17 {{ upvoteCount | shortNum }}
17 {{ upvoteCount | shortNum }}

Vast Difference Between Awstats data and Google Analytics data

by carolm
11 replies
  • SEO
    • |
I use Hostgator as my host and they have Awstats that runs from their cpanel. The data collected from Awstats is much greater than my reported stats at Google Analytics.

An example of this from one day this week:

Awstats:visits 33, pages 1398
Google Analytics:visits 10, pages 37

That is one of the best examples of this, but every day, Google Analytics is consistently only a fraction of the Awstats data.

I do regular re-submissions to google from the Attracta option in Hostgator cpanel

Any ideas? And do I have to find out why, and try to fix it?

Thanks in advance

Carol
#analytics #awstats #data #difference #google #google analytics #vast
  • Profile picture of the author carolm
    Well I found the answer to my own question - there are lots of reasons for this discrepancy, I'll list a few here:

    1. Awstats includes crawler visits in their data, and Google doesn't

    2. Google Analytics does not collect data from visitors without JavaScript , or with JavaScript disabled

    3. Many of your techy visitors will have blocked Analytics

    4. Awstats only counts those who download the page from the server when they visit, and so will not count anyone who has the page cached from their last visit, or who accesses the page via a caching service. Therefore, Awstats data will also be lower than the correct count, but by not as much as Analytics is.

    How about Alexa? Apparently it's even more unreliable than either Awstats or Analytics, due to limitations on their data gathering methods.

    Take home message: use either or both to track trends, and most visited pages etc but not one of these sites gives you accurate numbers.

    If looking to sell advertising space on your site, quote both sets of figures to prospective advertisers - or ask them which they would like.

    For my own sites the Ratios for Analyics:Awstats are:

    One website: using tomatosoft Analytics plugin
    Unique Visitors: 0.389
    Visits:0.165
    Pageviews:0.226

    Other website: using Joost Analytics plugin
    Unique visitors:0.297
    Visits: 0.124
    Pages: 0.015

    I personally find the degree of this discrepancy a bit depressing - it means that all the actual numbers are pretty much meaningless.
    {{ DiscussionBoard.errors[6100846].message }}
    • Profile picture of the author dburk
      Originally Posted by carolm View Post

      I personally find the degree of this discrepancy a bit depressing - it means that all the actual numbers are pretty much meaningless.
      Hi carolm,

      There is no need to choose to be depressed. Those two different analytic programs track 2 very different things. They track what they are supposed to track and they should be different if they are tracking properly.

      You just need to learn to appreciate the extra data that you have at your disposal, instead of lamenting that you have more (and different) data than you expected.

      The server stats show you server stats, which is useful to understand server loads and spider traffic. The browser stats give you browser data and tell you what you need to know about the people visiting your website.

      A very small percentage of browsers block Javascript and those that do primarily block third party scripts. If you want more accurate browser stats then install a hosted analytics package (i.e. Piwik) on the same domain (first party script vs. third party), they are less frequently blocked by browsers and will give you slightly more data.

      If you are looking to provide credible third party verification of your traffic to advertisers I highly recommend using Quantcast, it is as easy to install as Google analytics, and allows for trusted third party verification of not just visits but actual visitors and demographics of those visitors.
      Signature

      * Get Results - Outsource Your PPC Management |
      *       Don Burk Advertising & Marketing - www.donburk.com

      {{ DiscussionBoard.errors[6104836].message }}
      • Profile picture of the author carolm
        Originally Posted by dburk View Post

        Hi carolm,

        There is no need to choose to be depressed. Those two different analytic programs track 2 very different things. They track what they are supposed to track and they should be different if they are tracking properly.
        Thanks for the advice, dburk, but no need to worry - I probably meant disappointed, more than depressed - and I recovered within 24 hours once I knew why it happened, anyway.
        {{ DiscussionBoard.errors[6163547].message }}
  • Profile picture of the author Leo Wadsworth
    AWStats uses your webserver's log files. This gives it a picture of absolutely every access to your server.

    Google Analytics uses a Javascript function which is called after the page is rendered.

    If someone leaves before the page is finished rendering, GA doesn't count it, but AW will. Typically this is the biggest reason that AW numbers are higher. The faster your website, the lower this difference will be, but your numbers seem extreme. Are you certain you are looking at pageviews in AW, not hits? Each page represents several hits because a "hit" occurs with each piece of information, such as an image.

    The GA numbers will track closely with AdSense and AdWords impressions. I've also seen them track very well with impressions from DoubleClick ads.

    The thing to track is the change in the stats more than the absolute numbers. You should see corresponding changes in both at the same time.

    In terms of advertising, GA pageviews is the industry standard way of reporting them. As long as you use those, nobody will accuse you of overcounting your impressions.
    {{ DiscussionBoard.errors[6101240].message }}
  • Profile picture of the author Leo Wadsworth
    One more thing - the ratio on AW between visits and "pages" seems way too high. Google is reporting about 3.7 pages per visit, and that sounds pretty normal.
    {{ DiscussionBoard.errors[6101255].message }}
    • Profile picture of the author carolm
      Thanks for the heads-up. Leo - I'll double check my numbers.
      {{ DiscussionBoard.errors[6111013].message }}
    • Profile picture of the author carolm
      Originally Posted by Leo Wadsworth View Post

      One more thing - the ratio on AW between visits and "pages" seems way too high. Google is reporting about 3.7 pages per visit, and that sounds pretty normal.
      Hi Leo, yes they do seem extreme, which is what made me want to ask about them here. Yes, it's Page Views, not Hits.

      Dividing pages by visits and unique visitors, I get

      12.62 for march pages/visits
      26.70 for march pages/unique visitors

      for the other site:

      13.71 for march pages/visits
      30.72 for march pages/unique visitors

      'Extreme' is one way to describe this - I'm thinking incredible, as in, unbelievable!
      {{ DiscussionBoard.errors[6163674].message }}
  • Profile picture of the author Lucid
    As you and many others have found out before, they count different things and why the numbers don't seem to jive. Adwords counts actual clicks on your ads while analytics software counts visits and they're not the same things.
    {{ DiscussionBoard.errors[6103601].message }}
    • Profile picture of the author MikeFriedman
      I use a number of different traffic tracking software on different sites, but the ONLY thing I really use them for is to track trends. Things like what keywords are bringing traffic, how much traffic is increasing with rankings increases, sources of traffic, sites referring traffic, is traffic increasing or decreasing, etc.

      I never put too much stock into the exact numbers though. It is more just to get a general feel of what is going on month over month, year over year.
      Signature

      For SEO news, discussions, tactics, and more.
      {{ DiscussionBoard.errors[6103676].message }}
  • Profile picture of the author maricelu
    hey guys, don't want to make a new thread here for the thing that is closely related to this one.

    In Awstats, GAnalytics, Webalizer, etc. I get search keyphrases like "hummer car" which are far more than keywords the site actually is looking for, like "hummer car dallas" for example. SO, my local site gets searches like this and I tried to google with different locations, etc. but still I am not ranked on the first pages. How do you explain this? Is this something common? Basically, there it shows I am getting like 50 visitors from "hummer car" and only 8 from "hummer car dallas". Is the IP thing here involved?

    Waiting for your thoughs
    Signature

    I have no signature.

    {{ DiscussionBoard.errors[6111584].message }}
    • Profile picture of the author carolm
      Originally Posted by maricelu View Post

      hey guys, don't want to make a new thread here for the thing that is closely related to this one.

      Basically, there it shows I am getting like 50 visitors from "hummer car" and only 8 from "hummer car dallas". Is the IP thing here involved?

      Waiting for your thoughs
      Hi Maricelu, I'm certainly no expert, but I would expect there to be fewer visitors for 'hummer car dallas' than 'hummer car' because when you specify Dallas in the search, you are targeting a smaller market than when you target all the people who are looking for 'hummer car' anywhere in the world.
      {{ DiscussionBoard.errors[6117244].message }}

Trending Topics