Introduction to VeloStats

Privacy-first, cookieless web analytics built for performance and simplicity.

Introduction to VeloStats

VeloStats is a modern, privacy-first web analytics platform designed to give site owners deep real-time visibility without compromising user privacy or web performance.

Why VeloStats?

Traditional analytics platforms rely on invasive tracking cookies, cross-site profiling, and heavy client-side bundles that slow down page loads and require annoying GDPR consent banners. VeloStats was designed from day one with a different philosophy:

  • 100% Cookieless: No cookies, no localStorage identifiers, and no persistent device fingerprints.
  • Zero GDPR/CCPA Banners Needed: Because personal identifiable data is never stored, you don’t need user consent banners for analytics compliance.
  • Featherweight Tracker: The tracking script (velo.js) is under 2KB, loads asynchronously, and has zero impact on your Core Web Vitals.
  • Bot & Crawler Isolation: Automated scrapers, headless browsers, and datacenter pings are automatically classified and filtered so your human traffic metrics remain clean.
  • Enterprise-Grade Infrastructure: Powered by ultra-fast TimescaleDB time-series and Redis edge aggregation for instantaneous dashboard loading.

Core Architectural Concepts

VeloStats tracks user journeys through anonymous daily session hashes:

  1. Daily Rolling Salt: Visitor identifiers are generated using a one-way HMAC-SHA256 hash that rolls every 24 hours. A visitor cannot be tracked across multiple days or across different websites.
  2. Four Lifecycle Events: Rather than firing separate tracking calls for every minor interaction, VeloStats captures the session lifecycle in four lightweight pings (pageview, load, heartbeat, and leave), automatically collapsing them into a single consolidated record.
  3. Real-Time Edge Aggregation: Real-time dashboards refresh with sub-second latency via TimescaleDB hypertables.

Next, follow our Quickstart Guide to install your first tracking snippet in under 30 seconds.