You’ve been there. You launch a side project, you want to see if anyone actually cares, so you drop in the Google Analytics snippet. Within minutes, you’re getting data—but you’ve also become a willing cog in the surveillance capitalism machine.
You don’t need a privacy-friendly tracker. You need to stop tracking.
Recently, a developer named Gary asked Hacker News what people use for privacy-focused analytics. He wanted the basics: visitor counts, geo, referrals. The responses were the usual suspects: Plausible, Rybbit, Fathom. “Cookieless” became the buzzword of the day. But here’s the dirty secret about “cookieless” tracking: it still requires a client-side script, and it still often collects IP addresses or browser fingerprints.
We’ve been conditioned to believe that to understand our site, we must inject code into our visitors’ browsers. It’s a lie sold by an industry that monetizes data collection.
The modern web treats your visitors as data crops to be harvested, not guests to be respected.
Here is the twist: the best privacy-focused analytics tool doesn’t exist as a shiny new SaaS product. It’s already running on your server. Back in the day, we used tools like Webalyzer. They didn’t need JavaScript. They didn’t need cookies. They just read the raw web server logs.
Every time a browser requests a page, your server logs it. The IP, the referrer, the requested file. It’s all there. When you use a server log analyzer, you aren’t adding a single byte of tracking code to your user’s browser. You aren’t slowing down your page load. You’re just looking at the receipts your server already generates.
If you want true privacy, you don’t need a “better” tracker. You need to redefine what “enough data” means.
True privacy isn’t a feature you buy; it’s the data you choose not to collect.
We need to stop asking “which tracker is less evil?” and start asking “do we even need a tracker?” For 99% of side projects and small sites, server logs give you everything you need. You get your geo data, your referrers, your page views. You just don’t get to follow a single user around the internet like a stalker.
Next time you want to measure your site’s growth, skip the SaaS signup. Tell your server to parse its own logs. Build trust by refusing to collect data you don’t need. Your visitors will never know, but that’s exactly the point.
FAQ
Q: Aren't server log analyzers outdated and hard to set up?
A: They are older, but 'hard to set up' is a myth sold by SaaS companies. Running a log analyzer on your own server takes a few minutes and requires zero client-side code.
Q: What's the practical implication of using server logs?
A: You get the exact high-level data you need—page views, referrers, basic geo—without slowing down your site's load speed or adding a privacy policy disclaimer for tracking cookies.
Q: What's the contrarian take on 'cookieless' analytics?
A: Most 'privacy-focused' analytics tools are just surveillance with better PR. If you're injecting a script into a user's browser to collect their data, you're still tracking them.