With Google sunsetting Universal Analytics and notice that my historical data would be deleted in 2023, now was the opportune time to choose a new solution. In this post, I will cover some alternatives and a study of how they performed against GA.
I started using Google Analytics (GA) in 2011 and at the time it was a relatively simple product. Setup was a breeze and it was easy to find the data I needed. Today, GA has grown into an enterprise solution that requires some training to extract the insights I need. For basic websites, such as this one, it is more than I really need.
With the proliferation of analytics solutions in the last few years, there are a number of great alternatives to GA. For my use case, I was looking for a solution with the following:
- Simple Dashboard
- Infinite Data Retention
- GA Data Import
- Proxy or Back-end Support
Researching against these requirements, Plausible, Pirsch, and Fathom emerged as viable solutions. Moving forward with Plausible and Pirsch, I ran a 25 day study to compare the results against GA. At the time of writing, Fathom’s trial is 7 days which is too short for statistical relevance. To minimize implementation discrepancies, each solution was installed as an external JavaScript file as recommended by their publishers.
Visitor Aggregate
+ 8 | |||||||
+ 7 | |||||||
+ 6 | |||||||
+ 5 | |||||||
+ 4 | |||||||
+ 3 | |||||||
+ 2 | |||||||
+ 1 | |||||||
GA3 | |||||||
GA4 | Pirsch | Plausible |
Page View Aggregate
+ 13 | |||||||
+ 12 | |||||||
+ 11 | |||||||
+ 10 | |||||||
+ 9 | |||||||
+ 8 | |||||||
+ 7 | |||||||
+ 6 | |||||||
+ 5 | |||||||
+ 4 | |||||||
+ 3 | |||||||
+ 2 | |||||||
+ 1 | |||||||
GA3 | |||||||
– 1 | |||||||
GA4 | Pirsch | Plausible |
With Universal Analytics (GA3) as the baseline for comparison, I was surprised at how close the results were. In my experience it is not uncommon for GA to get caught in ad-blockers or a spam bot to skew the data. With this result, the deltas come down to how the technology in each solution logs visitors. Without analytics being captured server side, it is hard to determine which one is the most accurate.
Conclusion
With the results of the study this close, the deciding factor came down to developer support and product experience. I decided to move forward with Pirsch for the well designed dashboard, PHP SDK, and GA data import. As of writing Plausible and Fathom have a GA data import feature planned for a future release.
Leave a Reply