TV has the Nielsen ratings. Radio has Arbitron. But how is Internet viewership measured? The closest thing we have is Alexa, which ranks websites in order of popularity. The catch: It only monitors the activity of people who have the Alexa toolbar installed in their browser.
To their credit, Alexa readily admits that this skews the findings somewhat. They have millions of users, but obviously there are millions more who aren’t being counted. Then again, there’s no reason to assume that Alexa users’ web-surfing habits are terribly different from everyone else’s. (For a while, the Alexa toolbar wasn’t available for Macs. During that time, I would assume Mac-related websites were woefully under-counted in Alexa’s rankings. Alexa still can’t be used with Mac’s Safari browser.) Alexa says its top rankings are pretty accurate; it’s only after about #100,000 that they think the sample size is too small to be completely reliable.
I became interested in all this a few months ago when I installed the Alexa toolbar in my Firefox browser. It includes a nifty extension that displays, down in the corner of the browser window, the Alexa ranking of every site you visit. I love that, because I love ranking things. It’s fun to discover that sites you assumed were hugely popular are less-trafficked than you thought, and vice versa. It’s also nice for me personally to notice sites ranked lower than my own.
My site, EricDSnider.com, is currently at #167,273. Whether that’s high or not depends on how you look at it, I guess. Considering there are more than 80,000,000 websites on the Internet, 167,273 is pretty good. On the other hand, that also means there are 167,272 sites more popular than mine. Coming in 167,273rd in a contest is not very impressive.
If you don’t already have Alexa installed, why not install it? You don’t provide them with any personal information; no one is monitoring which sites YOU, Eric D. Snider (if that is your name), are visiting. All Alexa tracks is that someone from a particular IP address went to a particular site. Plus, if every single EricDSnider.com visitor has the Alexa toolbar installed, it will artificially boost my ranking, which is directly tied to my self-esteem.
By the way, Alexa says the top 10 most visited sites on the Internet are:
1. Yahoo
2. Google
3. MSN
4. YouTube
5. Windows Live (Hotmail)
6. MySpace
7. Facebook
8. Orkut (Google’s Facebook rip-off, which I have never heard of)
9. Wikipedia
10. Hi5.com (yet another MySpace clone, and one I’ve never heard of)
The Alexa download Web site is blocked by the powers that be where I work, which is unfortunately where I read you the most. (ericdsnider.com is NOT blocked!) But I think I will give Alexa a go on my laptop at home. I wonder where Alexa gets the site contact info from. (I guess I’ll have to wait to see what it says for your site, Eric.)
Orkut is awful…
I use Safari. Sorry. But, I do try and post links to some of your blog entries at Apple and Mac related forums that I frequent, hoping to help drive up your readership (and thus eventually have your readership exceed the Daily Herald [what is their ranking?] and boost your self-esteem).
My site is at 1,327,109, which I would think isn’t great, but then I look at other sites for people who do what I do… and are much more well-known… and they’re well below that. So…? I agree that the sampling isn’t sufficiently diverse yet, as they say beyond #100,000.
Jeez, in the top 200,000 out of 80,000,000 is pretty awesome, I would say. I think it helps that you’ve got a bit of an online community going on here, what with the forums and the fans. It means people come back and check several times during their pathetic workday.
Sample size is not really very important (unless it is way too small). The fact that they are touting sample sizes that big as evidence of their reliability makes me think they have never actually studied basic statistics (or they think that all those sheep — I mean masses — have no knowledge of statistics). The samples that were collected by the Literary Digest for their 1936 presidential election poll were enormous, but they didn’t use a random sample, so their numbers were awfully flawed. That helped lead to the eventual downfall of the magazine itself, and it went the way of the dodo. Only a certain collection of people are likely to put Alexa on their browsers. Will this collection of people be truly representative of the whole population? Maybe. But there is no way to know, which sheds too much doubt on the reliability of any of their numbers.
Sorry, the math teacher in me is coming out too much.
Okay, but you owe me one, Eric.
If you’ll answer my questions the way Mr. Google does, I’ll visit your site more often.
Congratulations! 167k-th out of 80M puts you in the 99.8th percentile–just shy of three standard deviations above the mean. That’s like having an IQ of 140 or so, like scoring a 1590 on the SAT, or like being 6’5″.
(One could go on about all the assumptions necessary to make these numbers meaningful, but that would spoil the fun.)
That’s crazy that Orkut is number 8. Everyone I know uses wikipedia and that lost to Orkut and barely beat another site I never heard of. Orkut proves how big Google is, I guess. I always wonder how legit Nielsen ratings are. In the last season of Arrested Development, Nielsen said that only 3-4 million people were watching. Yet, in my attempts to save the show- I told many people to watch. To my surprise, many them told me they arleady watched. None of us have Nielsen boxes. It must be right though. It’s like the voter polls. They ask 1,000 people out of millions, and it still comes out almost right. It amazes me though.
Ah Eric I’d go grab this “Alexa” crap, but turns out, I have a Mac (Darn me). That’s gotta be BS, though…how many computers may have Alexa whose owners didn’t switch Internet Explorer’s start page from Yahoo to something they’d actually want because, turns out, they didn’t know you could do that.
I don’t care what “Alexa” says…EricDSnider.com is gracing the #1 bookmark spot atop my Safari Web Browser…(and a bunch of other people I know as well.)
I think I read on TechCrunch.com that Alexa rankings are very inaccurate.
I’ve never posted here before – I’m a lurker – but I just wanted to say that Matt is so right: your website is one of the only things that breaks up the monotony of my “pathetic workday.” I work in a bookstore, and any time someone claims not to know what an autograph is, or asks me if there’s a “sequence” to a certain book, I come here to restore my faith in humanity.
Or sometimes just to watch that David Letterman/Paris Hilton clip. I could watch that all day.
Point is, you may not rank higher than google, but I have to delete this site from the history on my boss’s computer a lot more often.
Eric, now that you have posted this request, you will never know if it is real or not when your ranking rises. 😉
As I am sure you have noticed by now, your ranking has risen to #163,298. Boy, just by telling people to download the toolbar you rose nearly 4,000 spots.
I think the reason Orkut seems really popular is because everyone in India has an account there. I live in an apartment complex full of East Indians and ones I am friends with were always encouraging me to get an account with Orkut. I’d never heard of it before I moved here.
EricDSnider.com is blocked at work, so I have to read it at home, like a chump.