Looking at the TTLB Blogoshpere Ecosystem, I have been a slithering reptile for quite some time. This designation is based on the number of incoming links to my blog. I am currently ranked #1226 with 29 incoming links.
What is interesting is that my traffic has been increasing steadily but my ecosystem rank is stagnant. According to the TTLB traffic stats, I get 298 visitors per day and am ranked #224 by traffic (the traffic ranking does not include all the blogs in the ecosystem).
To check whether my ecosystem ranking and traffic are comparable, I decided to get the data for the top 447 blogs by traffic. This makes my blog the median by traffic in that list. The traffic varies from 76797 visits per day for Instapundit to 110 visits for Randgaenge. When I organize this list by ecosystem ranking (incoming links), I am #338 (or 114 positions lower than the median). The links ranking starts again with Instapundit (#1, 2195 inbound links) and ends with JenLars (#4579, 0 inbound links).
It seems then that I don’t get as many links to my blog as my traffic would suggest. That seems correct as quite a lot of my visitors nowadays come through search engines.
I wanted to do a regression analysis on visitors per day and number of inbound links but for that I would need to write a parser to extract the data from the HTML files. Too much work! I have grading to do.
I’m not sure what you can take from the stats of that ecosystem. For example, back in May it said I had 64 links and ranked #242. Today, it says I have 82 links and rank #479. Meanwhile, Technorati says I have 144 inbound links.
Go figure.
Frankly, I don’t think any of it means much of anything. Bloggers don’t line up well in a ladder orientation. They seem to be more in circles or pools. Therefore, a linear ranking can be misleading. There are “Top 50” weblogs I’ve never heard of or seen before.
1) there are many people who obviously link-whore.
2) many bloggers don’t change their blogrolls and add thousands of links (so it seems)-so old blogs with low traffic might be a bit over-ranked when comparing links vs. traffic.
3) you don’t talk about stupid shit like the war in iraq-so you get fewer links from the blog-mob. the blog-mob does a lot of blog-on-blog-commentary-circle-jerking over topics like “us in iraq.” this ups the links a lot. on your blog you tend to talk less about topics that might be circle-jerked about (like say ideofact or headheeb). on my bog (GENE EXPRESSION) there are topics that rarely get circle-jerked (genetics, science, commentary on the transition between paganism to christianity in the late roman empire, etc.), but when someone (usually godless) posts something on neocon foreign policy or iraq, we get a burst of links.
4) you don’t have an assinine & adverserial personality. a lot of bloggers get links by picking fights. we’ve actually tested this process at GENE EXPRESSION-and it is an automatic way to boost traffic & links. you bitch someone out, they bitch back at you, but link to your bitching, you bitch back, and so forth. other bloggers get involved in the bitch fest and all of a sudden you have 10 extra inbound links
5) be careful of sitemeter, i suspect it overestimates traffic, perhaps by as much as 1/3. i have 3 meters on my site, and while the low one has been giving about 1,000 hits/day the high one is giving 1,500 (sitemeter).
6)links do matter for google algorithms…
7)…or do they
in early september my blog’s traffic went from around 500/day to over 1,000/day in about 3 days. mostly because of extra search engine traffic. something happened with the google algorithm, but our links coming in were pretty static. our traffic ranking is #54 at about 1500 visits per day-but our ecosystem rank is 406-go figure!
8) you should really look at Average Visit Length as the real crucial metric. i think long visits are quality visits, people that pop in and out via search engines are less important (though some of those become regulars).
9) i’ve been thinking about this way too much 😉 to honesty-a lot of this recent obsession has been motivated by two things:
* we got kicked off one web host because our traffic increased to the point where the “rebuild” with the CGI scripts were hogging our server’s resources
* i am worried that the extra traffic will decrease quality of discussion on the message boards (this always happened when we an INSTALAUNCH-basically an ingathering of low IQ morons who want bite-sized opinions). right now our average visit length is 3 minutes and 28 seconds. and not too many morons have shown up.
btw, i meant that ideofact & headheeb are blogs that also don’t talk about stuff that might be circle-jerked, not that they do. 😉
“I wanted to do a regression analysis on visitors per day and number of inbound links but for that I would need to write a parser to extract the data from the HTML files. Too much work! I have grading to do.”
Or, you could just ask me for the data. I don’t bite, honest.
Drop me an email specifying what in particular you’d like, and chances are I can do a quick extract from the database into a .CSV file for you…
-NZB
PhotoDude: Your rank probably has slipped as a lot more blogs have been added to the ecosystem. About the difference with Technorati, I think Technorati searches more blogs plus it keeps the inbound links longer.
There are ‘Top 50’ weblogs I’ve never heard of or seen before.
Same here. Blogging does make sense more as communities with lots of intracommunity links but few intercommunity ones.
razib: Agreed generally with your points.
be careful of sitemeter, i suspect it overestimates traffic
I am not sure about that. Compared to some other counters it does, but when I compare it to my Apache logs, it actually misses some traffic. My average page requests per day over the last month according to an analysis of my logs is 846 (that includes my own accesses as well) while Sitemeter says 390 page views. These are not directly comparable but should be roughly equal.
you should really look at Average Visit Length as the real crucial metric.
True to some extent. But look at it this way: If all your visitors are regulars and you don’t use extended entries (I don’t) then they don’t need to visit more than 1 page (except for commenting). So their visit length would be zero. On the other hand, you do want new visitors to visit more than 1 page.
NZB: Thanks for your offer. I have sent you an email.