Forum Moderators: open
Does anyone have any concrete advice as to how to achieve "relevance?" (aside from the obvious: keyword in Title, Alt-tags, body text, etc.)
The only "dumb" question is the one you are afraid to ask.
Brett's post on Successful Site in 12 Months with Google Alone [webmasterworld.com] will tell you a lot about a strategy that many of us have persued with excellent results.
If you want to know how thw designers of Google were thinking (a few years ago) check The Anatomy of a Large-Scale Hypertextual Web Search Engine [www7.scu.edu.au].
Bret's detailed analysis appears to apply more to those SEO tasks related to selling from one's own site rather than the task of optimizing a client's site. Thus, he emphasizes the lengthy process of building content and optimizing for a broad range of keywords.
I am usually involved in the opposite: having a relatively short window in which to demonstrate results, and focusing on one or a small handful of keywords.
What I'm looking for in the forum is more specific advice on the "relevance" factor in respect to Google SERPS. That is, I'm working from the (correct?) assumption that incoming links from "relevant" sites/pages have more weight than links from "irrelevant" sites/pages. I'm not talking PR here, by the way.
That raises some questions: 1) am I right about the relevance factor? 2) How is relevance calculated by Google -- considering each of these elements: keyword, link text, originating page, originating website, theme of those sites that link to the originating website?
I hope this makes a lick of sense.
There are at least a couple of algorithmic approaches that Google might take when (or may be I should say if) they decide to factor in relevance:
1. The "recent" Google patent: [webmasterworld.com...]
2. Topic sensitive pagerank: [webmasterworld.com...]
As far as I can tell, what these approaches do is increase the weight given to relevant links while still giving some weight to others.
"Important, high-quality sites receive a higher PageRank, which Google remembers each time it conducts a search. Of course, important pages mean nothing to you if they don't match your query. So, Google combines PageRank with sophisticated text-matching techniques to find pages that are both important and relevant to your search. Google goes far beyond the number of times a term appears on a page and examines all aspects of the page's content (and the content of the pages linking to it) to determine if it's a good match for your query."
This strikes me as saying explicitly that Google is looking for a "theme", even if that is only location of a given keyword, or synonyms of that keyword, in the title, description, body text, alt-tags, link text, etc. of the pages it indexes.
My question to the group: does anyone have any specific notions as to what Google looks at, how it might prioritize the various elements of the pages, or any insight into this at all?
This would suggest that Google's criteria for "similar pages" is, in priority order, 1) Common inbound link, 2) Common keyword, and 3)(speculative) NO reciprocal link.
I'm leaving that last one as speculative--in could be quite the opposite for other sites. The rationale for NO reciprocal link might be to defeat the interlocking of your own sites, or the usual SEO link swap. I would doubt this based on some very solid results I've been getting with link swapping -- but that could be based on PR, not "relevance."
Now, if this can be interpreted as an algo for "relevance", Google is saying "a link to your page gets rated higher for relevance if its page-of-origin has common inbound links and at least one common keyword somewhere on the page. And maybe a few points for NOT having a reciprocal link?
This is, of course, exactly what Google says in the text I quoted earlier -- have I just rediscovered that the wheel is round?
