Experiment with Habr and Alexa. The truth is somewhere near!
Update : Habr won. Video Habr statistics here thehomepage.ru/job/post/iwanttobelieve
Disclaimer: this post is not a stone in the garden of Habr and its users (I myself am), this is an attempt to understand how Alexa works.
Precedent
Today I looked at Alexa.com on my project and compared it with HabraHabrom. The question arose, why such a small difference?
')
In Habr's statistics , more than 80,000 unique users per day are indicated, however with the addition: Data for graphs are imported from the Google Analytics system, as well as from the internal statistics system ..
In my project, everything ranges from 3000 to one day. The difference with Habr in visits is more than 20 times.
There are three explanations:
1. Alexa considers only foreign traffic. Toolwi.com has 50% of Russian traffic, another 10% of Ukraine and 20% of the CIS countries. Habr has exactly the same picture.
2. Everyone reads habr feedburner. Then it is no longer unique on the site and they should not be displayed in the statistics.
3. Everyone reads habr via iphone. watch point 2
Clean experiment
I open the websites of the Runet by liveinternet counter
I choose websites with an average attendance of about 100 thousand, 60 thousand, and 20 thousand from this section www.liveinternet.ru/rating/ru/internet with open statistics.
So the selected sites:
diary.ru - consistently 100.000+ unique per day. 70% Russian traffic. 2 million page views per day.
habrahabr.ru - consistently 80,000 unique users per day and 250 thousand page views. 70% Russian traffic
game-torrent.info - fluctuations within 50-60 thousand unik per day. 70% Russian traffic. Violent fluctuations, but 500,000 views on average per day.
abbyyonline.ru - not very stable 38.000 per day. 45-50% of Russian, 10-20% of Ukraine, 7% of the United States. From 200 thousand to 400 thousand views per day.
What happened:
Reach - the characteristic is not quite clear, but the projects are based on attendance, except for the habr.
Here game-torrent.info is crossed by abbyyonline.ru. All because of the apparent strong fluctuations in views and traffic. Habr again out of place.
And finally the views. The picture is not clear, all spoils diary.ru with its two million views. We remove it.
Now everything became clear. abbyyonline.ru ahead of game-torrent.info! This is evident from the statistics for today at abby a sharp jump down, at the game torrents a sharp drop. Yesterday, abby has 469 thousand views, game 482 thousand. Everything fits together perfectly! (for reference www.liveinternet.ru/stat/game-torrent.info and www.liveinternet.ru/stat/abbyyonline.ru )
Now add toolwi.com with an average of 5,000 views per day. no comments
Disclaimer: this post is not a stone in the garden of Habr and its users (I myself am), this is an attempt to understand how Alexa works.
Precedent
Today I looked at Alexa.com on my project and compared it with HabraHabrom. The question arose, why such a small difference?
')
In Habr's statistics , more than 80,000 unique users per day are indicated, however with the addition: Data for graphs are imported from the Google Analytics system, as well as from the internal statistics system ..
In my project, everything ranges from 3000 to one day. The difference with Habr in visits is more than 20 times.
There are three explanations:
1. Alexa considers only foreign traffic. Toolwi.com has 50% of Russian traffic, another 10% of Ukraine and 20% of the CIS countries. Habr has exactly the same picture.
2. Everyone reads habr feedburner. Then it is no longer unique on the site and they should not be displayed in the statistics.
3. Everyone reads habr via iphone. watch point 2
Clean experiment
I open the websites of the Runet by liveinternet counter
I choose websites with an average attendance of about 100 thousand, 60 thousand, and 20 thousand from this section www.liveinternet.ru/rating/ru/internet with open statistics.
So the selected sites:
diary.ru - consistently 100.000+ unique per day. 70% Russian traffic. 2 million page views per day.
habrahabr.ru - consistently 80,000 unique users per day and 250 thousand page views. 70% Russian traffic
game-torrent.info - fluctuations within 50-60 thousand unik per day. 70% Russian traffic. Violent fluctuations, but 500,000 views on average per day.
abbyyonline.ru - not very stable 38.000 per day. 45-50% of Russian, 10-20% of Ukraine, 7% of the United States. From 200 thousand to 400 thousand views per day.
What happened:
Reach - the characteristic is not quite clear, but the projects are based on attendance, except for the habr.
Here game-torrent.info is crossed by abbyyonline.ru. All because of the apparent strong fluctuations in views and traffic. Habr again out of place.
And finally the views. The picture is not clear, all spoils diary.ru with its two million views. We remove it.
Now everything became clear. abbyyonline.ru ahead of game-torrent.info! This is evident from the statistics for today at abby a sharp jump down, at the game torrents a sharp drop. Yesterday, abby has 469 thousand views, game 482 thousand. Everything fits together perfectly! (for reference www.liveinternet.ru/stat/game-torrent.info and www.liveinternet.ru/stat/abbyyonline.ru )
Now add toolwi.com with an average of 5,000 views per day. no comments
Source: https://habr.com/ru/post/51692/
All Articles