Does Google know what you mean?

Google’s ranking algorithm is evolving, constantly reaching new levels of complexity. SEO forums & communities swarm with often contradicting tips & advice, a resultant of research, experience, but also speculation & guesswork. In the attempt to capture a static value in Google’s algo, one may stumble upon LSI, or Latent Semantic Indexing.

In the early days, search engines would focus primarily on the presence and density of keywords on a page. The major fallback of that, exploited by ‘black hat’ SEOs, was the lack of analysis of relationships between keywords. Thus, a query: “walk in bath” could have a page of a Mobility Bathing Solutions website compete against Walking Tours of Bath, Somerset and a blog of a mum from Bath describing the first steps of her baby…

Latent Semantic Indexing allows search algorithms to build semantic maps of keywords, from which groups of potential semantic relatives are formed. Thus, when processing the search query “walk in bath”, an algorithm could rank the page according to the presence/density of such keywords as “disability showers”, “bath aids” or “nationwide mobility”.  An example of a semantically complete paragraph could be:

For a person with limited mobility, bathing may pose a challenge. Luckily, mobility companies offer ranges of solutions such as walk in baths, disability showers and bathing aids to meet all the bathing needs and bring ease and comfort into a disabled person’s daily routine.

In short – it is all about understanding the intent of the user and matching his search with semantically relevant, well-composed and genuine content. In the end, all search engines attempt to emulate a human user with their intrinsic semantic categorisation of the world.

---