70 Of the Most Popular Blogs In Malaysia
May 30, 2006 by Gaman
Syed Syahrul has published a list of the most popular Malaysian blogs at his blog.
Based on Alexa traffic rank alone, Sabahan.com should have been included on the list. With an Alexa rank of 52, 676, Sabahan.com is actually number six on the list. Not bad for a three (or is it four?) months old blog right.
No, I am not complaining for the exclusion. It just goes to show that you can’t relay on Alexa to gauge sites popularity.
How Alexa ranks a website
Alexa ratings are based on a level from 1 to 4,000,000 and it goes beyond 4 million sometimes. The lower the rating on Alexa the better. So if a website has a ranking below 100,000 then the website should be getting a good number of daily visitors.
3,000,000 to 4,000,000 ratings are for new sites mostly, or sites with no traffic. While a rating of 1 will be the website with the most traffic, a spot currently occupies by Yahoo.
There’s a formula I’ve found from Microsoft-Watch.org which estimates the number of daily visitors a website gets when Alexa’s rank is available. I won’t say it’s terribly accurate as the value could deviate from the real value by a wide margin depending on your website target audience, but still, it offers an easy way to predict the traffic levels of a website.
Daily Visitors = AlexaRank^( -0.732) x 7,000,000
How Reliable Is Alexa?
The main flaw in Alexa is that it only measures visits from people with the Alexa toolbar installed and Alexa assumes that this group of surfers are representative for all Internet surfers.
Another flaw in Alexa is that it may not reflect the average web surfer. The result can be biased as people who use Alexa toolbar are more likely to be webmasters, marketers or technically oriented people. The bias goes towards as to which sites these groups of people are likely to surf.
So it’s not surprising to know that two websites which have about the same traffic level would have a totally different Alexa rating each because one site caters more towards webmasters.