2
2 Comments

I built an analytics tool because I stopped trusting my own dashboard

Quick backstory: I ship websites and products for a living, and at some point I realized I didn't believe my own analytics anymore.

Two things broke my trust.

First, bots. Modern bots run real Chrome and rotate residential IPs, so they show up as normal visitors. I'd see a traffic spike, get excited, dig in, and find nothing behind it. No conversions, weird pages, 3 AM sessions. Once you notice it, you see it everywhere, and you start wondering how much of your "growth" is scrapers.

Second, traffic tells you almost nothing about money. My best-performing page by pageviews was worth less than a boring comparison page with a fraction of the views, because the boring page actually converted. No mainstream tool would show me that in one place.

So I built Kobbe (kobbe.io). Privacy-first analytics where traffic, conversions, and revenue live in one dashboard.

The two things it does differently:

Bot filtering in layers: user-agent checks, verified-bot detection at the edge, datacenter ASN checks, and behavioral signals. Headless browsers and scrapers get dropped before they hit your stats or your quota. VPN users stay counted.

Revenue attribution through checkout: Stripe, Polar, Paddle, Shopify, and others. Every page, source, and campaign shows what it actually earned, refunds subtracted. Traffic is a vanity metric; revenue per page changed how I think about content completely.

It's cookieless with no consent banner needed, honors Global Privacy Control, and stores no raw IPs.

On pricing I made one decision I'm stubborn about: every feature on every plan. No "upgrade to unlock funnels" games. You pick an event volume and that's it. Starts at $9/mo for 100K events, 15-day trial.

There's a live demo with realistic data and no signup wall at app.kobbe.io/demo/kobbe-studio, because I think you should be able to judge an analytics tool before creating an account.

Happy to answer anything about the build or the decisions. And if you look at it and something feels off, tell me, that's why I'm posting here.

on September 5, 2026
  1. 1

    This hits home. Our €150 of Google Ads looked fine in the clicks column and produced exactly 0 installs, while two weeks of quiet replies in live discussions brought the first real user. Same lesson as your boring comparison page: volume flatters, intent converts. The 3 AM session detail is the tell nobody warns you about. One question: how do you attribute a subscription that lands three weeks after the first visit β€” last touch, or does revenue keep accruing to the original source?

  2. 1

    The two differentiators solve pretty different trust problems: filtering fake traffic versus showing which pages actually make money. Have early users shown one of those is the reason they’d switch from their current analytics, or are both still hypotheses?