NPR: TechCrunch Rules. Web Stats: Uh ... No [UPDATED]

(MoneyWatch)  NPR's On The Media had an interview with Michael Arrington in wake of the New York Times piece on process journalism. But when I followed a link from Twitter about the story, blanket uninformed and unverified -- and easily disproven -- statements about the "dominance" of TechCrunch were the real surprise.

Here's the statement that smelled like week-old mackerel:

No news site dominates quite like TechCrunch, a network of websites and blogs that has basically buried all other outlets, online and off, in coverage of technology and tech enterprise.

Certainly I don't always agree with TechCrunch and have noted ethical problems in their coverage. I do, though, generally respect the outlet.

But "buried all other outlets, online and off"? Did anyone at On The Media, a show focusing on "the process of 'making media,'" do enough research to verify that statement, or was hype enough? Here's the disclaimer that should hardly be necessary, given that it's so obvious: I write for BNET, which is affiliated with CNET and ZDNet, so technically am in competition with TechCrunch, though I can't get into the whole us-versus-them thing.

Luckily, none of us have to rely on opinion or assertion. I immediately went to Compete.com to compare the web traffic of a variety of domains, because that would be a logical measure of how relatively important the sites were. (I'm not sure how you could accurately measure influence or the quality of coverage.) Here are the May unique visitors numbers:

Comparison of May 2009 Web Traffic

Domain	Unique Visitors
TechCrunch.com	1,862,612
ArsTechnica.com	988,578
ZDNet.com	2,337,743
CNET.com	32,597,141
Computerworld.com	787,650
InformationWeek.com	649,652

This is just comparing web sites. And what's the readership of the New York Times technology section? Or BusinessWeek's technology coverage? To say nothing of how many of NYTimes.com's 14,942,069 visitors last month, or Businessweek.com's 5,092,423, were reading technology coverage, given that both publications have some of the more prominent names in tech journalism. I'm also assuming that OTM was literal when it talked "tech enterprise." Otherwise I could have added such sites as Engadget.com (2,038,259 in June) or Gizmodo.com (2,166,978) or BoingBoing.net (1,902,371).

Granted, all such reported numbers can be off, but looking at them in total, they suggest that TechCrunch is simply not top of the heap.

Apparently being willing to do a bit more digging than On The Media, I went to Alexa and checked popularity ranks by the tech news and media category. All of a sudden, TechCrunch was number one. But there was something odd: neither ZDNet nor CNET even seemed to be on the list. When I did a comparison of these three sites even on Alexa, CNET was far and away still the leader in traffic rank:

So On The Media clearly went to Alexa, didn't talk to someone who had a real sense of who seriously covers tech, and didn't realize that traffic rank is a relative analysis and not necessarily one lending itself to definitive statements. Or maybe On The Media was trying a shot at process journalism and is waiting for sources to give a call.

[UPDATE: Simon Owens at Bloggasm got in touch with me, surprised at the difference in traffic. He asked whether I had looked at the other domains that TechCrunch owns. I hadn't but just did. Here are the results from its other domains:

Other TechCrunch Domains

Domain	May Unique Visitors
CrunchBase	467,226
CrunchGear	595,716
MobileCrunch	169,251
TechCrunchIT	123,704

That would be another 1,355,897 for the month, which would give a total of 3,218,509. That's still far shy of CNET. I also suspect that there's a lot of crossover in the readership. Also, CrunchBase is a database of companies and people involved in high tech, so that doesn't fall under covering technology per se. That reduces the number (artificially inflated by multiple counts of unique users) to 2,751,283. It's an impressive number, but simply not one that "buries" all other outlets. And, to be clear, this is no intent to slam TechCrunch, just to correct an inaccurate statement about tech media in general. And I guess this would be an example of process journalism.]