EDSolutions 15.05.2004 Google SEO Test Results

The results had to be reset and therefore we'll be waiting at least two more days for them to be indexed and the results posted. However, we have created a list that you can view to see what new tests are going to be available in a couple days. We have posted a total of 28 tests, and there will be more that we're not revealing until the actual unvealing of the results (think of it as a little surprise.)

Anyhow, if you want to view the current tests that are being performed as we speak, go to:

EDSolutions Google SEO Test Results - 15.05.2004

Feedback will be greatful!

 

How do you know that besides the top result, google orders the rest by rank scores? For example: when using the site:www.yoursite.com command google does not list the pages by rank but (it seems) by how long ago a age was crawled/indexed.
Because you are testing a query that does not return lots of results, it returns mainly pages from your site. Google's algorithm does not like to return a lot of pages from the same site. Google might have determined that the top result is most relevant and list the rest of them in arbitrary order. Something like in the final ranking stages google says: "hmm, there aren't enough results. let's see which pages we have removed from rankings (duplicate and almost duplicate content, pages from the same domain etc.) and add them in the end at no particular order (these removed pages at a previous stage of the algo, might have missed other ranking stages). For example: pages that were detected near duplicates might have missed being ranked at all, and just added in the end because of lack of enought results. Because no one knows internal stuff about the algo, I am just speculating about it.
I am not trying to be a *****, I am just providing my thoughts.

 

This is a good question, and to tell you the truth, we don't know for sure. But I'll tell you one thing:

Google wouldn't go putting any pages in an unsorted order for any reason. Why would they? This would allow Yahoo to scan Microsoft and return better results than Google.

Microsoft has a lot of pages. If I go search for "microsoft.com microsoft word" do you think I'll always be looking just for the first page? I may want the 5th best ranked.

Hexed

 

Well, if you search for "microsoft.com microsoft word" you'll see results from different microsoft domains (microsoft.de, microsoft.ch) and subdomain sites office.microsoft...
Google's idea is clear: return the most relevant page from a domain, not ALL pages from a domain. If you search for microsoft and google didn't filter out pages from the same domain, I bet that all the results should be from microsoft's sites. That is not "quality results".
Google does not want: lot's of results from one domain, duplicate page results, and query-specific duplicate results.

 

I think it's somewhat silly of google that

www.site.com/folder/ would not show as an additional/seperate result but

folder.site.com would

using this logic, would it be always better to use subdomains?