Fleen: The Awkward Christmas Dinner Of Our Obligation To Existence

Work has me crawling under tables doing physical inventory of computer assets; my back and my skull are tired.

If you haven’t seen the new Topatoco store, the bells and whistles are pretty much in place now. Of note: discounts apply to purchases (for example, three t-shirts), even from different creators that hail from different studios. While you’ve been able to get, say, a Dumbrella Val-u-Pak o’ Shirts at a convention, now you could do so with wearables from, say, Dinosaur Comics (Dayfree Press), Dresden Codak (of Koala Wallop), and Three Panel Soul (of nobody in particular). Remember all my blatherings about Aduz? With this common storefront service, Jeff Rowland has taken the first step towards that bright and shiny future. Now, who’s gonna step up and do the same on the manufacturing end?
Okay, so everybody knows that Alexa rankings lie somewhere between “fanciful” and “science fiction”, right? Which is why I didn’t spend too many mental cycles on Mr T‘s surveys of webcomics traffic rankings based on Alexa and similar services. Unreliable data in = unreliable conclusions out. But he’s done something that may actually wrest some value from that morass of non-metrics.
Namely, he’s pulled data from Alexa and Compete for his latest crack at analysis, but he’s also going to be retaining numbers from survey to survey instead of treating them as discrete events. The numbers in a particular survey may be utter crap, but the trendlines from month to month may actually be useful. If each survey has approximately the same bias in the data, the differentials may be somewhat reliable.

A’course, that’s a hell of an if, but time will tell. In six months, if we see the lines yo-yoing back and forth like a drunken clown, we’ll be able to abandon the entire exercise until the technology improves. If not, getting some of the sites in question to agree to disclose their traffic may help to refine the model to the point that it can actually tell us something. For the moment though, please look upon the numbers as requiring significant amounts of salt.

I still don’t think it’ll make much difference.

Alexa still only works on Microsoft Internet Explorers browsers — officially, at any rate. While Internet Explorer makes up the majority of the internet browsing experience (though not by as large a margin as it used to) it only accounts to somewhere between 25-33% of my site traffic.

So Alexa is basing my alexa ranking on a quarter of my readership. Oor a third of it at best.

Also we need to factor in how many new people start using Alexa each day compared to how many new readers you get each day. If your site is growing but not as fast as the Alexa userbase is, your site will slide down the rankings.

It’s bad math on the internet.

I don’t know much about Compete, so I can’t speak to how accurate they are.

Also, I hate your recaptcha.

Gary

Give me a little credit for exploring the Alexa idea before Mr. T. I went there and thrashed it out with many in webcomics before deciding that while it was interesting it was too problematic to do as a regular thing.

I have my own ideas on what would work better (maybe coming soon one day…), but just going to go to the Alexa well here again — I don’t see it as much of an improvement (from what I’ve pulled from Compete -out of curiosity – it’s even worse than Alexa).

Christopher:

There’s also an Alexa toolbar for Firefox called Sparky (I’ve had it installed for months), so it’s not just limited to Internet Explorer anymore. http://www.alexa.com/site/download/

My site stats tell me that 60% of my readers are using Firefox! I always used that to justify my relatively low Alexa ranking… I guess I’ll have to come up with a new excuse now. ;)

Well Sam, you also have to take into account whether or not your readership would be inclined to *use* the Alexa toolbar. Since a lot of my readership actually uses banner-ad blocking software and most are leery of spyware to begin with I’d say most of my regulars still don’t use it.

Hey, that’s why we call it “the unreliable survey,” Gary. I’m hoping that the more data we can gather on this matter, the more useful it will become.

But I’m also hoping that in a year or two, I can chuck Alexa and Compete for two or three sounder systems.

As for your desire to have the data confirmed by top-tier webcartoonists– I wish you luck, and you’re certainly better at the networking thing than I am, but I think you might run up against the problems I outline here (search “Alexa” to get to the relevant Q&A).

And Christopher: I don’t want to wave the Alexa flag, but I don’t think your objections are exactly sound either. Sampling is always based upon a fraction of the audience. For a comic of any size, I think less than 1% of the audience is in regular communication with the cartoonist, and that 1% tends to be the most dedicated fans. I have no doubt that the hardcore Help Deskers distrust “spyware,” but do they represent your entire audience in that? Because you are also being read by people like me, who at least used the Google Toolbar for a while.

[…] Let’s play ‘what if’ for a moment. This page has called for the creation of ‘Aduz’ or an ‘anti-Zuda’, a […]

RSS feed for comments on this post.

This Is Gonna Be Short

Contact Us

Archives

Search

A Good Start