Trusted by 3,000+ companies worldwide
Make rigorous experimentation easy for everyone
Experiment anywhere, your way
Customize metrics to your business
Apply rigorous statistical analysis
Built-in guardrails and workflows
Foster a culture of experimentation
Build more winners, ship fewer losers
Stop guessing which features matter. Use decision frameworks to know when to ship, when to rollback, and how long to test. Measure experimentation over time with meta analyses and insights.
Create trust through transparency
Query your data warehouse directly—Snowflake, BigQuery, Databricks, Redshift and more. Every query is visible through SQL. Every result is reproducible. Slice and dice metrics, add metrics retroactively to explore root causes.
Experimentation for everyone
Enable everyone to experiment like your best data scientist. No more waiting for engineering. No more ad hoc testing. Build a rigorous experimentation program anyone can deploy and understand.
GrowthBook Open-Source Platform
GrowthBook’s modular design works on top of what you have, or replaces what’s not working.
Integrates with your tech stack. Analyze your data where it lives. SQL visibility.
Same product, same features. On cloud or fully self-hosted.
No migration required. No changing tools. We work with your tech stack.
Connect our MCP server to Claude Code, Cursor, VS Code, etc.
Predictable pricing, flexible plans for every team
Explore free and tiered pricing and plans for both cloud and self-hosted deployments.
One platform to analyze, deploy, and measure
Feature Flags
Product Analytics
“We don’t need any code changes, we don’t need an app release. We just configure the new tests and launch right away.”
“Our goal was to consolidate everything into a single platform while saving money and ensuring compliance and security.”
“Being able to turn a feature on and off with a flip of a switch is fantastic... That’s so much easier than having to do a deploy or a roll-back.”
“Experimentation showed what customers actually do rather than what we assume they’ll do.”
“People only see the wins, but there’s actually greater value in avoiding losses. We’ve stopped changes that could have cost millions.”
"GrowthBook allowed us to uplevel our code, speed up decision-making, and focus on what we do best—building a world-class AI lending marketplace."
"The fact that we could retain ownership of our data was very, very important. Almost no solutions out there allow you to do that."
“GrowthBook lets us build experiments exactly how we want. The ability to target based on culture and geography, as granular as needed, is a major win for us.”
"The fact that GrowthBook offered us the ability to keep that data in-house was a key reason why we chose to work with them."
"The fact that GrowthBook offered us the ability to keep that data in-house was a key reason why we chose to work with them."
"The fact that we could retain ownership of our data was very, very important. Almost no solutions out there allow you to do that."
“GrowthBook lets us build experiments exactly how we want. The ability to target based on culture and geography, as granular as needed, is a major win for us.”
FAQs
“A/B testing" typically refers to simple two-variant tests. An experimentation platform goes further: multiple variants, advanced statistics, warehouse-native analysis, feature flag integration, holdouts for cumulative measurement, and tools to build an experimentation culture across your organization.
Look for: flexible experiment types (code-based and no-code), advanced statistics (CUPED, sequential testing), warehouse-native integration, transparent methodology you can verify, unlimited experiments without per-event pricing, and the ability to add metrics retroactively.
Warehouse-native means GrowthBook queries your data where it already lives (Snowflake, BigQuery, etc.) instead of copying it into our systems. Benefits: no data duplication costs, complete data ownership, use your existing metrics definitions, and total transparency into how results are calculated.
CUPED (Controlled-experiment Using Pre-experiment Data) reduces variance by accounting for pre-experiment user behavior. Less variance means you need fewer users to detect effects, so experiments reach statistical significance up to 2x faster.
Sequential testing lets you monitor experiments continuously and make valid decisions at any point—not just at a pre-set sample size. You can stop early when you have a clear winner, without inflating false positive rates.
Open source means transparency. You can inspect the statistical engines, verify the math, and trust the results. You can self-host for complete data control. No vendor lock-in—your data and your experiments stay yours.
Ready to ship faster?
No credit card required. Start with feature flags, experimentation, and product analytics—free.
