Get Real About Your Traffic
You would be surprised with the amount of bloggers and webmasters out there who believe that their sites get way more traffic than what they really do. And I am not pointing fingers here, as I have been there myself.
The reason for this is probably connected with the popularity of cPanel hosting, as the two web analytics programs that come with cPanel, AWStats and Webalizer, grossly overestimate traffic levels. Beginners often use one of these to get the stats for their sites, and they end up believing in a fairy tale of visitors and page views.
Around two years ago I was doing the same mistake. By that time this blog was probably getting some 100,000 page views monthly, but Webalizer was reporting 350,000 page views or so…. Needless to say that I wanted to believe in the latter number, but after a while a realized that it couldn’t be right.
Take a look at the traffic levels that Webalizer reported for this blog in August:
If you sum that up you will get 1,081,959 page views. I wish that was true! The real number is closer to a third of that.
So why is it important to get real about traffic levels? For several reasons. First of all it is the only way to know the popularity of your site and to compare it with the competition. Secondly, it is essential to know the real numbers when you are trying to monetize the site. Imagine selling banner ads on your site and claiming that you get 300,000 monthly page views when in reality you get 50,000…. Advertisers would get pissed and you would also hurt your credibility.
At this point you might be asking: “OK, you convinced me, but how do I get the real traffic numbers for my site?”
In my opinion that are just two ways to get the real traffic levels on your website. The first one is by using Google Analytics, which is the analytics software that most accurately interprets traffic data (it is free also). The second way would be to run third party ads on your site, like Google AdSense or ads from a CPM network. When there is money involved you can bet that people will track the numbers as accurately as possible.
40 Responses to “Get Real About Your Traffic”
I agree with you, AWStats shows more visitors than Google Analytics. I prefer GA because of great UI and data that it shows.
I am still confused a little because different statistics between different applications! but I use Google Analytics as it’s the most trusted program for me!
I have tried AWStats, Webalizer, Google Analytics, StatCounter, CPC and CPM charts and I never get the same numbers for my traffic ……… this is really confusing !
AWStats and Webalizer work by analysing the raw Apache server files. While there is a slight overestimation due to bots, this is by far the most accurate way to determine the real traffic to your site.
Blake @ Props Blog
I just signed up for Google Analytics. I’ve been using AWstats while I got started since it was quick and easy to setup. Google Analytics was actually very quick and easy to setup. I’m really excited to see how the data compares. I believe accurate stat tracking tool can really help page optimization. I also really like Google Webmaster Tools where you can see the top keywords on your site and where your site ranks in a variety of search terms.
I’m looking forward to seeing my real traffic stats.
I believe Google Analylitics is best tool. You can not only track your traffic but you can track your downloads, click and lot many things.
Google Analytics rocks!
I totally agree with you. Personally I trust google analytics more than awstats (which counts spambots as visits), webalizer etc..
@David, I am not sure if the JS is that slow. Even if it was, a user who does not stay long enough to load the Analytics GS wouldn’t have time to read or content, load your ads or anything else, so probably not a big deal if it was not counted.
Try putting Google AdSense on your site and comparing the impressions that AdSense will get with Webalizer, the latter will be 3 or 4 times as many.
@Daniel The problem is the GA files are requested from a server that is on a different network and different data centre to your own site. I have seen many situations where GA code is still loading long after the actual page finished rendering.
Let me put it this way. GA does not actually measure the visits to your page, but measures the requests to its code.
On the other hand, I agree AWStats and Webalizer are rather crude and count traffic that may be of no value. But I think 300% overestimation is far too much. The truth is probably somewhere in the middle between your server logs and GA.
If you are after a more accurate system, I would recommend Mint or Clicky. Mint is a self hosted app, so you are guaranteed accurate results. Clicky is more like GA but they have an option of hosting the JS on your own server, thus avoiding the problem GA has.
Anyway, just my 2c 🙂
Anyone know whether Sitemeter or Statcounter are any good; or whether I should just swap to GA.
I’m using WordPress Stats plugin which seems to be relatively conservative, and very close to what Google Analytics reports. I like having the stats reported in the WordPress admin interface as well, much more convenient for me at the moment when I’m tracking numbers of hits for various posts, and search terms directed into the blogs.
If anyone has any data on how accurate the WordPress system is, I would love to read about it.
I always wondered why my cPanel traffic stats were always way more that Google Analytics told me. I kind of guessed that the cPanel stats had to be wrong, but like you I wanted to believe they were correct.
I’ll just ignore them completely in future, and stick to checking out my Google Analytics stats.
Thanks for the info.
I use a mixture of Google Analytics, Awstats and a wordpress plugin called Cystats to measure my traffic. So far so good they’ve been pretty great in interpreting my traffic.
How do you think for wordpress stats and Google analytics comparison ?
Very true article; it could be very depressing to you to find out that you really have one-tenth the pageviews you thought you had.
thanks for let us know this issue, Google Analytics is probably the most reliable tool that we can trust right now.
I’m using StatCounter for some time but I’ve switched to Google Analylitics. Tracking so many things like downloads among any others. I believe this is the best tool around so far.
HTML Basic Tutor
I agree with:
-depending on how well the Google Analytics server is running, it can slow down the rendering of your page and may cause people to leave. Anyone try to surf the day Google’s servers did go down? It was awful. You sure could tell which sites use Google applications and services.
If you look at the Unique visitors and Number of visits columns and not worry about the hits column then you get some idea of how much your unique and valuable traffic your site/blog is getting.
Scrolling down the screen you next want to look at Vists duration. If people aren’t staying and reading your pages, then you have a problem that needs to be fixed. i.e. Content your targetted audience wants to read/see.
Everyone seems to be obsessed with traffic but if that traffic isn’t staying on the site and/or making purchases (if applicable) then all the traffic in the world won’t make your site successful.
Very well said. If you look at most bloggers, they don’t usually show the Google Analytic screenshots on their sites and instead show the cPanel traffic trackers. I actually dont have any idea why the traffic from these trackers are very inflated. Any idea why?
I notice a HUGE difference between the traffic tools 1&1 uses and Google Analytics.
It’s all pretty simple really, and both Google Analytics and server software like Webalizer are equally accurate though reflect differently.
Google Analytics counts people looking at your website using a browser.
Server hosted analytics, like Webalizer, track everything. This includes robots like Google. To a host, they don’t care if a person visits or if a bot scans your entire site. The host needs to transmit the data regardless so they track everything.
As usual, the truth is somewhere in between. Experience shows that Google Analytics (and other JS solutions) lose number of page views from legitimate visitors (not to mention other types of files like images, files, scripts, bandwidth, etc), while some log file analyzers show really ridiculous data.
With our Web Log Storming log analyzer (not free!) we are trying to fill this gap. It’s easy to exclude most robots while you still have info about every single hit to a server, regardless if script loading was interrupted, if JS is disabled, JS error occurs, etc.
What’s more important, you can actually check results by drilling down into details and see individual visitor’s (or robot’s) behavior and use live (on the fly) filters for advanced analysis.
YMMV, but if you are not satisfied with your current analytics solution, you might want to download free trial and see if this one suits your needs.
If this comment is inappropriate, please accept my apologies and feel free to remove it.
“thanks for let us know this issue, Google Analytics is probably the most reliable tool that we can trust right now.”
In short, if Mint records 200 visitors, Google Analytics will record 170. If Mint shows 600 pageviews, Google Analytics shows 500. The reason everyone is all over Google Analytics is the interface. They like shiny, and apparently will take a massive reliability hit in order to have it.
@LetUpdate and Dave Doolin, I believe WordPress stats are quite good actually, as they tend to be similar to what GA reports.
@Melvin, I believe they inflate numbers because they mix real traffic with search and other types of automated bots.
Overall, though, I could agree that Mint is slightly more accurate than GA due to JS issue.
Yeah, you’re right.
I never believe in web stats which is provided by cpanel web hosting or the others.
They all suck =.=
I just may believe in Google Analytics which is very detail and efficient.
This is something that has boggled me for a while – I use both WPstats and GA. GA reports about twice the number that WPstats does. It always has been about twice the amount, even as my traffic has grown (it’s a difference of about 50,000 page views monthly right now). I do have 3rd party CPM advertizing on my site and they pay very close to what the WPstats numbers are. I wish they paid in line with what GA says.
I always hear that GA is so accurate, but I have to wonder since my ads aren’t getting those high numbers. I only have 1 ad, so that makes me wonder too.
@Tiffany, I have seen a couple of ad networks who also report below GA numbers. Not sure what causes the discrepancy.
This is becoming seriously interesting. Goes to show that there is probably room for yet another stats application!
WordPress stats seems to overreport with respect to GA, for me, by about 30%. I’m cool with that. I’m mostly interested in orders of magnitude rather than factors of. Once I reach @tiffany’s numbers I’ll probably change my tune.
I was once in that same boat, believing the AWStats reports over GA :p
I’m so glad that I grew a head on my shoulders and faced the sad truth of my web site traffic.
Blake @ Props Blog
I’ve had GA up for about 5 days now. IT does seem to give me slightly lower numbers than WPstats, but not nearly as bad as I was expecting after reading this article.
GA is amazing for tracking traffic though. I really am enjoying being able to see all the details of where people link in from, which pages they go to, and all the other amazing tracking features GA offers. I don’t know if the other services offer the same details, but GA has me sold.
I use Google Analytics myself to review traffic levels. My traffic is no where near 100,000, but I am slowly building traffic each day. I just started my blog a couple of months ago. I like to see Loyal visitors and its really good to see new visitors each day. There is quite a bit of information on Google Analytics and I have to be honest I am still trying to understand it all. I still think its amazing that I see visitors from around the world.
Great Post Thanks!
I need to switch to Google Analytics. I have heard it’s better (and more accurate) than StatCounter, Site Meter, etc. You should do a “comparison” sometime. I’d love to read your take on the pros and cons of the various options available.
I never trusted or use Awstat. Since the beginning of my blogging adventure I have been using Statcounter and Google analytic. They are the best.
I see that Statcounter often shows a bit more than GA however. Well this is maybe because Statcounter count a unique visitor on 30 minutes interval and I think GA is longer.
My actual count with Statcounter is about 80000 per month while with GA it shows about 65000 to 70000 per month.
I left a comment here a few days ago, sharing that GA was giving me twice the #’s that WPstats does. After asking around, someone suggested that maybe the GA code was installed twice. I couldn’t find that myself, it was only installed in the header. But my blog designer dug around for a while and found that it really was installed twice in the same place. She fixed it and now GA is running very close to WPstats – a bit higher, but nothing out of line.
Just thought I’d leave an update, in case anyone else is dealing with this.
I have been using GA since I set up my site and have been aware of the under-reporting – it can’t even agree my adsense numbers! What annoys me about GA is that it is really difficult to exclude my own traffic. My traffic is low enough that I can seriously affect the stats if I spend a lot of time on my site. There are some work-arounds available on the web. None of them are foolproof. The biggest irony is that the fix advocated on the GA help site doesn’t really work on blogger (another google product!).
I’m considering trying out the Yahoo Analytics product to see the difference.
A large number of our customers have the same experience as many of the people above: They would put Google Adsense ads on their websites, and the number of views according to Google is only about 25% of the figure reported by the web server.
Since Google has a vested interest in this, I would certainly not switch to GA, simply to force the stats to match. The server stats does include visitors by search engine bots, but that is easy to identify if you look at the host names, and can NOT account for this huge discrepancy.
Comments are closed.