Best Practices for Split Testing with Bubble Apps

I’d love to hear how people are running split tests within their apps.

Are you using an external service that enables you to simply change text values based on element ID’s, or are you running the tests within Bubble? If you’re running them within Bubble, how are you tracking results for each variation?

Any particular gotchas, lessons learned, or other takeaways that’d be valuable for other community members?

@sridharan.s, interested in hearing about this as well.

I haven’t done much split testing, but I have found it immensely useful to utilize FullStory for understanding user sessions, especially during the debugging phase with distributed testers.

I wouldn’t say I have enough volume to make meaningful A/B distinctions. But running through tens to a hundred sessions on FullStory gives a pretty good picture of where users attention goes most, where they are rage clicking, etc.

1 Like