SEO Game Theory

While the SEO profession is fairly young, there is generally a common understanding that a site’s ranking for a given query is a function of PageRank, relevance, trust, and indexation. There is much discussion though, of which SEO factors are important, and which are not. Anyone who has researched SEO has run across forum discussions or blog posts that go along the lines of “Optimizing for X is useless! Optimize for Y!”. Certainly some of these are just trying to get some attention and cheap clicks on their site, but the argument bears discussion.

Should ranking factors deemed to be less important be ignored by an SEO?

To make a judgment on this, it’s important to both understand the context, the real meaning of this argument, and what can be inferred about search engine algorithms. Firstly, it’s important to note that when claims like this are made, any real evidence to back it up is quite scarce. When evidence is given, it is usually either unverifiable anecdotes or vague correlations. As well, overtures are rarely made to Bing! or third-tier search engines, and any data is almost exclusively from Google. It bears repeating as well, that just because a ‘celebrity’ SEO expert has a lot of blog traffic, that doesn’t necessarily mean they have any clue what they are talking about.

With that said, it is difficult for someone responsible for client ROI to take these kinds of claims very seriously, but let’s assume for the sake of argument that a claim like this is indeed factual, and that Factor X has been decreased in importance for the calculation of a site’s ranking. Let’s take a look at what we can generally assume about about the Google ranking algorithm.

To explain, the ranking of a page P for a query Q is dependent on the relevance of that query to the page in question, the Pagerank of that Page, and Trust/Authority of that page. Granulating further, Relevance of a page for a Query Q is a function of the headings, titles, content, layout, anchor text, meta tags, URL text, etc for the page P. Google states often that there are “over 200″ such relevancy factors that they weigh.

It should be noted that making an improvement in any of the components of page relevance will increase overall page relevance. In this model, if the importance of any factor is scaled down due to algorithm updates – it still makes sense to optimize all possible factors, as this will have a net positive effect on relevancy and therefore ranking.

Ignoring known or suspected relevancy factors is to sub-optimally optimize a client web page around a key search term.

So my response to people who harp on about certain relevancy factors making not as much difference anymore is twofold:

 
As there are no situations where leaving known relevancy factors unoptimized is the ideal choice, it should simply not be done.