When Google lead search designer, John Wiley, admitted in a Bloomberg interview that 15% of daily Google searches are about search terms the search engine has not seen before he quantified the problem with search for everyone.

semantic web searchAt its core function Google search is a match maker: it is designed to take a search query and match it with the content, on the web, that best answers it. But that is an imprecise, at best, task filled with the second-guessing that goes on when those searching are trying to imagine what keywords will best bring up the content they’re looking for and those who want to be found, are trying to imagine what keywords prospects are likely to use.

The 15% daily gap in Google search (which currently works out to a staggering 500 million search queries a day) shows just how hit-and-miss this is. Consider, as an example, the simplicity of selling sneakers. In order to be found, on the web, you need to take into account: sneakers, trainers, gym shoes, running shoes, sports shoes, athletic shoes and runners and that’s even before we get into the different types of running like sprinting, cross-country and long distance, each of which adds its own complexity.

How a specialist seller might describe any of this is probably completely different to how a dedicated enthusiast would, and both would use different terms to what you and I might.

None of this is new, but what is new is the change that’s coming. In a semantic web, semantic search does not just blindly bring searchers and web pages together and hope for the best. In the tradition of the most successful matchmakers it goes into the trouble of understanding what a web page is about (the way you and I would) and understanding the intent behind a search query and explicitly matching the former with the latter in a high-confidence result.

The reason we’re discussing all this, here, now lies in those two simple words: “high-confidence”. Semantic search is at its best when it knows user intent and can divine a website’s content. In order to do the former it usually needs access to a person’s personal profile, preferences, past search history, search patterns and social connections and in order to do the latter it needs to map all the relational connections of a website on the web.

As marketers the first requirement is outside your control but the second one isn’t and it can be reduced into a manageable set of marketing actions. In order for a website to begin to be understood outside the narrow band of keywords it uses in its content it needs to:

  • Have website copy that truly explains what the website does in terms of what issues it solves or informational needs it fulfils.
  • Establish a strong footprint in social media networks like Twitter, Google+ and Facebook.
  • Succeed in creating a buzz and a conversation around the website, the brand and the products or services it has to offer.

In a semantic web a social signal, clear, compelling content and lots of website visitor interaction by way of comments, mentions and resharing, create a matrix of connections that is now mined by semantic search. Marketers who understand this connection are best geared to take advantage of the 15% gap Google experiences and have their websites come up in searches that are related to their subject but not directly related to their keywords. And that is a truly radical departure from the traditional search marketing of the past.

For more detailed help on marketing in the semantic web check out Google Semantic Search.

(Semantic web / shutterstock)