Hypertune
Type-safe feature flags platform with A/B testing, analytics, and app configuration optimized for TypeScript
GrowthBook
Open Source Feature Flagging and A/B Testing Platform
Self-hosted or cloud platform for feature flags, gradual rollouts, and powerful A/B test analysis. Integrates with your existing data stack—BigQuery, Mixpanel, Redshift, and more.
Overview
Bring Enterprise Experimentation In-House
GrowthBook delivers the feature flagging and A/B testing capabilities that top-tier companies build internally, without the engineering overhead. Designed for product teams and data scientists, it combines advanced targeting, gradual rollouts, and statistically rigorous experiment analysis in a single platform.
Flexible Deployment, Powerful Analytics
Deploy via Docker Compose in under a minute or use managed cloud hosting. GrowthBook connects directly to your existing data warehouse—BigQuery, Redshift, Snowflake, Mixpanel, Google Analytics, and ClickHouse—so you retain full control over your data. SDKs for React, JavaScript, PHP, Ruby, Python, Go, Android, and iOS enable fast integration across web and mobile.
Built for Rigorous Testing
Go beyond basic metrics with CUPED variance reduction, sequential testing, Bayesian statistics, and SRM checks. Drill down by browser, country, or custom attributes, document experiments with Markdown and screenshots, and export results as Jupyter Notebooks for deeper analysis. Webhooks and a REST API support custom workflows and integrations.
Highlights
Advanced A/B test statistics: CUPED, sequential testing, Bayesian analysis, and SRM checks
Native integrations with BigQuery, Redshift, Snowflake, Mixpanel, and Google Analytics
Feature flags with gradual rollouts, advanced targeting, and experiment capabilities
Multi-language SDKs for React, JavaScript, PHP, Ruby, Python, Go, Android, and iOS
Pros
- Self-hosted deployment retains full data ownership and control
- Statistically rigorous experiment analysis rivals enterprise platforms
- Connects to existing data warehouses without ETL overhead
- Active community and responsive support via Slack and email
Considerations
- Self-hosting requires Docker and MongoDB infrastructure management
- Advanced statistical features may require data science expertise to interpret
- Smaller ecosystem compared to established SaaS vendors
- Documentation assumes familiarity with experimentation concepts
Managed products teams compare with
When teams consider GrowthBook, these hosted platforms usually appear on the same shortlist.
Looking for a hosted option? These are the services engineering teams benchmark against before choosing open source.
Fit guide
Great for
- Engineering teams wanting self-hosted feature flag infrastructure
- Data-driven organizations with existing warehouse investments
- Companies requiring advanced statistical rigor in A/B tests
- Teams seeking cost-effective alternatives to enterprise SaaS tools
Not ideal when
- Non-technical teams without engineering or data science resources
- Organizations lacking data warehouse or analytics infrastructure
- Teams needing white-glove onboarding and managed services
- Projects requiring immediate production deployment without setup time
How teams use it
Gradual Feature Rollout
Release new features to 5% of users, monitor metrics in real-time, and scale to 100% with confidence using built-in targeting rules.
Revenue-Impacting A/B Tests
Run pricing experiments with CUPED variance reduction and sequential testing to detect wins faster while controlling false positives.
Multi-Platform Experimentation
Coordinate feature flags across web, iOS, and Android apps using consistent SDKs and centralized experiment configuration.
Data Warehouse Integration
Analyze experiment results directly in BigQuery or Redshift, export to Jupyter Notebooks, and share findings with stakeholders.
Tech snapshot
TypeScript96%
Python2%
SCSS1%
JavaScript1%
Jinja1%
Handlebars1%
Frequently asked questions
How long does self-hosted setup take?
Docker Compose deployment completes in under a minute. You'll need to configure data source connections and SDK integration separately.
What data warehouses are supported?
GrowthBook integrates with BigQuery, Redshift, Snowflake, ClickHouse, Mixpanel, and Google Analytics. Custom SQL queries enable other sources.
Do I need a data scientist to use advanced statistics?
Basic A/B tests work out of the box. CUPED, Bayesian analysis, and sequential testing benefit from statistical knowledge but include sensible defaults.
Can I migrate from LaunchDarkly or Optimizely?
Yes. GrowthBook supports feature flag imports and provides SDKs with similar APIs. Migration complexity depends on your existing implementation.
Is there a managed cloud option?
GrowthBook offers managed cloud hosting alongside the self-hosted option. Both share the same feature set and SDK compatibility.
Project at a glance
Active- Stars
- 7,262
- Watchers
- 7,262
- Forks
- 660
Repo age4 years old
Last commit3 hours ago
Self-hostingSupported
Primary languageTypeScript
Last synced 3 hours ago