Alexa service is the first one which tried to show real number of visits to all Internet users. At the beggining it was very amicable service, but today you can see a lot of faults
in its system. Alexa is not objective and neutral service. It becomes clearly commercial service and its results are not objective any longer. In this text we will show you some search
engines, directories and how many Alexa visitors they had, and its deviate in regard on real number of visitors.
The most visited Internet site Yahoo!, have 12,933,162 Alexa visitors, but the real number of visitors is 20.5% higher. This directory is disabled for 2,651,300
visitors by service Alexa. HotBot have 1,797,891 visits by service Alexa, but the real number is for 9,1% higher.
Excite search engine have 3,731,628 of total Alexa visits. Real number of visits is 2,5% lower. Our next example is one of the best and the oldest search engine
AltaVista. By Alexa service, this search engine have 5,971,503 visitors. Real number of visitors is 8,8% lower. This is not too bad if you read next two examples: MSN Search have
4,806,846 Alexa visitors, but real number of visitors is for 22,5% lower.
Ask Jeeves have 2,274,887 total Alexa visits, but real number of visitors is 29,7% lower. Probably, the best
example of great deviations which make Alexa service very poor is Google. Google became one of the most popular search engines as the first one which have indexed 1 billion pages.
By Alexa's information, total visits are 1,816,553 , but real number of visitors is 68,1% lower.
All these data show that Alexa service has many faults in its system of counting visits. It is more noticable when we compare number of visits at middle
and small Internet sites. One thing is sure, and you can see it if you look at all these search engines: almost every search engine which have some URLs without www at the beginning,
get lower number of real visits. Alexa team will need many improvements in its system of measurement, if they want to be objective and neutral system of ranking sites by visits.