SEO (Search Engine Optimisation) is an important area for every web analyst as there are many things which can be done to optimise websites without having to have extensive knowledge of SEO. WDF*IDF is a new method of SEO and we will be taking a closer look at it throughout this article.

To start, all you need is a keyword and the corresponding best-ranking sites. This collection of sites gives an overview of how often the keyword has been used on those sites and which other keywords are also important in the same context. With this knowledge web managers can easily optimise the frequency of keywords on their own sites because successful keywords and high frequencies can be copied and it is not necessary to know the exact ranking criteria used by search engines.

The abbreviation WDF means “Within Document Frequency”, and refers to the frequency of a keyword within a single document. IDF is the “Inverse Document Frequency” and gives the frequency of a keyword in all of the documents that are available. If you manage a site for a very specific topic that only 11 specialists are interested in, you will clearly have a better chance of being found in the top 10 than someone who manages a “pizza delivery” site. What does this mean? Basically, if you have unique keywords and/or the right frequency, then the better the chances your site has of being clicked.

There are several tools for optimising websites using WDF*IDF. But as a digital analyst you are way ahead of other website managers, because you have access to additional information. You can, for example, say if there are strong off-page influences for your site’s ranking.

Take Google Translate as an example. Search engines list it in the top 10 if you look for the word “click”. The word click is a keyword which has got nothing to do with translations, but there are millions of links to be translated using the word “click”. If you want to optimise a page that should appear as a result’s page when someone makes a search for “clicks” and which provides information on the different types of clicks, then the results provided by WDF*IDF should not be trusted as they are ‘contaminated’ with off-page influences.

Additionally, WDF*IDF can’t provide new keywords or information, because you’ll only get what others have already included in their sites, and you won’t see what Internet users are really looking for. As a result you may be optimising your site for a keyword that your audience is no longer interested in.

Here your experience as a web analyst is required. Keywords from internal searches provide clues, for example, on how user interests or wordings change over time, or depending on location. “Mobile phone”, “cell phone” and “mobile” refer to the same object but the expression used will have a huge influence on your site’s search rank.

Other keywords obtained through search engine traffic are also useful, but searches in safe mode on Google make it harder to get clear results and there won’t be many visitors to your site who have been searching for more up-to-date keywords, even if you have the right product for them.

Our blog article on How to improve SEO despite an increase in the number of “not-provided” keywords. Web analytics methods and useful tips provides more information on Google’s not provided keywords and how to deal with them.

You can also use your AT Internet tools to keep an eye on the latest trends and buzz keywords. With Buzzwatcher you’ll know what experts and opinion leaders are talking about and how they talk about it. This gives you the chance to integrate new keywords into your sites before the rest of the market does. But your main goal is for you to be the opinion leader instead of just following the trend. By keeping your product or brand present in your customer’s mind, there will be less chances of them searching for something else.




Sebastian Keller was born in 1980 and started his career at McCann Erickson and other advertising & text agencies in Germany. He now is a freelance copywriter, translator and SEO texter. He works for AT Internet’s office in Munich since 2011.