Where Does Blog Traffic Come From?

Check out my about.me profile!

Galleries

Mapping The Blogosphere

GIS Data

Family

MaWaHa のピッツバーグ日記

Links

« The Digital Windfall | Main | Tracking People in the Blogosphere »

October 23, 2006

Where Does Blog Traffic Come From?

One of the areas of the blogosphere ecology that is perhaps less reported is that of referral statistics: how do readers get to individual blogs? Where do blog readers come from? By looking at publicly available referral logs we can get an idea of the distribution of traffic and some idea of trends over time. I've posted about this data before, but now that I have 50 days worth it is easier to see marked trends and make some more solid claims.

Search

The graphs below chart counts of referrals from the major search engines. These graphs show the count of referrals in the data against 51 days since the 2nd of September 2006 to the 26th of October 2006.

These graphs suggest that Google's traffic influence (from main stream web search) is increasing while Yahoo's is, for some reason declining. There are a number of possible explanations for trends of this type, including control over the placement of blog URLs in search result pages and basic coverage of the blogosphere in the index.

Feed Readers

Google's feed reader has received quite a bit of attention of late after a major interface update. Comparing traffic recorded from Google Reader with that from Bloglines we can see that while Bloglines is still winning by the numbers, the uptake of Google Reader is clear whereas Bloglines appears to be suffering a decline recently.

Meme Trackers

Memetrackers such as Memeorandum (including techmeme.com) and TailRank help people find stories that are being linked to including blog posts and news. The graph below suggests that traffic from the new comer TailRank is comparable with that from the market leader. Alexa stats suggest that the memeorandum.com domain and tailrank.com have similar reach - though this doesn't take in to account the combination of traffic from memeorandum.com and techmeme.com.

Blog Search and Portals

Google's blog search comes from either the blogsearch.google.com host or from search.blogger.com. Traffic from this source outweighs that from Technorati, considered the leader in the blogsearch space. It is interesting to compare these absolute numbers with those from mainstream search shown at the beginning of this post.

The Blogosphere

Of course, in this review of traffic data, it is important to include traffic to the blogopshere that comes from other blogs. The graph below shows traffic from the domains typepad.com, livejournal.com and blogpost.com.

Note that the graph is plotted with a log scale. Here we can see that traffic from blogspot is well above typepad and livejournal. Though livejournal accounts for many more blogs than typepad, it is connected to the core of the blogosphere (from which much of this traffic data has been drawn) comparatively weakly.

The key point with this data set is that traffic from the blogosphere to other blogs is a major piece of the story. This is wholly expected given our intuitions about the social nature of the medium.

Summary

Main stream search is still a major traffic driver to the blogosphere. However, there does appear to be movement from the major engines, with Google slightly gaining.
The feed reader space shows some similar, though more marked movement. Bloglines, the current leader, is being threatened by Google's feed reader.
Google, though lacking in sophistication, still has the numbers for blog search over Technorati, which is considered the market leader.
Traffic from blogs to other blogs is a very important part of the story and reflects in part the manner in which communities (at least at the gross level) are connected.

Update: I've had a number of questions regarding the data. The referal log data was gathered from blogs that have publicly visible sitemeter links. The crawl captures the referals each day at noon EST. This gets only a sample of the full referal data for each of the approximately 1, 500 blogs crawled. I've not yet worked through the full statistical implications of this approach, but I believe them to be sound to the above type of analysis. Some percentage (approximately 40%) of the referals are marked as 'unknown' meaning that the source of the click can not be determined. This percentage appears to be reasonably constant over time.

October 23, 2006 in blogosphere, socialmedia | Permalink

Comments

Can you give examples of these "publicly available referral logs"? This data is interesting, but I'm curious how biased the sample is.

Mihai Parparita
Google Reader Engineer

Posted by: mihai | October 23, 2006 at 12:59 PM

Mihai

You are right - I need to give some description of the data. I hope to get to that a little later in the day - stay tuned!

Matt

Posted by: Matthew Hurst | October 23, 2006 at 01:17 PM

Is the referal data coming from referers from mutliple blogs? Or just yours?

Posted by: Peter Caputa | October 23, 2006 at 04:27 PM

Alexa data just isn't reliable..... even if the data said that Tailrank was the #1 site on the Internet I'd say it's totally pointless. You might as well use a random number generator. :)

Also I believe that when sitemeter says "unknown" for the referer it's coming with the Referer HTTP header unset...

Good post though!

Kevin

Posted by: Kevin Burton | October 23, 2006 at 09:36 PM

Ah, mentions of Alexa and Sitemeter in the same post - my favourite! I wrote a bit about them here:
http://blog.simpy.com/blojsom/blog/2006/10/04/TechCrunch-vs-Lifehacker-Fishy-Stats.html

Interesting post. Is the crawl+parse script available? I'd love to look at some other numbers and other referals.

Posted by: Otis Gospodnetic | October 24, 2006 at 01:11 AM

Otis - thanks for the comment. There is no real magic to the crawling and parsing (wget plus some awk).

Posted by: Matthew Hurst | October 24, 2006 at 07:11 AM

Posted by: Bebo | October 24, 2007 at 11:39 AM

La mode du keffieh
A Paris, toutes les gonzesses se l’arrachent. Le Keffieh est revenu à la mode, et il est en ce moment-même un accessoire branché indispensable.
Jadis porté par les militants d’extrême-gauche (et accessoirement, plus tard, par les cailleras), le Keffieh est maintenant arboré par la jeunesse branchée, voire même friquée.
en savoir plus http://www.keffieh.fr
http://www.keffieh.fr

A la recherche d'un caftan marocain artisanal, d'une takchita classe et raffinée, d'une robe marocaine pour mariée, d'un jabador d'hôte ou d'une djelabba pour l'Aid? Nos tenues sont pour vous mesdames.
http://www.takchita.fr http://www.takchita.fr met à votre service le savoir faire des m3almiyas et les traditions des quatre coins du royaume.

Le meilleur de la série plus belle la vie
http://www.pblv.tv
http://www.pblv.tv

Posted by: Posao | February 13, 2009 at 07:49 PM

I recently came across your blog and have been reading along. I thought I would leave my first comment. I don't know what to say except that I have enjoyed reading. Nice blog. I will keep visiting this blog very often.

Alanna

http://www.craigslisthelper.info

Posted by: Alanna | April 04, 2009 at 10:45 AM

The comments to this entry are closed.

Subscribe to this blog's feed

Sun	Mon	Tue	Wed	Thu	Fri	Sat
		1	2	3	4	5
6	7	8	9	10	11	12
13	14	15	16	17	18	19
20	21	22	23	24	25	26
27	28	29	30	31

Data Mining: Text Mining, Visualization and Social Media

Galleries

GIS Data

Family

Links

October 23, 2006

Where Does Blog Traffic Come From?

Comments

Twitter Updates

March 2016

Recent Posts

Archives

Categories