Background

Google has been tracking click thrus on certain results since at least 2001 and I am pretty sure before that. You used to be able to see this in the url when clicking such as:

http://
www.google.com/url?sa=U&start=1&q=http://www.example.com/&e=747

This was usually done somewhat a random, but certain keywords – such as “Digital Cameras” were often targeted with a tracker. It was pretty rare for a user to get a tracker at that time – and the official line was that:

“ We use this information to understand and improve the quality of Google’s search technology. For instance, this data helps us determine how often users are satisfied with the first result of a query and how often they proceed to later results.”
Google Privacy Policy

Trackers are often annoying for reasons such as slowing things down, not being able to tell visited links, and not allowing you to right click and get the pure URL. To this day a click on the first result on yahoo for the term "Google" yields:

http://rds.yahoo.com/S=2766679/K=google/v=2/SID=w/TID=F587_122
/l=WS1/R=1/SS=19627813/MI=internal/IPC=us/SHE=0/H=3/;_ylt=Apc
ztF0PCSoG8Cy8Ic.TFphXNyoA/SIG=16o7po6he/EXP=1124876626/*-htt
p%3A//rdrw1.yahoo.com/click?u=http://www.google.com&y=023353
D163304F11&i=482&c=23069&q=02%5ESSHPM%5BL7xpp
xsz6&e=utf-8&r=0&d=wow~F587-en-us&n=8R6K5H31
PS546KQ1&s=21&t=&m=430AEFD2&x=01C02169E198
2CDE

Google has for the most part abandoned url based click tracking for the organic search engine results. However, they were smart - and came up with a way to do it in javascript which is much less transparent and harmful to the user. Last time I checked - not everyone (especially outside the US) was getting this, but my understanding is you most like have this if you are in the US.

General Theory

The General Theory is that if a user clicks on anything below the first result - then Google has failed to provide the best possible result for the user. I believe Google used this metric to gauge how successful various tweaks to the algorithm are working by measuring this behaviour. Now Google plans to do something much more powerful with this data. Look at this SERP that was done during a test of the

revisions inline for the term "prime":

It does not appear that Google is looking at any one source of data to come up with these revisions. However, the real beauty is they can see how well this works - and make it even bettter in the future. For example:

If more people click on the number nine result than click on the number eight result, then Google might no longer show inline results for that keyword - or pick another one to try in the future. Google could even choose to raise the inline revisions from positions 6, 7, and 8, to either a higher or lower position on the page based off of the data they collect. Combining this with personal search will make this even more powerful. I will get into that more with my next post, but will give one example - a search for the term "ir"

While insland revenue is very familiar to people in countries other than the US - in a perfect world - I should never have gotten this. Most US Citizens would have no idea what it even meant. However for someone in New Zealand it might be extrememly useful. Google is on the right track here and hopefully all users will see this in the future.

Posted by Chris on Tuesday, August 23rd, 2005 at 6:34 am.