What Have You

Some bloggers requested me to write about this article. The demand is so high I feel like Tom Cruise while writing this article (just kidding, lol!). Kidding aside, the topic regarding Alexa rank has been in my archive for about 2 weeks now and I think it’s high time it gets out of the dust bin and gets some polishing.

For a little info, Alexa is owned by Amazon.com and is responsible for measuring traffic gained by websites on the internet. Alexa is considered a good measure of how many visitors and page views a site gets on a periodic basis (daily, weekly and quarterly). In doing paid blogging, an advertiser usually considers the traffic a blogger’s site gets as measured by Alexa. In fact, even companies providing opportunities for bloggers to earn money (like Smorty, PPP, SponsoredReviews, etc.) require a certain amount of traffic from sites submitted to their networks before giving the site owner the go signal to grab opps. Advertisers do not differ significantly on how they view traffic measurements provided by Alexa. A site is evaluated by an advertiser to determine if his products/services can gain enough attention if in case he places his ads on that site. It provides advertisers some insights ahead of time, a projection, on how successful an ad campaign will be based on Alexa ranks of a site or blog.

On the other hand, there are various concerns over the ranking system of Alexa. Based on Alexa.com’s articles concerning how they measure traffic, data is collected only through the Alexa toolbar. The Alexa toolbar, for your information, is a downloadable toolbar from Alexa’s site which can be integrated in popular browsers such as Internet Explorer and Mozilla Firefox. This is the only way Alexa can gather information from internet users and what sites they visit. Hence, if you don’t have the Alexa toolbar installed, then your visits to and page views of a website will not be counted by Alexa. This is understandable since there is no other proper way to determine if a site has really been viewed without using a set of codes or scripts to track the users’ internet activity. However, as a means to gain traffic from Alexa toolbar, a lot of traffic-hungry website owner use a “redirect” where the user is redirected to Alexa’s website then back to their site. This is done by a simple URL replacement. For example, normally, when a commentator leaves a comment in your site, the URL he usually leaves behind the post is in the form http://www.his_URL.com, but when he wishes to have a redirect to Alexa (hoping that his Alexa ranking will increase), he’ll instead leave a URL like this http://redirect.alexa.com/redirect?www.his_URL.com. There is no proof that this technique works and reports of so many bloggers using this technique indicate a lowered Google Page Rank. In fact most gurus claim that this is a dirty strategy. Alexa, knowing that this has become a practice among bloggers wanting to increase their Alexa ranks in the shortest possible way made every redirect by any website owner to give a 404 error code. The lowered Google PR care of this redirect is logical since the person who clicks on the Alexa redirect link goes to Alexa first, not on the site. Hence, the credit goes to Alexa, not to the site. Moral lesson: Don’t use any redirect. Akismet also recognizes Alexa redirects as spams. If you don’t want to be banned from commenting in WordPress blogs, don’t leave an Alexa redirect in anyone’s blog.

On the other hand the validity of Alexa users’ representativeness (compared against the total number of internet users has been in question for a long time). Alexa’s user base sample is unknown up to this writing and many skeptics (including this author) question the validity of the sample (which refers to the total number of Alexa toolbar users divided by the total number of internet users). Simply put, what is the percentage of internet users who have installed the Alexa toolbar? Alexa even claimed that there is a bias in their measurement of traffic as websites may be getting more actual traffic than what Alexa’s instruments can measure and reveal to the public. This means that if you’re using statistical softwares to determine how many visitors and page views your site gets per day (e.g., Histats, Sitemeter, etc.), don’t expect that you’ll get the same results from Alexa. This is because I’ve never encountered any case where the actual traffic (based on Histats) was equal to Alexa’s traffic measurement.

Further, the Alexa’s rank is a combination of page views and reach (number of users who visited a particular site divided by the total number of internet users). However, the problem with this is combination is the actual distribution of weights on 1) page views and 2) reach. It has never been made public how Alexa puts weight on the two criteria. Is it 50-50, 60-40, 70-30? I honestly do not know. What is not clear to us is also the computation of reach. Is the equation in the form:

Total number of visitors to your site for that day
total number of internet users for that day

or

Total number of visitors to your site for that day
total number of internet users as determined by total IP addresses registered ?

In addition, I doubt the values of the ranking system itself. How is an Alexa ranking of 1,000,000 different from an Alexa ranking of 1,000? There is no clear definition as to how much traffic is represented in these figures. The general connotation about the Alexa ranking system is, the lower the value, the better. For a regular blogger, a lower Alexa figure simply means that his site got more/higher page views + reach for the day/week/quarter. Period. No questions asked. However, mathematically, this ranking system is vague. As a statistician, I sometimes devise a method of interpreting a certain value (either individually or group-derived) when data gathered for a study is ready for processing. For example, consider the following:

INDIVIDUAL VALUES

Value                     Interpretation

5                         Very good

4                         Good

3                         Average

2                         Poor

1                         Very Poor

GROUP-DERIVED VALUES

Actual Scores             Grade

97-10                         1.00

94-96                         1.25

91-93                         1.50

88-90                         1.75 and so on

In view of Alexa, how can a rank value of 1,000,000 be interpreted? Does this mean your site gets 100-200 visitors with 1-2 page views per visit? Or, does this mean that your site gets 150-250 visitors with 2-3 page views per visit? If you do not know, then until now, it’s not really clear to you as it is to me.

But to cut the long story short, if your Alexa rank is high, then the value is represented by a low number. Increasing your Alexa ranking is another story. For now, understanding your Alexa ranking should be your first concern and I hope I did my part in making you understand what a simple number can do to what-used-to-be ideal bloggers into traffic-hungry ones.

RELATED POSTS:

1. Did Google Strip Off Your Page Rank Sooner than Expected?
2. Google Page Rank Update
3. Entrecard is the New Trend to Get Traffic to Your Website