I've been using Google Analytics on several of my websites, it's a wonderful tool that provides you with invaluable information about user behavior on your website.
The aim of this post is to show the big difference in measuring unique visitors, pageviews etc between this two tools.
Militz.nu is a domain availability tool, it lists all quarantine domains under the top level domain .se
This is a chart of statistic of this domain: Militz.nu for the period: 2/11/2007 - 2/17/2007
| Messures | Google Analytics | Webtrends |
|---|---|---|
| Unique Visitors | 521 | 2132 |
| Pageviews | 2638 | 16287 |
| Average Visit Length (sec) | 0-10 | 23:58 |
| Returning Visitors | 84,9% | 67% |
Top 5 referring sites according to Webtrends
| Sites | Visits |
|---|---|
| No Referrer | 1,812 |
| http://www.militz.nu/ | 932 |
| http://www.google.com/ | 182 |
| http://www.google.se/ | 130 |
| http://search.msn.se/ | 19 |
Top 5 referring sites according to Google analytics
| Sites | Visits |
|---|---|
| google[organic] | 415 |
| (direct)[(none)] | 109 |
| msn (search.msn.com) | 20 |
| eniro.se | 7 |
| yahoo[organic] | 6 |
Top 5 search keyword
| Google Analytics | Webtrends |
|---|---|
| laserab.se | www.happybirthday.se |
| webgenerator.se | fanerdun |
| www.horoskop.se | www.salomon.se |
| www.salomon.se | www.bygganytt.se |
| www.facebook.se | seatclubsweden |
The main problem with this big difference is that when u need to publices your weblogs to compare your website with others there are no real numbers to rely on. Personally i prefer google analytics as a tool to track user behavior, but webtrends as a tool for facts and figures.
2 kommentarer:
there is also a difference with the two programs in that WebTrands is looking at your weblog files for all files on the web server and Analytics takes javascript calls from each page.
So on WebTrends you will get the errors you will not get on google from missing files but also it will record spiders and robots as hits when google will not as they do not trigger the JS.
what you should look for is that the two stats programs report the same ratio of visits to page views and then you know the systems are set up the same way just recording the hits differently.
Interesting comparison...
Webtrends though can do log file analysis as well as page tagging. We use Webtrends only in the latter sense. I would be very interested in finding out how exactly the comparison was done?
Could you shed some light on that please? Thanks, Kees
Skicka en kommentar
<< Hem