In Praise of Alexa

Now that Google has stopped ranking websites - yes page rank is long dead - there is a real need for some public auditing and ranking.

 

stonetapes.

 

Alexa has a long history as a unique and very public way of measuring web site success. 

The way it works is simple and the way that it gets things wrong is equally simple. You might well not like it, but I would argue that it or something like it is a very necessary part of the web landscape.

You might well ask what makes Alex-style rankings so different?

After all, there are no end of web analytics services,  most notably Google Analytics. Each one will record the details of the traffic to a site down to the time spent on a page, which menu items were clicked and so on. Why bother with an alternative service that is admittedly not as accurate or as detailed.

The key difference is that Alexa, and Alexa-style rankings, are public.

You can look up the Alexa rank of any website and a website can't opt out of being included. A website doesn't even have to put itself forward to be included in the ranking.

Why is this important?

Because websites, especially big commercial websites, keep their traffic figures a closely guarded secret for reasons that are not difficult to fathom. Web traffic varies in ways that are unpredictable and reading too much in to a slight drop in traffic over a few days would be a really good way for the wider media to create some good headlines. Also correlating a site's web traffic with public management changes would also lead to speculation of success or failure. In short, making accurate and detailed traffic stats public isn't a good idea because the very accuracy and detail can be used against you. It gives too much away to your competitors.

Compare this to making public data that is slightly doubtful and not at all detailed. You will might not like it, but its deniable and anyway there isn't anything you can do about it. This is Alexa's role.

But why is this a good thing?

Why do end users and others need to know how popular a site is.To be honest they don't, but they often like to know a sites rank. It was also suggested recently that users could be reassured by checking a site's Alexa ranking to discover if it was trustworthy or some scam set up a few days before. This is an interesting idea, but probably not a good one to rely on at the moment.

As a website publisher you need to have some idea where you sit in the ecological niche your site occupies. This is difficult to ascertain if you are only looking at your own data and, no matter how detailed this is, it cannot tell you how much more traffic you could be getting.

If your site is designed to appeal to widget fanciers and you are getting 100 unique visitors per day after a few years of existence, you really have no way of knowing if this represents a good proportion of the entire population of internet-using widget fanciers or that you are simply not penetrating the potential market of 1000s of unique visitors per day. You can do some theoretical estimates, and you might even be tempted into some market research, but what you really want to know is how well are you doing as compared to other sites targeting widget fanciers. One way of discovering this information is to simply compare the success of your site with those of your competitors and at the moment Alexa provides about the only way of doing this.

Of course, it is also obvious that advertisers would like some way of judging if it was worth approaching a website for advertising data. An informal look at a site's Alexa ranking is one way of deciding how much time it is worth spending considering using it for advertising.

So Alexa is a great idea but now we come to the problem.

Currently Alexa is flawed.

The original idea of Alexa, setup in 1996, was to offer users a useful toolbar in return for data on which sites they visited. The data was, and is, collected anonymously and overall it seems like a good deal. A nice shiny toolbar in return for letting the world know what are the best sites.

But Alexa was born in more innocent times. It didn't take long for the paranoia of the Internet regarding privacy to start to cause problems. Some people, including Microsoft, labeled the Alexa toolbar malware and spyware and all sorts of nonsense. Users reacted badly to having it on their machines and, to be honest, as browsers improved the Toolbar didn't really offer much that a user would want.

Today the situation is even worse. Amazon took the company over in 1999 and it has been downhill from there. A revamp in 2009 offered some hope that the service would be reborn, but the Toolbar still remains uninspiring. When you include problems of keeping up to date with the rapid release schedule of Chrome and  Firefox, and the difficulties of finding useful widgets to include on the toolbar, then it is surprising that any user has the Alexa toolbar installed. As a result the Alexa ranks are based on a very small and biased sample of users, making the rankings even more doubtful. There is also the problem of the "long tail". Websites that rank over a million get so few Alexa toolbar-using visitors that one visit can make a big difference in their ranks.

There are currently no real alternatives to Alexa, even though it is currently broken. There is Compete.com, which uses a "panel" of users in much the same way as the TV viewing data is gathered. However, the panel probably isn't big enough to really do justice to the huge number of websites that need to be monitored. It like Alexa uses a toolbar and a range of other measures but it has similar problems.

We do need something like Alexa.

The newspaper and magazine industry has long had mechanisms that provided audited circulation figures. It is time that the web had something similar. It would be possible to organize some sort of scheme whereby websites joined an audited page impression scheme, but this has its problems of getting enough sites to join and there are problems in being too accurate with such data.

There are advantages in having public site rankings that are not too accurate so that not too much can be read into their random fluctuations. This may seem like an odd idea but deniability is a strong point of Alexa rankings.

Rather than some standard scheme for auditing site traffic, making the Alexa Toolbar desirable seems a much better idea. Perhaps Amazon could be persuaded to give away a book or a downloadable game on a monthly or weekly basis via the Alexa toolbar to a lucky winner.This might transform a toolbar that is viewed as malware by paranoid users into something that might win them a prize.

Whatever the mechanism, the industry needs something like the Alexa ranking system and it needs it to work.

More from:

stonetapes.

 


Google's War On Links - Prohibition All Over Again
03/08/2012

Given that Google's life blood is the hyperlink you might find it strange that it has just declared war on said link. Not only that, but in its attempts to make search better it has placed a weapon in [ ... ]



Reddit Admits Banning Major Sites
18/06/2012

Reddit has caused some surprise by announcing bans on Business Week, Phys.org, The Atlantic and more. Some Redditors are happy that evildoers are punished, some are not so sure, but this news shouldn' [ ... ]


Other Articles

Last Updated ( Friday, 21 October 2011 )
 
 

   
RSS feed of all content
I Programmer - full contents
Copyright © 2014 i-programmer.info. All Rights Reserved.
Joomla! is Free Software released under the GNU/GPL License.