growthbook  by growthbook

Open-source platform for feature flagging, experimentation, and product analytics

Created 4 years ago
7,351 stars

Top 7.0% on SourcePulse

GitHubView on GitHub
Project Summary

Open Source Feature Flags, Experimentation, and Product Analytics

GrowthBook is an open-source platform for feature flagging, A/B testing, and product analytics, aiming to provide enterprise-grade capabilities without the cost of SaaS or the effort of in-house development. It empowers companies to implement data-driven product development and experimentation workflows.

How It Works

The platform offers feature flags with advanced targeting and gradual rollouts, alongside a sophisticated experiment stats engine supporting methodologies like CUPED and Bayesian analysis. A key differentiator is its "Warehouse Native" architecture, enabling direct querying of data sources such as BigQuery, Snowflake, and Databricks. Metric definitions are flexible and SQL-backed, allowing for complex calculations.

Quick Start & Requirements

Self-hosting is achieved via Docker Compose. Clone the repository, then run docker compose up -d. The application is accessible at http://localhost:3000. Detailed instructions are available in the official "Self-Hosting Instructions".

Highlighted Details

  • Feature flags with advanced targeting, gradual rollouts, and A/B experiments.
  • Support for 24 SDKs across major platforms (React, Python, Android, iOS).
  • Advanced experiment statistics engine (CUPED, Sequential, Bayesian, SRM checks).
  • Warehouse-native querying for 11 data sources (e.g., BigQuery, Snowflake, Databricks).
  • Flexible SQL-backed metric definitions.
  • Integrated product analytics suite.

Maintenance & Community

Contributions are welcomed via CONTRIBUTING.md. Community support is available through Slack, with email (hello@growthbook.io) as an alternative.

Licensing & Compatibility

GrowthBook is an "Open Core" product. The majority of the codebase is under the permissive MIT license. However, specific directories are governed by a separate "GrowthBook Enterprise License," which may impose restrictions on commercial use. Consult the LICENSE file for full details.

Limitations & Caveats

The "Open Core" model means certain advanced features are likely restricted to the GrowthBook Enterprise License, not covered by the permissive MIT license. This requires careful review to understand the scope of open-source availability versus proprietary or licensed components for adoption.

Health Check
Last Commit

17 hours ago

Responsiveness

Inactive

Pull Requests (30d)
180
Issues (30d)
21
Star History
85 stars in the last 30 days

Explore Similar Projects

Starred by Alex Graveley Alex Graveley(Creator of GitHub Copilot, Dropbox Paper, Mobilecoin, Hackpad), Gregor Zunic Gregor Zunic(Cofounder of Browser Use), and
4 more.

lmnr by lmnr-ai

0.7%
3k
Open-source platform for engineering AI products
Created 1 year ago
Updated 1 day ago
Feedback? Help us improve.