Compare Google Analytics, Rambler Top100 and Mail Counters

Finally, they got around to understanding the features of attendance counters, more precisely, the basic data that they give out. And more precisely - how the data of different counters correspond to each other.

Compared:
- Google Analytics
- Rambler Top100
- Mail


Technique
')
We take a regular corporate site with an average attendance (what a site, for obvious reasons, I do not indicate). We put GA, Rambler and Mail counters on it (for now we omit LiveInternet, since it is not often used on corporate websites). It is clear that the counters are on all pages of the site, that is, they are on an equal footing (the location in the page code is also the same).

We compare data only on attendance (visitors, hits), as the most accessible data even in the case of closed statistics.


results

The data collected for the period from February 26 to April 1 (5 weeks) is as follows.

Google Analytics :


Rambler Top100 :


Mail :



OK. We are trying to understand what is happening here.

The GA help gives a comprehensive answer:

• Absolutely unique visitors - how many real visitors (computers with a unique cookie) opened the site at a specified time interval.
• Visitors are absolutely unique visitors per day, summed up on specified days.
• Visits - the number of sessions (timeout not more than 30 minutes).
• Page views are hits, of course.

Help Rambler is not so verbose and gives a definition only for hits and hosts:

• Hits - the number of user hits on the site.
• Hosts - the number of unique IP addresses of computers from which the pages of the site were accessed.

Help Mail gives the following information:

• The visitor is a unique cookie. The same thing as an absolutely unique visitor at GA, only users with disabled cookies are not counted (4% according to global statistics HotLog).
• Hosts - the number of unique IP-addresses of computers from which the pages of the site were accessed for a specified period of time.
• New - the number of new (from the beginning of the calendar month) unique IP-addresses of computers from which the pages of the site were accessed. From this definition, by the way, it follows that if you look at the statistics by months, then “hosts” = “new”.
• Hit - he hit everywhere. Here, however, Mail indicates that the hits of some bots are not counted.

So, what data can be equated between the counters:

• Absolutely Unique Visitors (GA) = Visitors (Mail) . The experimental Mail data differs from the GA indicators by about 5% mainly downwards, which can be explained by failing users by Mail without cookies and robots.
• Hits (GA) = Hits (Mail) . In practice, again, however, with a slight amendment of –5%. Rambler for some reason gives 20-25% less hits - why ???

Judging by the figures, Visitors (Rambler) = Visitors (GA) with an amendment of –5-10%, that is, this is not the number of unique visitors for a specified period, but the sum of unique visitors by day. Generally speaking, this indicator is completely useless, since it does not reflect either the number of real visitors or the number of their visits.

From the useful data we can also mention GA visits , from which it is possible to determine the frequency of users entering the site. All other indicators (hosts Rambler and Mail, as well as New Mail) - absolutely useless thrash.


findings

• Google Analytics is driving :)
• Mail - more or less normally considers visitors and hits.
• Rambler - complete crap. Its data can only be used for relative comparison of site traffic.

There remains the question of how to convert the data of the Rambler into the desired data for unique visitors and hits. There are no other options, except how to build a proportion with indicators of another site (the same subject) with installed Rambler + Mail or GA counters ...

By the way, if someone can add a comparison with the LiveInternet counter to the topic, it will be very cool!