Difference between Google Analytics and Web server logs

Hi,

I developed a website and following are the stats of the first 11 days. You will see that there is a big difference between stats of web server and Google Analytics. I want to know:

1. Why so much difference S
2. Secondly is my website's progress good?

Google Analytics
Visits: 838
Page Views: 1432
Visitors: 157
Avg. Page Views: 1.71

Web Server Logs
Visits: 928
Page Views: 11429
Visitors: 3003
Avg. Page Views: 8.07


Thanks

 

At a guess I would say that your web server logs includes bots and google analytics does not.

Plus your web server logs are wrong somewhere along the lines.

A visitor is one unique person, a visitor can make many visits (one to many if your really geeky!)

So therefore visitors should always be less than or equal(extremely unlikely) to visits.

I would trust google analytics.

 

this is something that interests us as well. we often see large discrepancies as to what Google analytics reports and the server logs and / or analytics based directly on server logs reports and as yet are undecided about which is more accurate.

any expert input appreciated.

 

I got large also discrepancies between Google Analytics and my log analyzer Awstats

 

And what do you guys say about my website progress?

 

To be more precise :
Month of April:
Google Analytics
45,373 Visits
237,129 Pageviews

Advanced Web Statistics 6.7
76 590 Visits (unique visitors)
2183 970 Pageviews

I do not think bots can create such difference
plus AWSTAT give me the bot traffic separated

I would love to have some logical explanation on this

 

well i too have seen these kind of differences but not this big

generally the diffrerences are minor like google shows less visitors compare to histats[my other analytic tool] which shows more....

lets c maybe some expert can give us proper answer

 

we have another major customer site showing even greater variation than that as well.

 

if the visitor doesn't have javascript enabled google analytic code won't work, also you should check that you included the goog code in ALL the pages maybe you missed some.

 

no, all pages included and working. on the site I am talking about Google says the uniques are something like 4x as high as their other software.

 

On my site all the menu javascript and other javascript component are doubled with a simple noscript html code so i do not think that is the issue
So javascript is not an issue for me

 

I have included code in every page and I am using the new code.

 

Another big discrepancy
April month
Traffic Origin
Google Analytics:
Search Engines 24,020 (52.94%)
Direct Traffic 13,631 (30.04%)
Referring Sites 7,707 (16.99%)

Advanced Web Statistics 6.7 :
Search Engines 44,448( 6 % )
Direct Traffic 678,243 (92 %)
Referring Sites 13,561 (1.8 % )

Which one i can trust Google or my web logs analyze by aw stats?

 

I used Google Analytics most of the time to judge my traffic, so I trust it more than web base statistic.

 

I understand , i did not mean to say Google Analytics was a bad but i try to understand why it is so much difference between my web log and and my Google Analytics results . I am aware of the inaccuracy of the web log i just try to find something i can trust to be as accurate as possible
made many post on Google Forums but right now i did not find any logical explanation of these discrepancies
What to trust to have a correct image of your traffic?
Any lights on the matter will be welcome

 

suppose you have one index page, that includes 3 other files.

so when this file will open once, google analytics will show it one page, but your server log will show it 4 pages.

that's it.

 

Understand but do that explain the difference on unique visitor ?

 

interesting, we have this going on too

 

SomethingÅ› broken there.
Visits should be higher than visitors.
You should look into that before considering a comparison.

Your menus <noscript> elements have nothing to do with Analytics.
If a visitor is using something like Firefoxes noscript extension, or a proxy strips Googles Analytics JS, the visit simply can not be tracked by Analytics.

You might want to consider setting up custom logging on your server.
At least have an inhouse log parser developed.
That's the only way you can be 100% sure the server has done what the logs say.
Past that there's no magic bullet, you just have to keep up with the tricks of the trade.

 

Regarding visits and visitors it could be that I may have read it incorrectly and it is the other way around i.e. Visits: 3003 and Visitors: 928