About Hitwise
Hitwise is the leader in online competitive intelligence. Contact Hitwise to maximize your online marketing programs.
Hitwise Intelligence - Bill Tancer - North America
Analyst Weblog
« Ron Paul for President 2.0 | Global Warming: A Search Term Analysis »
Facebook: Confusing Slowdown with Seasonality
October 18, 2007
I've seen several posts circulating in the blogosphere postulating on the demise of Facebook given a drop in the sites traffic over August and September of this year. My immediate reaction was to turn to Hitwise data and provide a little more granularity to provide insight to the question of Facebook slowdown v. seasonality (my hypothesis on the recent drop in traffic).
Here's a monthly Facebook chart with two years of historical data:
Now look at the same chart on a weekly basis:
Finally, lets zoom in on August - October 2006 on a weekly basis:
The 2006 dip is minimized in the two-year chart given the growth of Facebook during the timeframe. At Hitwise, with a sample of 10 million Internet users (which include home, work and educational users), we have the advantage of granularity... even down to producing daily traffic charts. Based on the previous year's pattern, and the uptick in Facebook traffic over the last week, I think the argument that the dip over the last two months is seasonal is certainly plausible.
On another note, I'm speaking at the Web 2.0 summit today at noon, if you happen to be at the show swing by and hear our predictions for the next hot 2.0 sites.
Posted by Bill Tancer at 08:10 AM
| 
(3)
| 
(1)
In Categories Social Networking
TrackBack
TrackBack URL:
http://weblogs.hitwise.com/movabletype/mt-tb.cgi/620.
Links to weblogs that reference this entry:
- Google to Facebook: If We Can't Buy You, Then Join You (PardonMyFrench)
I thought this announcement by Google in the post called Google News Goes Social was a brilliant move on their part. Google created a Facebook application that lets users have news and customized searches appear on their profile pages. Of [Read More]
Comments
Bill: While it's great that Hitwise is releasing some data, this post is about a week late, e.g. blogger Andy Beal already made the same point with Hitwise data from last year.
Chuck: that's possible, but seems like justification rather than explanation. I think it's more plausible that actual traffic increases as college students connect with new folks before tests start cutting into their free time. :-)
I'm inclined towards Quantcast's explanation (per GigaOM): it's a flaw in the panel methodology. Hitwise and others who use panels may not like that conclusion, but they could do a real service to their clients by digging deeper rather than pretending that a single approach (no matter how hard they work at it) will suffice for every subset of internet users.
Posted by Scott Lawton (Blogcosm) | October 18, 2007 05:04 PM
Scott,
Thanks for your comment. I don't buy Sutter's explanation for a couple of reasons. First, our methodology differs from the others in that we partner w/ ISPs and supplement with panel members. With our sample we're capturing home, work and educational users. Second, If it were true that Facebook dropped because kids were going back to school, then the trend would be for a continued drop or flat-line till the next vacation. As the charts show Facebook is back on the rise.
Posted by Anonymous | October 18, 2007 05:40 PM
Archives (view all posts)
Categories
It is fairly clear why the dips occur.
August to September - People return to school, want to hang out with real people. October - start using it again to hook up and stay in touch.
December to January - major break at school. people are enjoying christmas and the new year holiday.
Growth in between - people don't like working or studying. Facebook is a much more entertaining way to spend your time....
It will be interesting to see if these dips occur on a 3 to 5 year rolling basis.
Posted by Chuck Rosendahl | October 18, 2007 01:22 PM