Hello there! 👋 Is there a possibility of defining nested namespaces, mutually exclusive to each other? Seems that this feature is not supported currently. We want to run mutually exclusive experiments on level 1 (the namespace that we currently define) - but below it on level 2 define experiments that should be grouped and mutually exclusive to each other (including the experiment that’s running on level 1), and so on… Is there a different approach we could take to achieve this? We would do this as our org is expanding experimentation to multiple teams and each team wants to run specific experiments related to their use cases without overlap and to achieve proper data isolation. Thank you so much!

Hi there! You're right that we don't currently support nested namespaces. In your use case, I'm understanding that you want to run mutually exclusive experiments on the same data without overlap/interference. Is that right?

Hey Ryan! 👋 That’s exactly right, our teams need a way to move autonomously and we want to enable them, while also having a scalable approach to manage their experiments. I found that coupling a Saved Group (Attribute) directly with a namespace - creating each and attaching both to the override rule to each ff can segment traffic in a single piece of the flow, that the Saved Group covers. And then we need alignment in our organization based on these limitations. I found this issue that might be covering some of this functionality - but I don’t know how much work it is exactly to contribute! https://github.com/growthbook/growthbook/issues/490 Let me know if you have any other suggestions on what we can currently do. Thanks so much!!

I chatted with our engineers and data science teams. At the moment, there's not a clear way to achieve this; however, they recognize the validity of the use case and are investigating the possibility of nested namespaces. If I get any additional info, I'll share it with you.