Traffic analysis problem

[marketraise001]

Ok,this one has got to do with traffic.I am currently working on a couple of websites A and B.

For Website A I usually measure my traffic with the means of a statcounter,Google analytics account and the Alexa rank.The problem is that the traffic results from the statcounter and Google Analysis always differs regularly.For example if the statcounter shows that I have had 50 unique visits and 80 raw visits,where as in Google analytics the traffic stats show that for that very particular day I have had only 35 visits.Why is this happening???

For website B it shows that the unique visits are 150-200/daily.The Analytics show that the daily visits are 170.Where as the Alexa rank for this one is way done six hundred thousand.Why does this happen.I consider something is wrong.The reason being via stat counter and Google Analytics I see that I have more Traffic for website B then For website A,where as in Alexa Rankings I see that Website A has a better ranking than website B.

I would preffer a detailed and simple explanation.Something that will help me understand the problem better.

[Linda Buquet]

EEEEKKKK Web Stats.

You know even if you have the 2 most popular stats programs that most hosts include - Webalyzer and Awstat, the numbers will be way off from each other and different than all the ones you mentioned too.

Several reasons:

1) They all handle search engine spiders differently. Some filter them or try to, so the report only shows real human visits while another could be greatly inflated due to spider visits. Since spiders don't buy stuff it can really throw off your conversion rates.

2) There are a variety of cookie issues. If a visitor hits, leaves and comes back in half hour some may count 2 visits, some may only count 1. Then again if one of the stats programs is counting by cookies, some visitors have cookies blocked, so those visits won't show up on that stats program, but they show up on the other.

3) In the past (not sure if it's still the case) Some count by IP. But then you have 1,000 people all on the same AOL IP, so the stats programs that count by IP only count 1 visitor. On the other hand, you have some visitors on an ISP that swaps out IPs every few minute, so the one visitor gets counted multiple times on one stats program, but not the other.

There are a bunch of other reasons too. But essentially there is no standard and they all do it "their" way. Crazy!

You can't look at any of them as totally accurate and it's hard to tell which is closest to the truth. The best thing to do is look at trends. If they all show your traffic is up this month, even though the numbers don't match, then you can feel confident you are on an upward trend.

[Larwee]

marketraise001, you also mentioned Alexa.

Alexa is far from being accurate or reliable for so many reasons. First it obtains information from people who have the Alexa Toolbar installed. People without the toolbar who visit your site aren't considered by Alexa.

Your own visits to your sites play a role also. If you happen to visit one of them yourself more than the other, Alexa could give it a higher ranking.

A lot of people are under the false impression that Alexa is accurate. As a result there is a lot of abuse to manipulate the Alexa rankings in order to make a site look more popular than it actually is and it is very easy to manipulate the Alexa rankings. So, Alexa rankings aren't the best way for people to try to accurately measure their traffic.

[Linda Buquet]

Good point Larry. Above I was talking about various web stats programs.
Alexa is a different animal and can be manipulated.

It also only measures traffic that comes from visitors that have the toolbar.
So if there's a big article about Alexa being spyware, a bunch of people uninstall so it looks like your traffic went down. OR a SEO blog may look like they have lots of traffic because lots of webmasters use the toolbar, but a fashion blog that has lots of traffic may not look like very busy because maybe not as many of that type of visitor has the toolbar installed.