I read a post in SEOBOOK about a cheap was to rank using inline query refinement.  I just want to say that I find it difficult to read the SEOBOOK blog since the template was changed - there’s too much going on and I can’t focus on the content on Arron Wall’s posts and the blog has become too blatently commerical for me.

However, the information about query refinements was very good, but a little too intensive for me to read in full.  When Google finds a query is ambigious it take the the top pages (the first 100 or the first 1000) and calcuates term vectors (how close the main terms one the page compare with an ideal page for that query - one that Google has indentified and stored).  Google also come with suggested terms based on those calculations.  

For example, with the term "Jaguar"  you might get:

jaguar car,

mac os x jaguar,

jaguar racing, and

jaguar cat

Here’s my point: understand the details of how the process works is not really going to help anyone rank better - if any thing, it will inhibit creating anything.

I think SEO work can get carried to the point where some people think begin to reverse engineer what the computer scientists programmed to figure out how to create content that will rank well.   I’ve found that approach leads to paralysis.

When I was at Webmasterworld this week I posted my notes on each session directly to my blog as the session was going on - often posting several times.  In many cases, my posts were being searched for and coming up in the top of search results within two hours of my posting (i even came up as #2 for "google" in Google Blog Search that way - within two hours of my posting).

Now, all I did was write with an idea in mind - and I could create content that way pretty quickly. 

If I worried how the vector on my content terms matched to some ideal page for a query I wanted to rank for …I’d never write anything.