I’m presenting Piwik to a company (https://catracalivre.com.br) wondering to replace Google Analytics (GA), but when trying a proof-of-concept, I’m getting very different results comparing numbers on Piwik and GA. I would like some tips to diagnose that.
Currently, the real time users counting on Piwik is something like 2100 users and in GA is 5000 users. It is too different and I’m trying to understand the reason.
Other scripts can break Piwik script on the footer? The way Piwik count users is different of the way GA count users?
Piwik is running on a cluster of 3 servers on Azure, each one with 4 cores and 8Gb RAM. I’m using Percona XtraDB Cluster and Redis, but I tried without Redis too. No errors on servers logs and the load average is very Ok.
I have never used GA, so I don’t know which vistors count to live visitors. If GA e.g. uses visitors in the last 15 minutes and Piwik in the last 3 minutes, it could explain the difference.
Piwik is configured quite privacy aware by default. Users who opted out of the tracking or have enabled do-not-track in their browser settings are excluded by default.
If you search around on the forum there are some more posts about why Piwik shows less visitors than GA.
Well, GA counts real time users in the last 1 minute, so the average when counting the last 1 or 3 minutes should be not so different, I guess. Also, I had disabled privacy options to check if this could be the reason, but no luck.
I will dig in forum to check if there is any clue about this.
!