applications. That's when you're bound to see differences based on how different apps define a visit or a page view, and of course there are likely to be discrepancies between applications that read log files and those that run from scripts embedded on the page.

What I'm not accustomed to is seeing discrepancies in the data of a single application.




0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users

SPAM FREE FORUM!
 
If you are just registering to spam,
don't bother. You will be wasting your
time as your spam will never see the
light of day!