Introduction 

  AELXA know that the director is that more and more, but I only recently discovered that many of the station and did not carefully read ALEXA statements, and only looking at a figure that ranked.    In fact, very often, AELXA data provided by the other two - each of the millions of people each of the hits and page views - is more worthy of study. 

  ALEXA data sources 

  While almost is well known, but they must emphasize that, ALEXA statistical data from their ALEXA toolbar users visit the web site information gathering, that is to say, only when a Web site is "to install and use the tools of the people" visit, AELXA will know, otherwise, a person in their own home to visit a site, how could AELXA know? 

  Obviously, if a person had not been installed tool, but not installed or use (such as the closure of the temporary use of his or other browser), then his visit any sites are not ALEXA know. 

  In the following text, I would like to "install and use ALEXA tool of the people" as "the tools of the user." 

  Understanding "of millions of people each hits (Reach per million users)" 
  We first give an example of this, and B duo, the two companies to work A: We have 100 individual companies, I understand, B: We have 200 companies know me.    Then, who in their own companies were more?    No way judgement, because we do not know their companies have a total of how many people, if a company a total of 100 individuals, companies and B 500 individuals, then it is clearly a company in its own name more point (all People who know him). 

  ALEXA is the same, because every day the number of users of the tool is not the same as (or unloaded every day of the installation, the installation tool of the people do not have the time to the Internet), so if ALEXA indicated that the actual figure is not conducive to click The contrast between the two days. 

  For example, yesterday there are 20 tools of the user visits your site, today or 20, you may think that your site's popularity is no change, in fact, is not, because this two-day total of the tools users is different, and if today is Thursday Yesterday was Friday, today the total increase in the number of Internet a lot, and you visit the website of the people has not increased, and this means you are losing. 

  So ALEXA introduction of a "user per million hits" more scientific parameters to measure. 

  For example, today, a total of 500,000 active users of the tool, including 50 visited your site, that is one in ten thousand people have visited your site, then if there are 1 million personal, it should have access to your personal 100 The Web site, so ALEXA will show you the number of hits millions of users is 100.    If today there are 2 million active users of the tool, which you visited 50 sites, calculated the same way, the millions of users who click on the number should be 25. 

  So every day how many ALEXA active users?    I passed on to you my statistics "is installed ALEXA toolbar" functional analysis, can be inferred from all over the world every day there are about 60 to a total of 800,000 active use ALEXA toolbar users.    Because ALEXA displayed hits millions of users is often "visit our website ALEXA actual users of the tool," 1.3 to 1.5 times. 

  I would like to you statistics users can try contrast to its own statistics and AELXA of millions of users click on the ratio between the number of relationship is not the case. 

  But attention here may be some deviation, the reasons for that: 
  ① I would like to you know that statistics can only visitors of the tools have been installed, and he could not judge whether or not to use the tools; 
  ② change with a user IP on your site, you would be I would like to statistics count twice, but it will be twice ALEXA count is still unknown, it is difficult to say in the installation tool of the tools of their time will generate an independent marking; 
  ③ network between China and the United States will affect the degree of smoothness of the leakage ALEXA statistics; 
  ④ daily tools of the total number of users are changing, we can not know that this figure. 

  Understand the basis for the website ranking 

  AELXA ranking in accordance with the "millions of users click on a few" and the "per user Pageview" comprehensive evaluation by the two data, but also the proportion of no way of judging.    But at least can know: 

  ① equal to the number of millions of users click on the two sites whose pages attractiveness of the category in the front row; 
  ② per user Pageview equivalent of two websites, the millions of users who click on a few, will come in the front row; 
  ③ these two parameters is the two, and their ranking is considerable. 

  Data analysis AELXA 

  For example, we have it.    Chelseacn.com Yesterday, the director told me that he suspected someone helped him brush rankings, said he was 100 IP, on top of more than 100,000.    So I read and watch his ALEXA data: 

  Millions of users click: 5.5 
  According ago, we said that the ratio can be judged daily visit his website "tool of the user" should be about 34.    He IP site about 100 a day from 100 individuals, 34 of the tool installed just normal people.    This data is correct. 

  Per user through the pages: 11 
  I read about, chelseacn.com is an information station, and the content is, the people who visit this site may be very interested in the corresponding content, they see a 89 articles and columns with Home Page like, a person can have a normal 11 months.    This data is not wrong. 
  Therefore, no one helped him brush rankings.    In other words, as long as there are three or four people per day to your website and browse 11 per person, you can also ranking as he did reach 100,000. 

  Read ALEXA attention matters we already know, the tools of daily active users only about 800,000, and every day is the actual difference between the number of Internet great.    And a small number of installed users of the tool is a lot of distribution in the English voice of the state.    In China a few of the tools users, Webmaster also occupy a large proportion.    Uneven distribution of minorities and so can be said that they do not have representation, and sometimes simple look at the rankings is meaningless. 

  When you need to ALEXA based on the data, it may be necessary to pay attention to: 

  ① top 30,000 after the Web site should not be compared. 
  The ranking of these Web sites is actually a few installation tool of the people affected, great chance. 

  ② between different types of sites should not be compared. 
  Their customer base may not be the same, these tools of the user to install the ratio will be different.    We can imagine 10,000 IP is the same site, providing free space website and the website of planting flowers stresses the rankings may be worlds apart. 

  ③ contrast not only ranking. 
  If you want to put in advertising, you should be more concerned about the millions of users who click on a few because the parameters reflected more the number of site visits.    If you want to evaluate the popularity of the content of the website, you should be more concerned about a few pages per user visit, always attractive website allows users to surf many pages. 

  ④ attention should be paid to a week of data and the data today. 
  ALEXA write directly in the results page (OverView page), that figure is a three-month comprehensive ranking Some websites have developed rapidly, such as the recent one to two months, with another three months of data and comparison site is not reasonable. 

  ⑤ look at the trend of flow map. 
  Flow ups and downs of the site is likely to rely on the flow of advertising, advertising will be lost once the flow.    Of course, advertising is not necessarily brought about, we can at least come to this site flow instability conclusions. 

  Postscript 

  Many people think that my data is cheating ALEXA out, and sometimes the more people it will explain my feeling in the excuse, really Huomaohuohei flavor.    Although I do not value my ALEXA data, but I really do not want to be put on the hat of data cheating, I always want other people's minds, there is a real me, and I am not being distorted.    But I hate myself, did not have time to take back that has been Ruguangbaoan, good to go back with them to make it Clear that…… 

  Has finally do so today.