Why do different web site tracking programs give wildly different results? Our web host for my main blog, Politics in the Zeros, provides Webalizer and Awstats to track web hits, yet for several years now, Webalizer has shown about twice as many hits as Awstats, which seems baffling, as both reside on the server. Clearly, they are measuring differently. So which is more accurate? And is there any way to get their results closer?
Yes. After some experimentation, I’ve made a few changes that brought the results closer.
First off, I looked at the referring sites section in Awstats. Sometimes you’ll see a site listed that has no pages but hundreds, maybe thousands of hits. These generally are a site that is linking directly to an image on your site, thus stealing your bandwidth without crediting you. If they are major pigs about this (such as a certain radio station in NYC and a right wing site), then I use the block IP utility and they never see the site again. This cuts visits and hits down, but they were garbage hits anyway.
Spam comments were getting absurd, sometimes 1500 a day. I installed stronger anti-spam software plus turned off trackbacks and pingbacks, as they are highly abused by splogs. This also cut down on the number of garbage hits and visits.
Before these tweaks, Webalizer showed 5,000-6,000 visits a day with Awstats at 1,800-2,000. Now it’s 4,100 vs. 2,300, still a wide range, but definitely lessening.
BTW, online tracking software like Google Analytics and the internal WordPress tracking underreport by an enormous factor. Google generally shows Polizeros at 400 visits a day and WordPress at 900. These programs will show better detail on what posts are being linked to, but obviously miss a way too many visits.
Bob Morris 
Bob,
I’ve recently been diving into several websites of a new employer and webalizer is extremely inaccurate and overstates traffic perhaps even tenfold or more…
It’s used by a lot of hosting companies on smaller sites who think they are getting thousands of hits and visitors per month when in reality it’s a few hundred.
Rule of thumb.. Less is more when it comes to Analytics. You always want to be working harder to extend your sites reach and there is no shortcut for hard work! 🙂
Bo-Banna
http://www.fanclubtickets.net
LikeLike
I agree that Webalizer overestimates, but many hosts have both it and AWStats as a part of Control Panel.
The important things. I think, is to look for trends – rising or dropping, and what are the most popular read posts.
LikeLike
I was reading your blog and ran into this myself.
Analytics is based on JavaScript embedded in a web site’s web page… This Javascript is ran by a client browser. AND the client browser reports back to google of the site it is looking at. Don’t get me wrong Analytics is a very nice tool. I use it myself. BUT I don’t leave all my eggs in one basket either…
ALL traffic like spiders/BOTs and spammers, will not show up on Analytics…. They do not run the Javascript that Analytics relies on. And rightfully so. Analytics is meant to measure user traffic like normal peeps like us… Not to report on NON normal web serfing..
Webalizer monitors all traffic. If an image is downloaded or a file downloaded, right down to the little spider crawling the website. This information is very useful for system administrators. Not so useful for content developers..
Also keep in mind you want to make sure the JavaScript for analytics is on ALL pages of your site. The header or footer is a great place to start. Warning, sometimes in WordPress themes the “Header” or “footer” that you have the code in, may not traverse across your whole site. It depends on the theme that you are using and how they wrote it…
So if I may… let’s take a web page. Web pages have s a background image, a banner image, maybe a footer image, and maybe even a sidebar image… In Analytics that would be ONE page hit. However, in Webalizer, that could show up as 4 to 6 hits.. Why the extra hits? If you have advertising on your site like Adsense, you will get a return hit for scanning the page that the ad is on when a user so nicely clicks on those links.
So Analytics see 400 visitors, And let’s say 4 images etc. per page so that’s 1600 hits seen by Webalizer.. Then you have the spiders and BOTs.. I can tell you that can often be double your visitors… And they will index all your pages… And again and again.. Don’t forget about your RSS feeds, trackbacks etc.
You may want to take a look at your sitemap and reduce the amount of spiders on your site if it seems like they are scanning all the time… Personally, I say let them crawl my site… That’s how we get found by REAL people.. But there are spiders out there that are of no use but use up bandwidth.. You can try updating your robot.txt But chances are that the spider is “bad” and will ignore the file. You can then also limit the spider by blocking it’s IP.. Those darn things just keep on moving around and get new addresses…
Or you can just sit back, and as long as your real visitors are not getting affected, let those virtual pests crawl all they want…
Regards,
Jason Brundage
MyITkb.net
LikeLike
Jason,
Thanks for the highly useful ideas and tips.
I generally try look for trends. If several ways of reporting all show increasing traffic, then I assume I’m getting more visitors.
Hits seems a useless number to me because, as you mentioned, it includes images. Visits are what I look at, even if the various analyzer programs define visits differently.
LikeLike
Sorry guys.. The webalizer does not overestimate hits.. a ‘hit’ is defined as a request to the server. Don’t think it’s correct? Count the number of lines in your server log to verify. Other stats programs like AWstats that try to filter out non-human requests are actually giving the inaccurate numbers. As for Google, it relies on javascript being enabled on the visiting browser, which a lot of people turn off for security reasons. Those uses will travel your site completely unseen by Google. And regarding ‘visits’.. there is no way to determine them, so any stats package reporting them are guessing.
LikeLike
I’ve also found Webalizer to give stupidly inaccurate stats. For example its currently saying that my sites has just over 1000 uniques this month. AWStats says its had just over 800.
LikeLike
I DONOT like WebAlizer.
Main reasons:
1) Not very correct stats
2) Refspam through webalizer logs
Refspam is popular in my country, and in case they make it more often the site with WebAlizer may me ddosed.
Thats what i think
LikeLike
webalizer is best with full detailed stats..
LikeLike