SEO and Web Analytics: it takes two!

Do you optimize your web site for search engines? If so, you probably notice that the competition in search engines is intensifying. It takes more and more energy to obtain the desired positioning on key terms. The large number of blogs, as well as social and informational web sites is what is responsible for this increasing competition to be on a search engine’s first page.

It is of utmost importance to give special attention to the selection of these terms. However, several companies still base themselves on information such as the positioning of keywords that come to mind to check if their SEO team is doing its work effectively. Is this really the best way to measure effectiveness?

Probably not. In effect, is it advantageous to be first on a keyword that is searched for only twice a week? Worse, is it really useful to be in the first 10 results for an expression which does not generate any conversion? It’s this last point that holds our attention throughout this article.

Several company managers not selling directly online will be tempted to stop reading at this point, thinking that because they do not sell online they cannot measure conversion. Error. In reality, there is a great number of possible conversions for any given web site (visits on an important page and downloads of a document are two very simple examples).

To come back to conversion, we are always astonished to notice the quantity of companies using powerful analytic tools, but without having properly configured them to check the conversions from the natural results of search engines. A lot of them limit themselves to the analysis of conversions associated to only keyword purchasing or even worse, not analyzing this indicator at all.

How to improve yourself using the measuring instruments?

All self-respecting web analytic tools make it possible for users to know which keyword generates conversions. In the majority of cases, it is even possible to relate it to the product bought and its value (in the case of an electronic commerce’s web site, of course!). Thus, by collecting these data, it is possible to define the words that are worth the effort of being optimized.

It is not rare to find that more than 50% of the keywords are unique. Internet users are searching with more and more precision (« long tail »). This diversity in the expressions used requires a thorough data analysis in order to come to the best conclusions as to which keywords to optimize.

An example is worth a thousand words!

Let’s take the example of a company selling computers. By looking at the data from the statistics of an analytic tool, we learn that 10 000 key-expressions generated sales in the past year. By looking at the column $/visits, we learn that:

• The name of the company (e.g. Future store) generates $/visit of 0, 50$/visit.
• Several unique searches generate $/visit of more than 1000$ (e.g. Toshiba laptop with a dual core processor).
• Certain searches on more or less precise terms generate average sales of around 250$/visit (e.g. “Toshiba laptop” or “19 inch flat screen monitor”).

Thanks to our analytic and keywords mining tools, it is possible for us to highlight important keywords according to their conversion factor and to define the keywords that deserve to be optimized. It is then necessary to proceed to the analysis of the positioning of these keywords in order to determine whether an optimization for them is necessary.