"top search query clicks" how does google know this?

In the Query Stats in Google Sitemaps there is now a box named " Top search query clicks ". In Google's words:

But how could they know this? If my site was #1 for the query, then fair enough there is a pre-cache URL in its place which they may be monitoring, but if it was lower down the list (which is the case for some queries) then how could they know the URL was clicked?

 

How do they do this? Yahoo! is obvious about it.

 

Logging clicks? It is simple as web statistics. In these day there are many web statistics scripts doing it.
These results are only based on google. Of course google doesn`t know anything about what people are searching on other search engines.

 

I could understand that if the URL for a result was hxxp://www.google.com/redirect?http://result.com. But it's just the original hxxp://result.com.

Now, looking at a weblog, I can understand that where my trail ends I have chosen a result. But if I search for something, click a result, then click Back and try another result - normal searching behaviour - what is telling Google that I tried 5 or 6 links before I left to go to result.com?

Apologies if I'm just underestimating the amount of data an apache weblog contains .

 

Even though Google seems to only have a simple href to outgoing links, at closer inspection of the source code on the results page there is a "onmousedown" command that will "reflect" this choice to a google log.

You can implement something like this yourself and would be referred to as an outgoing link analyzer. See this link below

http://www.radiation.com/products/reflector/users_guide.html

edited 21.40: further reading here http://www.google.com/support/analytics/bin/answer.py?answer=27246

Regards

Stuey

 

Thanks Stuey, that is what I was looking for.

 

They are tracking all data taken from everywhere. From adsense clicks&views, toolbars,gmail,etc.
Big brother is wathcing us!

 

I was just looking at their HTML code now - very messy + hard to read as they don't use newlines - and I see some JS that might be doing it. But I've never learned javascript and to me it looks like a lot of functions defined that aren't called at all. The links themselves actually point to the direct pages but there is some code that I expect rewrites them to point to Google's linktracker:

<a class=l href="http://www.testprepreview.com/">
notice the class 'l', undefined in Styles

Code (markup):

and here I expect is the rewrite function:

function rwt(el,ct,cd,sg){el.href="/url?sa=t&ct="+escape(ct)+"&cd="+escape(cd)+"&url="+escape(el.href).replace(/\+/g,"%2B")+"&ei=_HYdRJ2RMqDc-AHKhthd"+sg;el.onmousedown="";return true;}

Code (markup):

All this Javascript is way over my head, but it looks to me like links are only counted if JS is turned on on the client's browser. Am I right in saying this?

 

BWDOW... not sure if you are joking or not, but I agree that Big Brother is watching your every step on the web....
I have disabled Google cookies and removed their toolbar just because of that.
We had a Big Brother (Orwell's 1984) but he has grown up and his name is no Big Daddy!!!

 

Why, are you doing something you shouldn't be ? Don't you track your website visitors, use statistics software, check referrals, pageviews, geographic location? We all snoop, Google just collects data on a bigger scale - because it operates on a bigger scale.

 

They usually do it through personalised search, which tracks all the URLs you click in the serps. There's no need to go through Analytics, which just brings a warped picture anyway. Without personalised search Google does not track serps-clicks (perhaps through the toolbar, I don't use that). No magic here, just does what you tell it to do

 

Why do you think they launched Google Analytics?

 

I am sure that they are tracking us. It's a part of their vision