|
|
|
|
|
|
Website frequentation:
an accentuated seasonal dip for July 2007
|
|
| The frequentation of websites is seasonal, with notably, the usual summer break. This study compares the seasonal effect of the month of July 2007 to the average effect of the months of July for the 5 previous years. |
| Perimeter: |
- Study conducted from January 2002 to July 2007
- Range of 206,399 websites
|
|
| Could Internet users be leaving earlier on holidays? |
The analysis of the website frequentation (see our study “Websites in 2006: slowdown of growth) ”, allowed us to observe the existence of audience seasonality for websites, with key moments over the course of the year like the summer dip or the January revival.
Thanks to non-parametric methods of deseasonalization, we have been able to extract these monthly, seasonal effects.
The graph below presents the monthly, seasonal effects recorded on website frequentation between 2002 and 2006. The months of January and March experience strong seasonal effects with, respectively, peaks of +10% and +11% in website frequentation. Whereas it is the month of August that is the most marked in terms of dip: -13 % in websites frequentation, followed by the month of July.
|
Between 2002 and 2006, an average month of July is characterized by a dip of -9% in website frequentation. What about for the month of July 2007?
In order to characterize the seasonality of the website frequentation in July 2007, we compared the seasonal effects of the months of July 2002-2006 to that of 2007.
Thus, the seasonal effect of July 2007 is more important than the average. It indeed corresponds to a dip of nearly -12%, i.e. more pronounced for a third than the average of seasonal effects of the months of July 2002 to 2006. This phenomenon also occurred during the month of July 2006.
|
|
The summer dip, until then less pronounced in July than in August, reached nearly -12% in July 2006 and 2007, i.e. practically the same monthly average seasonal effect recorded on the website frequentation for the months of August 2002 to 2006.
It is also interesting to note here that during a period of one month the weather does not necessarily have an influence on website frequentation: with poor weather in July 2007, the dip is as marked as in July 2006 which had hotter weather. Whereas over shorter periods, such as the first beautiful weekend of spring, that can effectively have more impacts (see our study available only in French “More sun, less Internet…”).
How will the seasonal effect of August 2007 be characterized? Is the trend observed over several years between July and August in the progress of evolving?
Rendezvous soon on XiTi Monitor with our latest evolution on the seasonality of websites frequentation.
|
Methodology:
The audience recorded on the websites by XiTi made it possible to detect very early the existence of seasonal effects, that treatment methods allow to extract.
In this study, the series that we were interested in is the frequentation of Internet sites. It regards the evolution of audience generated by a fixed number of websites. The series of data successively integrates the daily evolution rate, calculated at a constant perimeter over the totality of the websites audited by XiTi. The evolution rate thereby integrates into the mode of calculation the arrivals and departures of websites from the XiTi perimeter.
Tools adapted* to the treatment of chronological series allowed us to proceed with the decomposition of this monthly series. The components are:
- The series trend that represents the long-term evolution of the series.
- The seasonal composition representing the fluctuations that are infra annual, monthly, and in our case, those that repeat themselves more-or-less regularly form year to year. It highlights the phases of growth and recession.
It is this seasonal component that interested us in this study.
*Non-parametric method of deseasonalization.
|
|
|
|
|
|
|
|