The web is all about content. Finding success in the search industry is all about putting the right content in front of the right user at the right time. How do they do this? Keywords!
There is no denying that all of the popular search engines available today have come a long way from simply counting links and words on pages.
Images, videos, pdfs, flash and other rich media are also being indexed these days along with text-based pages… but the fact remains, users still search using keywords. This is why it’s important not to drop the ball when it comes to keyword strategy and optimising every element on every page.
There’s so much more to optimisation these days than simply making sure that sufficient keywords appear on a page.
Debunking the myth: there is an optimal keyword density
Be very suspicious of SEOs who tell you that the correct keyword density for a key-phrase is between 3% and 8% (or any percentage for that matter). This is simply not true.
Naturally, you cannot hope to rank for a keyword or key-phrase if it does not appear anywhere on your site. However, sheer volume of keywords is not the secret to good rankings.
Many optimisation specialists acknowledge that keyword densities play no role in how commercial search engines process text, index documents or assign weights to keywords.
Indeed, keyword density says nothing about a document’s relevance to a search. It is completely separate from a document’s quality or semantics, and therefore, relevance to a specific topic.
Search engines use complex algorithms with a number of weighted variables to calculate the rankings of sites relative to a specific search term. In fact, in any type of automated information retrieval or text-mining, a weighting system is used.
It’s actually more about weighting than repetition
This weighting is most commonly presented as a tf-idf (term frequency-inverse document frequency) weight. This is a statistical measure which evaluates how important a word is to a document or a corpus of information.
The basic principle is that the importance of any given word increases proportionally to the number of times the word appears in a document; however this is offset by the number of times the word appears in the corpus, or body of documents.
Thus, tf-idf is a fairly efficient way to rank a page’s relevance to a given user query.
Know your corpus
What does this mean for your content? Well, when considering the frequency of a keyword in a document, the best approach from a strategic point of view is to seek to be as natural as possible.
In other words, if you are trying to optimise your page for « PVC paint » – you should consider what the normal frequency of usage is for that word in the pages that rank well for a search for that phrase.
Naturally when it comes to ordering pages in the search engine results pages, the tf-idf weight is not the only variable considered in the algorithm, and aspects such as latent semantic indexing undoubtedly play a pivotal role in determining a document’s relevance to a search term too.
Regrettably, because no SEO has insights into the exact mechanics and specifics of any search engine’s algorithms, the most they can do is to strive to be the least imperfect.
Keyword strategy is still important
This does not negate the importance of having a proper keyword strategy in place – including thorough keyword research and allocation.
Segmentation is a very important part of any keyword strategy. You need to have a clear structure in place for both your human visitors and the search engine bots. Think carefully about your subject matter and build focused, relevant and original content around those areas.
The bottom line is: websites which are clearly structured consistently outrank those which are not – even if they cover similar subject matter.
A few closing points
There is far too much danger of being labelled a spammer if you stuff your content with keywords. Steer clear of black-hat SEO tactics such as hidden text and try to keep things natural.
It’s always best to keep the user in mind when writing content. Natural word usage, solid information architecture and unique content will undoubtedly pay dividends in the end.