GrowthBook 5.0

About GrowthBook 5.0

GrowthBook 5.0 is an open-source, warehouse-native platform for product development and experimentation. It combines feature flags, A/B testing, product analytics, and AI evaluations into a single system that works with data already in a company's existing data stack. The platform can be deployed in the cloud or self-hosted.

Review

GrowthBook 5.0 is the second major launch of the product, and it represents a significant expansion from earlier versions focused primarily on feature flags and experimentation. The 5.0 release adds product analytics and several AI-powered features while keeping the open-source, warehouse-native architecture.

Key Features

• AI Visual Editor - Builds no-code experiments directly in the browser without writing code.
• 25 open-source Skills - Let AI agents create feature flags and draft experiments programmatically.
• In-app AI Assistant - Allows users to ask questions about experiments in natural language, such as what changed or where to dig deeper.
• Product Analytics - Now generally available, providing product data exploration within the same platform.
• Streamlined experiment workflow - Reduces the number of form fields by about 20, making setup faster.
• Stronger feature flag governance - Adds controls for safe rollout of changes.
• Faster queries - Performance improvements for data retrieval and analysis.

Pricing and Value

GrowthBook offers free options, including the open-source self-hosted version. Cloud-hosted pricing details are not explicitly defined in the reference content. The platform's value comes from combining multiple tools-feature flags, experimentation, analytics, and AI evaluations-into one system that works with existing data warehouses.

Pros

• Open-source and self-hostable, giving teams full control over their data and infrastructure.
• Uses a Bayesian statistical approach for experiments, with transparent query logic that users can inspect.
• AI Assistant and Skills make experimentation more accessible to non-technical team members and AI agents.
• Warehouse-native architecture means data stays in the company's existing data stack.
• Active community support available through Slack.

Cons

• Interpretation of experiment results still requires good metric definitions and healthy skepticism; the tool does not automatically solve trust issues between teams.
• Not well suited for teams that want a fully managed, closed-source SaaS solution with no self-hosting option.
• As a product that combines multiple categories, it may have a steeper learning curve for teams only needing one specific function, such as basic feature flags.

GrowthBook 5.0 is best suited for product teams that already use a data warehouse and want a unified platform for feature management, experimentation, and analytics. It works well for organizations that value data ownership through self-hosting and need to involve both technical and non-technical stakeholders in the experimentation process.