Why We Built Our Own Analytics Instead of Just Using Google's
By Anested Team
Let's get this out of the way: Google Analytics is a fantastic tool. It handles traffic, sessions, referral sources — the big-picture stuff. We still use it. But somewhere around the third client project, we realized we needed something more granular. We wanted to know which specific component on a page drove the most clicks. We wanted to track scroll depth per section, not just per page. And we wanted to own the raw data.
What off-the-shelf analytics miss
Most analytics tools are designed for marketers who care about traffic and conversions at the page level. And that's fine for many cases. But if you're building intelligent infrastructure — systems that learn and improve — you need much deeper signal. You need to know that the pricing section on a landing page gets scrolled past by 68% of visitors, or that a specific CTA button gets 4× more clicks when it's above the fold.
What our in-house system captures
- Total and unique page visits, with proper session handling
- Per-component interaction events — clicks, hovers, scroll-into-view
- Area and locality derived from IP geolocation
- Exact coordinates only when the user explicitly grants permission
The architecture behind it
We store tracking events in MongoDB as a flexible event stream. Each event carries a timestamp, session ID, component identifier, and action type. We aggregate on demand using pipelines — daily uniques, weekly trends, heatmap-style interaction density. Running this alongside Google Analytics gives us the convenience of a hosted tool and the depth of a system we own.
The best part? Because it's built into the infrastructure, there's no extra integration step for new projects. Every site we ship comes with this baked in from the start.