Skip to main content
D The Drew Archive One house, many rooms
  1. The Terminal/

Adding Analytics Without Becoming the Product

Why I chose small, self-hosted, privacy-respecting analytics for The Drew Archive.

How AI was used

Drew worked with ChatGPT and OpenClaw to add privacy-respecting analytics to The Drew Archive. After the work was complete, he asked ChatGPT to write a blog post about what they had accomplished. Drew found the initial draft too robotic, so he asked Claude Sonnet to preserve the content while making it more readable. Drew reviewed the finished article before publication, and OpenClaw prepared and deployed it through The Drew Archive’s publication workflow.

Systems involved: ChatGPT, OpenClaw, Claude Sonnet

About the disclosure levels

RealmThe Terminal

TrailTechnology in Service of Independence

Reading modeEssay

Table of Contents

For a long time I avoided putting analytics on The Drew Archive. Not because I didn’t want to know whether anyone was reading, but because I didn’t want to build a site around collecting data on its visitors.

There’s an irony to much of the modern web: the tools built to help you understand your audience often end up knowing far more about that audience than you ever meant to collect. That never sat particularly well with me.

What I Actually Want to Know
#

I don’t need detailed behavioral profiles. I don’t need advertising identifiers. I don’t need to know what someone did before arriving here or where they went afterward. Mostly, I’m curious about a handful of simple questions:

Which articles are people reading? How did they find the site? What topics seem to resonate? Is the structure of the Archive helping people discover related ideas?

Those questions help me build a better archive. They don’t require turning readers into datasets.

Choosing GoatCounter
#

After auditing the server and looking at a few self-hosted analytics platforms, I settled on GoatCounter. It’s intentionally small — it runs on my own server, stores its own data, and doesn’t need cookies or persistent visitor profiles.

It also respects browser privacy preferences like Do Not Track and Global Privacy Control. If someone has explicitly asked not to be tracked, the site simply honors that. That feels like the right default.

Keeping Things Simple
#

The setup stays intentionally modest: no custom events, no heat maps, no session replay, no advertising integrations, no attempt to follow visitors around the internet. Just enough to understand whether the Archive is becoming easier to navigate and more useful over time.

I suspect I’ll glance at the dashboard every so often, but I don’t want to spend my days watching graphs.

The goal isn’t to chase traffic. The goal is to build something worth returning to.

Where to go next

Keep moving through the archive

This piece works best as part of a larger path. Pick the next doorway that matches how you want to keep reading.

Follow the trail Technology in Service of Independence Move through the wider sequence this article belongs to and keep the thread intact. Stay in this realm The Terminal Browse more writing from this section if you want to keep exploring the same part of the archive. Read the adjacent piece Hardening The Drew Archive How I upgraded, hardened, and built a tested disaster recovery system for The Drew Archive.

Keep exploring

Explore beyond this room

These are the closest nearby pieces in the archive, with extra weight given to the same realm and trail when possible.

The Terminal Essay

SearXNG

Why Drew self-hosted SearXNG as a private search engine instead of relying on proprietary search providers.

The Terminal Essay

Comments Are Now Live on The Drew Archive

Why I added a self-hosted Remark42 discussion section to The Drew Archive, and how email login, Resend, Hugo, and a little production hardening brought it together.


Conversation

Comments

Join the discussion. Be thoughtful, stay on topic, and treat other readers as neighbors.