AWStats Vs Webalizer Vs Google Analytics Visitor Numbers

Updated on 04-Dec-2011

Web stats programs tend to have different purposes and give different results. I don’t worry too much about the discrepancies between the various programs because comparing say Google Analytics with Webalizer or AWStats doesn’t make sense as it is like comparing a speedometer with a rev counter.

If you like this article, please click on one or more of the social media buttons on the left. Thank you!

On any particular day a website could have 90 visitors reported in Google Analytics, 170 visitors in AWStats and 220 visitors in Webalizer – but there are good reasons for the different figures.

A Real Example – Google Analytics, Webalizer and AWStats Visitor Numbers

For example, on 15th July, these two sites had the following stats.

Example Site 1 – Visitor Stats on 15th July 2011
Stats Package Time Zone Used Unique Visits Recorded
Webalizer EST 56
AWStats EST 32
Google Analytics EST 18
Example Site 2 – Visitor Stats on 15th July 2011
Stats Package Time Zone Used Unique Visits Recorded
Webalizer EST 821
AWStats EST 576
Google Analytics EST 249

The main difference between Google Analytics, AWStats and Webalizer is that Google Analytics gathers visitor information via some JavaScript code you have to place on each of your site’s pages, but the other two interpret the site’s server logs which are generated and held, on your webserver.

This means that the point of data collection is different – for Google Analytics data is collected on the visitor’s browser (and we know how many of those there are with their infinite configuration, version, platform and addon possibilities), and for the others the data is collected on one single place – your webserver.

In fact, all three may even collect daily data in different time zones so ‘days’ begin and end at different times if you host your site in a different time zone to the one you live in. Your Google Account will most likely be in your local time zone, and your server based reports will be based on the server’s time zone.

Think about that for a moment. In the case of Google Analytics data collected from many thousands of your visitors, from a wide variety of locations and in the other, the information is passively collected in one location, directly from your webserver. And further, there is the possibility of time zone discrepancies. It is obvious that the data cannot possibly yield the same answers.

Google Analytics

How Google Analytics Determines Visitor Numbers

  1. Google Analytics relies on the website owner adding special code to each page of their website. This is easy to do when you use a CMS like WordPress to build sites, but not everyone does of course. Consequently, some sites may not have the appropriate Google Analytics code placed on every page. This would cause inaccuracies if visits to some pages were not recorded because the code just isn’t there.
  2. Even if a site does have the Google Analytics code on all its pages, if the site loads slowly, the Google Analytics code may not have a chance to run. This can be the case if the code is placed at the bottom of the page (as we’re advised to do).
  3. Significant numbers of people have fears of the malicious use of JavaScript by some websites. These people routinely block some or all JavaScript from running in their browsers. If you are visited by such a person, their visits will no be tracked by Google Analytics.
  4. Google Analytics relies on cookies to work out a number of things about a visitor. Cookies determine for example, if a visitor is a new one or a returning visitor and also allow Google Analytics to figure out how long a visitor’s visit lasts amongst other things. So, if a user cleans out his cookies, or if he doesn’t allow them to be dropped, Google Analytics will inaccurately record his visit.
  5. Google Analytics does not record search engine bots and other crawlers on your site. This is good because you then don’t get robot visits confused with human visits. It means you need to look at your server logs if you want to understand which robots are visiting your site and where all your bandwidth is going. For this Webalizer or AWstats is perfect (you’ve just found a reason to use both types of statistics program).
  6. Google Analytics starts counting a new visit after a 30 minute delay between visits. So if you visit a site, then go to lunch for 31 minutes, then come back to browsing the same site, that will count as two visits.
  7. You get under-counting of visits in Google Analytics because not everyone allows Google Analytics to run on their browsers and because the old synchronous Google code (which many people are still running), if placed at the bottom of all your pages may not load before your visitor leaves. To fix the code issue, make sure you use the newer asynchronous code for your Analytics.

Google Analytics is a tool for website owners to market their sites online. It helps you understand where an indivudal visitor came from, how they found you, what they did on your site, how engaging they found your content to be, how they left your site and which website goals were acheived. With this information you can make changes to your site to further optimize it for marketing, sales or other conversion.

AWStats

How AWStats Determines Visitor Numbers

  1. AWStats interprets your server log files and as such it defines what it thinks a visitor is, based solely on IP address and user agent. So if something visits several pages on your site with a user agent indicating it is a browser such as Firefox say, and on a single IP address, then that will count as a human visitor to AWStats. But if the user agent is GoogleBot, it will be defined as a robot and not count as a human visitor.
  2. However, some robots don’t identify themselves and although AWStats tries to keep a database of known robots, it can’t know them all. Sometimes robots are counted as human visitors and sometimes, a person who visits a site on a number of different IP addresses in the same session will count as more than one visitor too. And if someone views a page they looked at yesterday or last week (and it’s in their browser cache), AWStats won’t measure their visit at all.
  3. You can use it to see how much bandwidth various robots and crawlers are using, and where they come from. It is useful for finding out who is crawling your site, not how your vsiitors behave (or even what a visitor is) once on your site.
  4. The other thing is that without cookies to help it determine which visitor is which on the same IP, AWstats uses a period of time of 60 minutes to gauge a visit. So if one person visits for 30 minutes, then 35 minutes later, visits again, this counts as one visit. But if they visit now for 55 minutes and then visit again 10 minutes later this counts as two visits.
  5. Mostly, however, you get over-counting of visits on AWStats due to the mis-identification of robot visits as human visits.

AWStats isn’t a marketing tool. But it does allow you to get a good understanding of visitor numbers from both humans and robots, and how much bandwith is being used. You might also be able to detect if someone is hotlinking to your images or document files too, and taking your bandwidth that way. It is a tool for network adminstrators more than marketers. AWStats won’t help you with conversion issues, website goals, nor measure the effectiveness of your SEO or marketing campaigns.

Webalizer

How Webalizer Determines Visitor Numbers

  1. Webalizer works in a similar way to AWStats – in that it interprets server logfiles. In my tests it consistently reported more visitors than either Google Analytics, or AWStats. The main reason for this this that it doesn’t try by default, to differentiate between robot and human visits. So when people have been used to Webalizer, and then get Google Analytics, they are often perplexed by the steep (apparent) drop in traffic.
  2. Webalizer sets a visit at 30 minutes duration, rather than AWStsts’ 60 minutes. So more vists will be recorded due to this fact alone.
  3. You can set Webalizer up to ignore certain spiders or robots, but maybe this isn’t worth doing as you can’t eliminate them all so why bother?

Like AWStats, Webalizer isn’t a marketing tool. It does however provide a simple answer to the question – where has all my bandwidth gone? By default, it lumps all visitors together – human and robot and has only a 30 minute visitor window. It is hardly surprising the numbers it reports are generally higher than those reported by Google Analytics and AWStats.

This article was updated on December 4, 2011

About Elizabeth

Elizabeth is an experienced web developer and WordPress SEO Coach. Connect with her on Google+ and Twitter

Comments

  1. sixstringsensei
    Twitter:
    says:

    I always wondered why the discrepancies between stats software. This is great info Lizzy. Now, instead of just touting webalizer and AWStats as over-inflated and inaccurate ways to read traffic, I can pick the info I need out of them and then rely on GA for the rest of my numbers.

  2. @sixstringsensei And now that Google Analytics has social media engagement tracking we can measure Google+1 activity and even (I believe) activity on other platforms like say your account sixstringsensai on Twitter. Note how I can refer to you in LiveFyre by your Twitter handle. I think it tweets something when I do.

    My latest conversation: http://www.devonwebdesigners.com/mini-course-01-choosing-a-theme/

  3. Thanks for clearing that up for me! I’m pretty new to this whole blogging lark and this is pretty confusing. I am going to use Google Analytics after reading this. Thanks.

  4. @ruizml33 Good – glad to have helped.

  5. Thanks for sharing this article. I really like the way you made this list of how they count visitors. I always use Google Analytics.

  6. This is great information and one of the first times I’ve seen an explanation as to why these different services give different results. I stick with Google Analytics although the others have their uses.
    Nishadha recently posted..Draw Gantt Charts Online Easily Using CreatelyMy Profile

  7. Really great stuff Liz. I always knew AWStats and Webalizer were wildly inaccurate, but I didn’t know the reason why. I don’t like Google Analytics much either though, so I actually use Piwik which works pretty well instead.
    Chris recently posted..Domain SearchMy Profile

    • Yes – avoiding Google Analytics is what some prefer to do, but, I think as time goes on, if you want traffic from Google at all, it’s a case of joining them on all fronts. Analytics, Webmaster Tools, G+ …. If you have good alternatives sources of traffic – as we all should – then you can disengage …

  8. Thanks for the info. I use Analytics and was wondering if I should even both with setting up the other two programs right now since my site is in its infancy. You’ve convinced me to hold off right now and focus on other things. Now I know which I’ll choose when it comes time to worry about where my bandwidth is going. Thanks.

  9. Thanks for the comparison. This is great. I was confused why Google was giving me different, lower, stats than other options and your article definitely cleared it up. I’m using Google for the most part but I guess I’ll continue keep the other two in the backseat in the meantime. :) Thanks!
    Dewey recently posted..[Video] Kindle Fire vs. iPad – Couldn’t have said any better really @ChrisPirilloMy Profile

  10. Gary Ashton says:

    I was also confused about why Google gives a different, lower, stat than other options but its certainly clearer now and I’ll look forward to experiment more with the help of the information provided here.

  11. I try not to overthink analytics in general, Elizabeth.

    Sure it’s great info and should be definitely used to analyze/improve our sites, but way too many site owners go overboard on this one.

    After all, it’s just a guide, no the Bible!

    Thanks for the great rundown on different tools.
    Ana recently posted..Internet Marketing ToolsMy Profile

  12. Thanks Liz, I was wondering what the difference was. Now, the real question is, how to present the disparate numbers to potential advertisers or partners?

    • Well, the only numbers to show advertisers are the real visitor numbers. They won’t be interested in how many robots came looking. If you can also show a relatively low bounce rate and evidence of user engagement (comments, likes etc), then more proof your site has traction. Has to be Google Analytics (or similar) for all that.

  13. I would add that a simple call to an image without even consulting the blog is also considered a visit for AWStats. The script for Google Analytics is not able to see this direct appeal, it also explains the difference.

  14. Your discussion clears up a lot of questions I’ve had for years, mainly why my Webalizer stats are so inflated compared to Google’s. I also use GoDaddy analytics, which are close to Google’s. Do you have any feel for how GoDaddy figures their stats, compared to the others? Thanks.

  15. The best software at the date of this writing for my use is Statpress. It works with WordPress as a plugin. It monitors unique visitors, spiders and robots and can differentiate between the different types of traffic and IP addresses
    Chris Alexander recently posted..How Do I Know If I Am Spamming ?My Profile

    • Glad you are happy with it, but I wouldn’t be interested in any WordPress plugin with that much functionality in it, that’s not been updated since January 2010.

  16. This was a great post, cleared a few things up. But I still have outstanding questions.

    My wife has a popular blog
    Last month Awstats said she recieved 47,366 unique visitors.
    Last month Google said she recieved 6,909 unique visitors.

    For me the difference is outragous. We want to sell advertising on the site and of course it is better to tell them awstats figures but the difference is just too big from google to say it is right and google is two small.

    Advice?

    • AWStats is likely to be counting some robot visitors as human visitors. If your wife has a popular blog I guess she works quite hard on it most days? Is she allowing Awstats to count her own visits?

      Are you running the new(ish) asynchronous Google Analytics code? If you still have the old synchronous code installed it will under-count your visits if your site is slow.

      Does your wife exclude her own visits from AWStats?

      I would think your Google stats are close to the truth.

  17. Just to add to my previous post,

    I have a stat system in the word press also called StatsPressV
    It says unique 7625 last month, whch makes me think google is closer to the mark but still awstats is well off .

  18. Wonderful article. I was wondering if you have considered the impact of Googles changes in privacy that came into effect March 2012 and the impact of people “pausing web history” will have on the accuracy of Gooogle Analytics(GA).

    I use GA and I noticed after March 1st 2012 a couple of strange occurrences
    i: time spent on site nosedived from 2 minutes+ to 34 secs
    ii: the number of new visitors shot up on some days to 100%
    iii: as I have paused my webhistory and clear my browsers cookies daily, GA no longer tracks my behaviour on my site when I visit. I know this because GA has a specific place name for my geographical position and it has the most visits and the longest time and the most reduced bounce rate. Lets call this geographical location x. Prior to pausing my web history geographical location x had over 120 visits, 25% bounce rate and time spent on site 6 minutes +, which is in keeping with my behaviour on my site. I post around 80 posts per month. I must add I have been blogging since Dec 2011 so my visitor numbers are small enough for me to detect outliers. As of March 2012 geographical location x, does not show up on my GA. So my question on the accuracy of GA is,if GA only tracks the percentage of visitors to your site that have
    i: enabled javascript
    ii: have not paused web history and
    iii: presumably do not clear their browsers cookie history daily

    So as of March 2012, not only will GA be under reporting your numbers, their behavioural analysis will also be way off as we do not know the number of visitors to your site that have paused their web history. At the moment GA is under reporting my numbers by 120+, that’s just accounting for me(I am the biggest stat outlier on my site). I have no idea who else visits my site that has paused their web history. To further complicate matters as of May 2012 all sites that track users behaviour via cookies will have to get explicit permission to track them, if your site is based in the EU, apparently BT have trialled this and discovered that 90% of people ask not to be tracked. So we can assume that post May 2012 GA and all other stat estimators will become utterly useless.

    So judging by your article will using webalizer or aw stats produce more optimum results? Because presumably all one needs to do is to discount for robots – say 10% of your figures? More importantly how does the aforementioned points affect bloggers and business models based on ad revenue. Presumably affiliate marketing is affected as the 30 day cookie window will not function all the time, display ad revenue will always be way off if an advertiser is insisting on GA and GA is under reporting your numbers due to people pausing their web history. So in my case if GA is saying 100 visitors (it doesn’t, but for example purposes) it really should be say 200 visitors +, that will affect the rates bloggers can charge for sponsored reviews and display ads.

    I’d really like to have your thoughts on this, as I have only just started blogging (Dec 2012) and I am trying to get my head around the various stat measuring tools.

    Thank you.

    • It is ludicrous to expect site owners to do this when most people who have websites are not capable of finding a solution by themselves. There are hundreds of people posting comments like “is there a WordPress plugin for this?”.

      If they wanted to do this properly the ICO would force the implementation at browser level – it is the only sensible place to do it. Which makes me think the ICO is staffed by incompetants.

      So what do you do if you want to comply – given that actual compliance is costly, requires you to do stupid things to your website, wastes time you can ill-afford, reduces the user experience and hands an advantage to anyone who doesn’t comply?

      There are WordPress plugins you can use. We are looking at developing a plugin that helps with compliance but is also as unobtrusive as possible. I don’t know if we’ll come up with anything better than already exists.

  19. Thanks for the response, if there are no plugins, I’d probably disable both Google Analytics and Quantcast, as at the moment I believe the information they are giving me is wrong. Are my suspicions about GA and pausing web history accurate?
    E Ipila recently posted..Clarins Enchanted Summer Collection 2012My Profile

    • Don’t think pausing your web history affects your GA. I ignore my own visits by using Google Chrome.

      You say you post 80 posts per month. At that rate, I would be more worried that my visits were so low that my own visits skewed the numbers that much. If increasing your visits is what you want to do, I think you are worrying about the wrong things. There are many other things to do to your site if you want to see improvements that do not include stressing over GA’s precise numbers and which analytics tool to use.

  20. Thank you Elizabeth, I have been doubling my visitor numbers since the launch of the blogazine in December 2011; so visitor numbers is not my problem, the nosediving of time spent and extreme increase of new visitors(100% on numerous days) since March 2012 that is bothering me and the disappearance of the stat outlier since March 2012 as I never installed GA’s blocker, but did on statcounter. Anyway these things work themselves out in the end. Thanks for everything have a wonderful weekend.

  21. That makes a lot a sense. I know a lot of people that shut of java and I can see why there would a discrepancy
    Corey Feldman recently posted..Server CrashMy Profile

  22. Hi Elizabeth,

    I was looking for the difference among these three web stats programs and found your article. It is very informative and well-written. I’ve posted it on my facebook fanpage. Thanks for sharing!
    Mary recently posted..Google+ Local Tries to Pull Local Businesses Into Google+My Profile

  23. Hi Elizabeth,

    What a great post! I’ve recently started my affiliate marketing journey and have set up some sites with wordpress. I was considering using AWstats rather than GA but you say it’s not too useful for marketing. I’ve been advised not to use google analytics because of the Panda and Penguin updates. Apparently by giving google your website details, they say it’s more likely your site could get caught in their net, especially now they are doing human reviews.

    I was wondering what your view is on this and also what you think of Piwik as an alternative.

    Thanks again for the info

  24. Have you ever used WUSAGE?

    I havent looked at it in a few years, but I used to think it was pretty good.

  25. Succinct and to the point – thanks Liz. Was searching for a no-fluff explanation without the heaps of technical jargon that normally goes into a writeup on this. Dee
    Dee recently posted..Local Internet Marketing : New Post SeriesMy Profile

  26. Good info. I tell people not to bother with Google Analytics until they get to 2500 per month. Before that, it may not give a lot of useful information. Until that point, you can usually rely on the server analytics like AWstats. I’ve had the two interfere with each other too on some hosts. For example, when GA is installed, I lose the AWstats for some reason. I haven’t been able to figure out why. Anyone have experience with this?
    Ryan Bowman recently posted..Paypal Fees vs Other Merchant AccountsMy Profile

  27. Thorough article, thanks. I have a similar situation to Daniel (above), where AWstats reports almost 1000 visitors a day to one of my websites, yet Google reports about 50. My only guess at this point as that there are a lot of robots getting counted up by AWstats, though that is not confirmed yet. If anyone has any other insights about such large discrepancies in stats numbers, please pass them along.
    Joe Moody recently posted..How to watch your website traffic in realtimeMy Profile

  28. This thread has been around a while and some of the things have also changed over time. One thing that has not changed for me is that I am not a Google Analytics fan.

    I use Piwik. It’s free and it installs quite easy. They have made some recent jumps and are looking really really good now.

    Google of course notices that others are jumping on the train and they have started to not pass the “Search Words” most of the time, which is a bit anoying to say the least.

    Some of the numbers mentioned above are quite incredible. One other reason though why Analytics shows less visitors can also be that someone hast installed a Plug-In, killing the Analytics code as he / she does not want to get tracked. Or you entered the Analytics code at the bottom of the page and the visitor already left the page again, before Google could get him tracked.

    One more reason I use Piwik, it is on my server and I know it has got enough power to track the visitors and doesn’t hang up every now and then.

    As for webalizer and awstats … if I have to choose I take awstats. With a bit of configuring the numbers get closer to reality as webalizer and webalizer is just the pits in my opinion, too old and no where near up to date piece of software.
    Steffen recently posted..perl Module updatenMy Profile

Ask a question or leave a comment

*

CommentLuv badge