There’s a term in SEO called Latent Semantic Indexing, abbreviated “LSI,” that essentially means that the software engineers are trying to program the algorithms to understand the message a web page conveys based on different variations of language used, as opposed to on repetition of specific keywords.
Hmm, that wasn’t as clear as I hoped.
Here’s an example:
If I was giving a friend advice on choosing the best Denver SEO company, apart from my obvious recommendation that they use NEWMEDIA 😉 , I’d probably say something like this:
“When you’re interviewing an SEO provider, in Denver or anywhere, make sure they understand the rules. White Hat only; focus on backlinks, ideally from Authority sites; make sure they actually implement the SEO recommendations for you, and not just consult; consider rearchitecting your site in SILOs, etc. Make sure that the Denver SEO company you select has a proven track record of high rankings for their clients, and if possible get your results guaranteed.”
Do you see what I did there? I described the situation with a bunch of different terms that, collectively, conveys the message “Denver SEO company.” That’s normal language. That’s what the search engines, especially Google, are trying to do too. Well, in reverse. Google’s not talking to your site — your site is talking to Google, based on the content on it. So, given that we know that Google is trying to understand your site’s message, based on this language principle called LSI, you really should be structuring your content appropriately.
Historically many site owners, and even some bad SEO firms, tried to tell Google what their site was about by doing something like this:
“Denver plumber, denver plumber, denver plumber, denver plumber”
You get the idea.
Now, in really bad cases that can get you in trouble, but even in a best case, techniques like this, called keyword stuffing, won’t help you at all. You need to consider LSI.
Here’s a really interesting post from Matt Cutts on the topic: http://www.mattcutts.com/blog/google-synonyms/
And here’s the article he’s referencing, with some actual examples: http://googleblog.blogspot.com/2010/01/helping-computers-understand-language.html
Really, when you think about it, it’s a lot easier using natural language than it is to try and jam in as many keywords as you possibly can.
Just my $0.02. 🙂