How it started
I started building Helion in 2024 because I kept running into the same problem: analytics is the first thing you instrument and the last thing you want to think about — until the bill arrives, or until the vendor changes their pricing, or until a lawyer asks where your users' data is stored and you do not have a good answer.
I had been using Mixpanel for side projects and noticed the pattern every developer eventually hits: at low volume, it is free; at meaningful volume, the cost becomes the thing you are optimizing around rather than your product. More importantly, I wanted to understand my users, not just count them — and I wanted that data to live in infrastructure I controlled.
So I built it.
What I wanted Helion to be
The goal was not to clone an existing analytics tool. It was to build the analytics layer I would actually want on a product I care about:
- Self-hosted first — your ClickHouse instance, your data, your queries
- Cookieless by default — privacy compliance without requiring a consent banner
- Both web and product analytics in one deployment — no juggling Plausible for pages and Mixpanel for events
- Open source under AGPL — so the code you run is the code you can audit, fork, and improve
- MCP-ready — because event data is training signal in the AI era, not just a dashboard metric
What shipped in 2024
The first version of Helion was a working end-to-end analytics pipeline: SDK, API, ClickHouse storage, and a dashboard. From there, the year was about filling in the gaps between "works for my projects" and "works for others."
Key milestones:
- Core event tracking, user profiles, and session management
- Funnel and retention analytics
- Self-hosting via Docker Compose with a single setup script
- SDKs for JavaScript, Next.js, Astro, Nuxt, Express, and React Native under the
@helionlabs/*scope - Cookieless tracking with daily-rotating device identifiers
- Export API for raw event access
- Revenue tracking alongside product events
- MCP server integration (query your analytics data directly from Claude or Cursor)
Roadmap for 2025
The foundation is solid. What comes next is depth:
- Warehouse connector — sync events to BigQuery, Snowflake, and Redshift
- Session replay improvements
- SDK publish pipeline to npm (
@helionlabs/*) - More notification and alerting capabilities
- Additional integrations (Google Search Console, Slack, webhooks)
- Performance optimizations for high-volume self-hosted deployments
A note on open source
Helion is licensed under AGPL-3.0. I decided early that the code should be fully public — not just open-core, not just a public repo with most features behind a paywall. If you want to run Helion on your own infrastructure, you can. If you want to contribute, the codebase is there. If you want to verify what the tracking script does, you can read every line.
The project is at github.com/Shrotriya-lalit/helionlabs. Feedback, issues, and contributions are welcome.
If you have not tried Helion yet, the fastest path is the cloud or the self-hosting guide. Either way, reach out on Discord if you run into anything.
