  |
  |
I have received a lot of messages, and now I see some blogs about the topics, on why there are differences in the amount of hits on the blog and stats tab.
I am going to explain this once again and then create a posting in the support forum which will be linked to this blog, basically since I cannot remember what I called the previous blog explaining this :).
Hits on the Author Accounts
Our policy have always been to only pay actual hits and not those created by robot crawlers and other programs that aren’t managed by a physical user.
To determine an actual user we look at a few things that need to be present:
a) a session cookie
b) Browser identification
c) session lifetime
Session cookies
The session cookies are created so that we can see whether you have visited our site or not. We store a cookie called “mdl” in your browser cache to identify whether you have viewed an article or not.
Browser identification
If you are not using a browser to access the site and view articles then we do not count it as an actual user.
In our code we determine whether your browser is:
a) Opera
b) Firefox
c) MS Internet Explorer
d) Safari
e) Chrome
based, there are a few others we also check for.
Session life time
The cookie and session works together to determine whether a hit occurs by a specific user with in a 24 hour time frame.
So how does this all fit together?
If one of these 3 conditions aren’t met then we do not record a visit on the page as a hit. We follow the following procedure:
1) Is the user using a recognised browser YES
2) Is the mdl cookie set YES
3) Is the cookie present in the session YES
Now we say cool, this is an actual person viewing the page, record one hit and allocate it to the blogger. If you return to the page on the same day then we already have you logged as viewing the page and we do not record a hit, (Google works the same on unique hits).
If one of those conditions are not met then we do not record a hit.
Why?
In the past we have had a few users who tried to by pass the system and cheat their way into more hits. This affected not only our blog hits but also the hits on the sponsored ads running on the site.
So by applying this method that is used in general through out the industry, we are only paying for Unique Views and not on overall impressions. Mainly because overall impression can be manipulated.
Hope this helps all the authors who had questions.
Subscribe to this comment's feed
written by The Source, June 22, 2010
-
- +0
-
-
written by barrmar, June 22, 2010
I just have three questions/observations:
the first is that the hits count has been removed for individual posts - It is very useful to be able to see the hits per post - i.e. to which posts get more views etc.
the second observation is that return visits to a post no longer count. If there are many comments, this often attracts return visits, further comments etc. These always counted in the past - there was no stipulation of unique hits.
The third is a question - how certain are you that the filter works correctly? I find it very hard to accept that the majority of hits on my posts are from web crawlers or spiders. Of 1401 hits (on the posts themselves before these states were removed) totalled 1401 for June, the Stats page about 500. I would have expected perhaps 5% - a spider won't go to the same post more than once as far as I understand.
A fourth point - some of us look at MyDL from our cell-phones. I have read many posts on my Blackberry (not very satisfactory but manageable. From your explanation above it seems that your system will exclude these hits although they are in fact genuine hits, and mobile access to all sites is becoming more common by the day!
-
- +0
-
-
written by The Source, June 23, 2010
I am working on the hit count meter for the posting, I removed it for now to stop the confusion until it reflects the unique views.
We have tested the system to ensure that it works the same way it did before. It always worked on unique views, we just never explained it in detail as with this posting. Mobile browsers are working on the same principles so that should now influence the lower hits. Mobile browsers also store cookies and create sessions. By default they also send a browser identification which we pick up.
Nevertheless we are constantly monitoring the system to improve it.
-
- +1
-
-
Some people delete all cookies at end of session. I presume this does not affect anything.
Obviously bots and web-crawlers are less discerning than the average user - they are the only ones who read my blogs!