| 1:11 am on Jul 10, 2006 (gmt 0)|
Google does know when people make clicks - not sure the exact mechanism, but in Google Sitemaps they can tell you which terms most often result in clicks to your site so they're collecting the data somehow.
| 2:02 am on Jul 10, 2006 (gmt 0)|
People have been reporting Google click-counting urls here off and for a pretty long time. They seem to run for a very limited period (often only hours) for any particular type of search, then they get removed. I assume they know when they've collected statistically significant amounts of data and then the SERP reverts to a simpler style of result url.
| 4:42 am on Jul 10, 2006 (gmt 0)|
Inktomi search experimented with this hit count,
I think 1998 or 1999.
They had also a search engine, where listings had been modified based on the hit counts.
I think they gave up the method, as it became known and manipulated.
| 4:49 am on Jul 10, 2006 (gmt 0)|
Tedster wrote : "They seem to run for a very limited period (often only hours) for any particular type of search, then they get removed"
Clicks have always been counted (even when the click-counting urls are off).
Look at the source of any Google Search, and you will see the following code :
The clk() JS function is nothing else than the click-counter :-)
[edited by: Frederic1 at 4:54 am (utc) on July 10, 2006]
| 8:44 am on Jul 10, 2006 (gmt 0)|
Have to say i dont like this sort of thing as i dont trust google one bit.
Lets assume your site gets 10,000 clicks a week for the search term blue widgets to your blue widgets page. You rank in the top three of the serps and you start being recognised in the market as a site about blue widgets and you start selling blue widgets in bulk. Google knows this as its tracking clicks.
Next thing google puts a 20 placement condition on your site for this key word which makes your serps position 22 rather than 2, you now have next to no traffic for this term - You now need to buy adwords to get the same traffic back.
Now call me a cynic but this is the sort of way i would see google using this kind of data, after all its all about making more and more and more money - they are not doing this for surfers or webmasters advantage are they?
Anyway thats my two pennys, i dont like it one bit - the only answer for us webmasters is for google to lose some market share so that they no longer dominate so that search traffic is spread more accross the internet, but currently we have no one in the market even within a mile of them hence why they will continue working on issues to no doubt increase adwords revenue rather than search quailty.
| 9:56 am on Jul 10, 2006 (gmt 0)|
|... after all its all about making more and more and more money |
What are you in it for?
|they are not doing this for surfers or webmasters advantage are they? |
The important thing is for Google to ensure the surfer doesn't suffer. As long as that doesn't happen, fair play to them.
| 11:49 am on Jul 10, 2006 (gmt 0)|
Thinking about it further, GoogleToolbar is obviously a way Google can count hits - it gives them a sampling of data which is sufficient to draw conclusions on many search terms.
| 2:42 pm on Jul 10, 2006 (gmt 0)|
|And will this be part of the ranking algo? |
A touch of click through rate thrown into the secret sauce would make sense;
If a lot of people visit your site thereís got to be something of value there.
In a way it's a little like adwords, take your click through rate and add it to your bid and thatís where you place. If no one clicks on a high ranking site, why is it ranking high?
More to the topic though, Google is a data mining enterprise at the end of the day, and you have to assume they are collecting this sort of information given how close it is to their core product.
| 3:43 pm on Jul 10, 2006 (gmt 0)|
|Look at the source of any Google Search, and you will see the following code : |
Not here, Frederic. The only time I see any onclick scripting in the Google SERP source code is when I am signed in to my Google Account in some way. When I'm signed out, there is none.
| 4:00 pm on Jul 10, 2006 (gmt 0)|
I see the onmousedown clk() code even when not signed in to any Google account. I do have the Toolbar installed though.
It's been around for a long time [webmasterworld.com...]
[edited by: Iguana at 4:01 pm (utc) on July 10, 2006]
| 4:42 pm on Jul 10, 2006 (gmt 0)|
I have this suspicion for quite a long time now. I think google is particularly tracking user behaviour via the toolbar much deeper than by the means mentioned here before. This is the reason why from time to time a link to download the toolbar appears on the main page.
A friend of mine, who has a highly informative page on widgets has been particularly struck by 27th june. He is a thorrough expert in his niche, and has put particular attention to improve his content and exchange only highly theme-relevant backlinks for quite some time now. In the past two years he continuously moved from selling widgets directly towards sending his visitors to a similar affiliate site. And this affiliate site is also running adwords.
Bad perspectives for sites generating revenue by any sort of affiliate program whilst attracting visitors from organic serps alone.
And if I put myself into the position of the searcher, I must admit I cannot ban this policy at all, actually: Why should I perform two clicks, if google manages to lead me to the desired page with one?
BTW: My friend's affiliate partner does not appear higher in the organic serps, but quite often on spot #1 among the adwords on the right. All this is only logical.
| 8:53 pm on Jul 10, 2006 (gmt 0)|
I uninstalled the toolbar, closed all browser windows, and I still get the onmousedown=clk() function. I'm not logged in to Google in any way. So why do I see this function and Tedster doesn't?
| 1:46 am on Jul 11, 2006 (gmt 0)|
| 8:56 am on Jul 11, 2006 (gmt 0)|
I don't know about Google counting hits, but I know that as soon as Google drops my site in their updates etc ... MSN & Yahoo elevate my site to page 1 for most of my keywords, resulting in maybe a 50% compensation for any Google loss. With all of the webmaster interest that is generated by Google page 1 listings, I'm convinced that I lose very little from these periods of Google Gloom.
My sites these days are ad free and for personal interests etc, the situation I describe may be different for those with affiliate schemes etc, as Yahoo never liked me having those.
All the Best
| 9:28 am on Jul 11, 2006 (gmt 0)|
So, it looks like Google has around three years of click data from Internet Explorer users. Useful data. It means they know how people refine search strings when they don't find what they want (as shown by those middle section results). Possibly they are used in some way for ranking as well.