Google reads between the lines to help determine site relevancy

Google reads between the lines to help determine site relevancy

Have you noticed lately that Google is delivering search results that have ‘matching’ keywords in bold that are different to the keywords used as the search criteria? Google’s continued pursuit of what they believe are the best search results for a given search query has yielded their beta attempt at Latent Semantic Indexing. What this basically means is that Google is using meaning and intent much more as a way of determining relevancy.

What does this mean for domaining? Well, it means that writing SEO content without using semantic language is like going to a gun fight with only one bullet. To illustrate with a real world example, lets go to the development whipping post domain case study as a site that suffers from severe Semantic Index Deficiency.

When searching for ‘inline skate’ in, the first position goes to the landing page on Wikipedia for ‘Inline Skates’. The bold keywords in the Wikipedia search result title and the description are ‘inline skates‘ (plural) and ‘inline skating‘, neither of which are exact matches to the query.

Semantic SEO is a technique that will garnish both a greater breadth of organic search results and make the site more relevant for a variety of keywords, each giving strength to the other.

For Google in this case, ‘skate’ could be both a noun and a verbal adjective, also called an active participle, as it is describing the noun (skate) performing the action in the verb (skating). Without any additional data or modifiers, the Google algorithm basically reads the index for all the sites that could warrant a place in the search results and then scans for latent context to determine the most relevant results for the active participle ‘skate’. Aside from their authoritative rank, Wikipedia’s ‘Inline Skate’ page is a strong competitor for a wide variety of active participles and why it gets consideration for a broad search result over other sites with exact matches.

When searching for the semantic ‘skating‘, if the content was optimized properly and all other things being equal, the search results would almost be interchangeable between ‘skate’ and ‘skating’. The sites that appeared in the top five search engine positions for both ‘skate’ and ‘skating’ include Wikipedia (same lander in both), (different landers) and (same landers), so their content, intentional or otherwise, is unique and indexes well for semantic language.

Semantic keyword search results tend to suggest that pleasing Google with exact keyword matches isn’t vital to be considered for first place in the search engines. To write purposeful SEO content for a site means writing enough that you can identify the primary keyword(s) to Google, and then use semantics to shape and coerce Google’s understanding from a number of angles so that the site has the best chance of being discovered for a particular topic, not just for a keyword. It also has to be written in a way that you can do all this and make is readable for the content consumer.

Active participles aren’t the only part of the language that Google takes into consideration. Synonyms are another way to teach the search overlord more of what the page is about, and as you know, are simply other words that basically mean the same thing, such as ‘coaching’ and ‘teaching’, however the index is less likely to respond to synonyms given the heavy use of participles and the greater the relative strength that they offer to the query keyword.

Put that in your pipe, mass development.