Ejecting Google Analytics for Open Source Options

While aware of the holiday winding down from work (remember to participate in our end of the year Holiday Pulse poll), a technical issue is on my mind, even if it is something to tackle in the new year.

Like many others our OEGlobal web sites make use of Google Analytics to look at the trends and patterns of our site visitors. In looking to implement better GDPR compliance for cookie management (testing now on the OEG Voices site) one of the culprits is Google’s tracking that comes with the pretty charts and data.

We have put on our long list of technical todos to seek an open source web analytics tool, one that could be self managed so data is not sent elsewhere.

Somewhere in my link clicking this week, I stumbled on maybe one to consider

which does offer a self hosted version. I actually am not a real server admin who can suss out these platforms, but have brought this up for consideration-- as an open organization we ought to use as much open source tools as possible (one can always aim for the gold standard of the OER Foundation).

I did some across a listing of five open source web analytics platforms from Opensource.com

One was Plausible as well as Matamo (once known as Piwik I believe I saw in use at BCcampus??), plus Open Web Analytics (has general versions plus Mediawiki and WordPress plugins), something called Count.ly. It was interesting to see listed as well AWStats which generates reports from server log files (I remember using this on web servers like 15 years ago!)

I am curious and hope others can share direct experience with any of these platforms-- I see @DelmarLarsen responded when a posted a link to Plausible in that twitter platform people are talking about :wink:

Thanks for sharing any technical insights on open source alternatives to Google Analytics.

Thank you @cogdog!

If anyone has experience with PostHog, feedback is very welcome. PostHog is written in Django (a web framework we use heavily at OEGlobal) and is extremely feature-rich, incl. support for heatmaps, session recording (as video), etc… Its primary use is product analytics and I wonder how easy it is to set it up for web analytics, as a drop-in replacement for Google Analytics. Has anyone from the community used it? Are there any caveats to be aware of?

More Django-related packages for analytics:

I’m going to implement a Plausible.io instance next week to provide an independent alternative to our Matomo instance (which we’re not sure is working properly in every case)… this provides a means to verify consistency of analytics. Happy to let you use it, @cogdog, as well, when I’ve got it going!

1 Like

Thanks, Dave, I am keen to see if Plausible.io is more than plausible :wink:

FWIW a recent review

1 Like