GrowthBook is an open source tool for feature flags and experiments. It reads results from your own data warehouse.
It has a free plan and a self hosted option. Paid plans add more features.
A summary of patterns across popular review sites and community discussions
Other tools worth comparing, based on how teams typically choose between them
How much setup and in-house expertise this tool needs, and how much you would depend on outside help
Independence checks 5 of 5 met
✓Can start on your own. Free or trial start, self-serve signup.
✓Works without marketing know-how. Simple, guided use.
✓Your team can run it day to day. Works for small teams.
✓Low risk of costly mistakes. Low risk, results are easy to review.
✓Affordable to start small. Affordable to start small.
Product teams that want control over their data.
Teams wanting a no code editor.
A few walkthroughs to see the tool in action before you commit
Practical answers, plus what users report on review sites
GrowthBook is an open source platform for feature flags and A/B testing. We could use it to run experiments and read results from our own data warehouse, so we do not need to move data to another vendor. It is mainly developer led.
It can be, if we have at least one developer or technical person. A free cloud plan for up to 3 users and a free self-hosted open source edition are listed. G2 reviewers mention a learning curve and an interface less intuitive for non-technical users, so marketers alone may struggle.
Toolradar lists a free Starter cloud plan for up to 3 users, a Pro plan at about 40 dollars per user per month for up to 50 users, and custom Enterprise pricing. Self-hosting is free under the MIT license but needs infrastructure expertise and maintenance. We would confirm on the vendor pricing page, and we could not confirm our own usage costs for a warehouse.
Listed features include unlimited feature flags and experiments, warehouse-native analysis, and on Pro a visual A/B test editor, advanced statistics, sticky bucketing, multi-armed bandits and safe rollouts. We could run server-side and client-side tests. G2 reviewers praise the flexible implementation and feature flagging.
It helps rather than replaces. We still need someone to design experiments, add the code or tags and interpret the results. It handles assignment and statistics, not strategy.
Toolradar names Snowflake, BigQuery and Redshift as supported data warehouses, and says it works with other SQL sources and event trackers. We could not confirm a full integration list, so we would check the documentation for our analytics stack.
G2 shows about 4.6 out of 5 from 26 reviews, with praise for support and ease of testing, and complaints about the learning curve and no preview for server-side tests. We would check that we have a data warehouse or can set one up, and who maintains it. If self-hosting, we would plan for upkeep.
You must be logged in to submit a review.
No reviews yet. Be the first to review!