How easily could Google match me with similar searchers?

[pavlovapete]

Personalization aside, I assume Google has several years of my query & clickstream data sitting on a server/s somewhere.

How easy would it be for them to match me with people who have simialar clickstreams all around the planet.

Easy I guess. What say you?

[tedster]

In a general "demographic" way I'm sure they do this and are working to fine tune it. This is ultimately what personalized ad delivery is all about, after all. And have you heard about this Adwords enhancement last summer - session based broad match [adwords.google.com]? Yes, they are already collecting and using some outrageous search session data!

So if the Google SERPs ever start to show DoubleClick ads, then we're all in for it.

[TheMadScientist]

Yeah, it would be relatively easy to do with the computing power and time to write the code to do it... IMO it's much easier than pulling apart the English language and trying to organize the pages you have access to based on relevance to a 3 word string. If you want to know how good the people writing the algos (heuristics) for G are at what they do sit down and try to think about matching the most relevant 10 pages out of 1,000,000 to a 3 word string. You've got a 1 in 100,000 chance of getting one of the 10 most relevant results in the top ten randomly. They do so much better it's not even funny.

<MiniRant>
I know neither of you are complaining, so this is a bit more of a random mini rant, because sometimes it seems so silly to me for some many people to complain about a bit of spam anywhere in the results when the numbers they are working with to return the results at all are incredible. You have a 1 in NUM of results chance of getting the correct result at the top let alone getting the top ten in anywhere near the correct order and people complain because they aren't what a site owner thinks is perfect? As if what they do is easy, and doing it as fast as they do is remarkable... Want to know how good the heuristics at Google are? Go check the odds of winning the lottery (those are similar to the odds of getting the correct result at the top for many terms) and then see how many times out of 100 Google's heuristic is even close...
</MiniRant>

No offense to either of you... This is the thread I was posting in and I felt a little ranty and there was a nice lead in. LOL.

I guess the short answer to the original question is:
Relatively easy... By relatively easy I mean you're matching similar short strings to each other, not a short string to N of 1,000,000,000, 20 to 10,000 word pages.