Algorithms that measure on-page factors look at the elements of a page that would lead a user to think the page is worth browsing. This includes how keywords are used in content as well as how other words on the page relate. For example, for any given topic, some phrases are common, so if your web site is about beading, an on-page algorithm will determine that by the number of times the term ‘‘beading’’ is used, as well as by the number of related phrases and words that are also used on the page (e.g., wire, patterns, jump rings, string or stringing, etc.).

These word patterns are an indicator that the algorithm results — that beading is the topic of the page — are, in fact, correct. The alternative, no related patterns of words, suggests that keywords were entered randomly on a page, just for their value.

The algorithm will also likely look at the proximity of related words. This is just another element of the pattern that validates the algorithmic results, but these elements also contribute to the quality score of a page.

The on-page algorithm also looks at some elements that human visitors can’t see. The back side of a web page contains special content designed specifically for web crawlers. This content is called meta tags . When a crawler examines your web site, it looks at these tags as definitions for what you intend your site to be about .

It then weighs that against the other elements of on-site optimization, as well as whole-site and off-site optimization, too.

Whole-site algorithms

If on-site algorithms look at the relationship of words and content on a page, then whole-site algorithms look at the relationship of pages on a site. For example, does the home page content relate to the content on other pages? This is an important factor from a user’s viewpoint, because if users come to your site expecting one thing and then click through a link and wind up in completely unrelated territory, they won’t be happy.

To ensure that your web site is what it claims to be, the whole-site algorithm looks at the relationship of site elements, such as the architecture of pages, the use of anchor text , and how the pages on your site are linked together. This is one reason why it’s best to have separate web sites if you have a site that covers multiple, unrelated topics or subjects.

How your site is architected — that is, how usable it is for a site visitor, based on the topic it appears to be about — is a determining factor in how useful web site visitors find your site.

Understand that one of the most important concepts in SEO is how useful site visitors find your web site, and a recurring theme throughout this book is building sites that visitors want to spend time on. Do that and SEO will (usually) fall naturally into place.