At the end of July, I set up both Fathom and Netlify Analytics for this site, because I wanted to develop a sense of how big my ongoing audience is (setting aside big spikes from things like being on the first page of Hacker News for some article or another). I chose these two because they are both totally privacy-preserving. Fathom runs as JS in the browser; Netlify runs on Netlify’s servers; but both track nothing about you — only the pages which got visited and what the source was if that information was sent along.

The question I wanted to answer by setting up both Fathom and Netlify Analytics was straightforward: How much fall-off is there in using client vs. server-side analytics here? I expected there to be some difference between the two, because my writing skews to the technical and thus my audience to the kinds of folks who are likely to have a content-blocker installed. What I did not expect was just how large the difference would be.

Netlify measures five times as much traffic as Fathom does. Put another way: Fathom sees only 20% of what Netlify sees coming to my site — it is blind to 4 out of every 5 page views and visitors. (That holds equally for high-traffic days and normal-traffic days, and for high- and low-traffic pages.) My audience does not skew toward tech-savvy types with content blockers installed: it is dominated by them.

Your numbers likely look different from mine unless you have a similarly tech-dominant site. But you should check — especially if (unlike me) you actually make your living in any way directly related to the volume of traffic to your site.