Birth of A Web Site – II

On from Analytics Culture, Web Analytics

This is the second instalment of this series about the analysis of since the day of its launched. If you missed the first one, well, look down on this page! or find it here.

For this post, I will examine in more details what’s going on with the organic search engine traffic, keyphrases, etc, since search engines, namely Google, are responsible for 85% + of total traffic.

In the period we examine, those visits came via 10,121 different keyphrases. Only 391 expressions have generated 5 visits and more, which is 3,9% of total. And what percentage of keyphrases brought 10 visits of more? A mere 1,4%. So, you got it, we are talking about a long tail. In this case 81,3% of all expressions generated 1 visit, a single one, in 3 months of activity!!

So, if we have a look, nothing surprises us with the line shape:

The 5 red crosses represent the top expressions where we find “cuisine du quebec” (see box). Note that we cannot speak of “brand” searches here since they are very generic expressions, except for the, which obviously is the site. This means that the very top keyphrases are generic topic terms too, by that I mean that those organic search visits were not from a navigational use of Google, which I have commented a lot here and here. This is not surprising, because the site is not known in its market, and that’s why organic search here is true search, and brings so many new visitors. It is important to differentiate that phenomenon from what Avinash Kaushik presents in his analysis of keyphrase Head-Tail attributes; I think Avinash negelcts the navigational use of Google, which I believe makes his interpretation of the Head questionable.

The dotted vertical line is where the 10–visit plus keyphrases end, which I define as being the Head of the graph. So, as you’re telling yourself I am sure, the graph presents no surprises, since it is quite the typical head-long tail representation.

However, it does not tell the whole story of where is at with organic search. The main expressions, the Head, again represent only 1,4% of total traffic from searche engines (the tail should normally be 20 times longer than what it appears here on this graph). If you look at the number scale, that Head still does not produce that many visits. Why? Because the site does not compete well yet for the high volume keyphrases. If it did, I believe that not only the number scale would be higher, but the relative contribution to the site traffic from the Head would be much more important than 1,4%. If we expected to be Pareto here, we would say that it basically should bring 80% of all visits!

This site is currently building a deep, long, and very diversified Long Tail. The variety of keyphrases from which it gets most of its traffic grows everyday. But it still gets a lot of competitive pressure indirectly, since it does not position itself yet in the high-volume keyphrases. This means that when it does, traffic will explode.

Question is if and when this will happen. In three months? Two years? This remains to be seen. I certainly wish I had a predictive model here, but external conditions are always hard, if not impossible, to factor in.

In the next post, I will examine entry pages, which are strongly related to what’s going on with keyphrases, as you might expect.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.