Alexa’s front page has been undergoing some changes over the last little while. For a comparison have a look at the old interface and the new interface. It’s a pretty significant overhaul and not just stylistically.
Look at the old interface. It’s geared primarily as a search engine and/or portal. It’s meant to help the general public find information on the web, show them popular pages and so forth. What was the problem with that? That’s not what people used Alexa for. As a search engine/portal, Alexa stood as a dinosaur left over from the boom days. So they made a change, a big one.
When you look at the new interface you’ll see something that would be rather hostile to a user just poking around the web looking for random things. No this interface looks more like a very technical tool and that’s not by accident. You see it seems as though Alexa has realised that the people who really do use their website are the webmasters, advertisers and advertising brokers all over the web. The rest of the world doesn’t really care.
Interestingly enough this new role that Alexa finds itself in, as one of the major sources of 3rd party data on internet traffic is one of significant value. There just aren’t very many good metrics in this field that someone can comb through freely while exploring a good space to advertise. Conversely, there’s currently no better way for a webmaster to demonstrate their volume of traffic in an honest and open way. This is Alexa’s new audience, even though they’ve been there for a long time. The only difference is that Alexa has now conceded that it cannot compete in the search/directory/portal space and is putting its core users up front.
So where does the Alexa rank come from? Anyone on the web can go and download their toolbar and become a contributor to the Alexa rankings. It works in a similar manner as TV ratings do, extrapolating a broad population’s behaviour based on a smaller subset, but in Alexa’s case anyone is free to participate. Can it be manipulated? You bet! Since anyone can participate, there are lots of ways to manipulate it which is why advertisers only use it as one of many metrics to decide where to spend their dollars. But the closer you get to the top of the pile, the harder things become to manipulate. It’s generally suspected that gaming your way into the top 1000 is exceptionally difficult and probably just a waste of time really.
There’s a flip side to people manipulating rankings higher though. It’s also possible that your Alexa rank will be terrible despite the fact that you’re pulling lots of traffic. If none or few of your visitors have the toolbar installed then Alexa has little or no data to go by and can only assume that your website is not visited. So what’s the answer?
I’ll be honest with you, I don’t know. Obviously pulling more traffic should statistically bring in more toolbar users. But you should be able to increase your statistical chances by tackling topics that toolbar users are more likely to be predisposed to visiting. What are those topics? I have no idea. So post your thoughts and let’s talk this out.
Alexa, Alexa Rank, usability

3 responses so far ↓
1 mike // May 18, 2007 at 12:32 pm
The Alexa toolbar was a terrible piece of software for a long time. It may still be. Spyware/adware programs used to forcibly remove it, no idea if they still do or not.
I would expect that someone with traffic comprised of tech-savy users would be shortchanged. Similarly, a blog about privacy-related issues might suffer.
Someone with a lot of links from MySpace, on the other hand, would probably benefit.
Alexa doesn’t release things like “how many people visit your site”, so there is really no way to compare its accuracy with reality.
Personally, I ignore Alexa rankings completely. I think their data gathering methods are crap, their reporting is crap, and their toolbar is crap. If I want general information about the web, I’ll go to Nielsen/NetRatings. If I want information about my site, I’ll fire up any one of a dozen different log analysis tools. If I want to see how I am doing relative to others, I’ll fire up a search engine. My methods may not be accurate, but at least they don’t rely on a 12 year old installing a toolbar.
2 Dave // May 18, 2007 at 4:14 pm
I think that they look at page impressions if you put the html code on your site. Of course that would still only leave toolbar users and sites with the counter in contention.
3 Marc // May 18, 2007 at 10:23 pm
Mike: You’re right, Alexa is certainly inaccurate and I don’t put much stock in it either, but advertisers do. If you’re not involved in selling ads without a contextual broker (e.g. Adsense) then it’s not an issue. If you are in that business, then Alexa rankings become an issue. It’s not the be-all-end-all metric, but it’s one of the ones they use.
Dave: I had no idea that you could actually dump Alexa code on your site to have yourself counted. I don’t know if that would do me good or do me harm though