Part 3: Building My Own Privacy-First Analytics

I’ll fill this post during/after implementation. The blog itself will be the result — each post and list page will show the live analytics counts.

What I expect to learn:

• How much traffic my blog actually gets
• Which posts hold readers the longest
• What a simple, public analytics system feels like from both sides (publisher and reader)

Open questions going in:

• Does the hashed fingerprint produce reasonable unique counts, or does it over/under-count?
• Does scroll-depth tracking add meaningful signal, or is it noise?
• At what scale does the analytics service start mattering for my tunnel’s bandwidth?

If you’re reading this, the numbers below are real. Let’s see what they say.

---

Sources: (added during implementation)