{"type":"video","version":"1.0","html":"<iframe src=\"https://www.loom.com/embed/f3484d6f211d490ba3a502867482ed77\" frameborder=\"0\" width=\"1918\" height=\"1438\" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>","height":1438,"width":1918,"provider_name":"Loom","provider_url":"https://www.loom.com","thumbnail_height":1438,"thumbnail_width":1918,"thumbnail_url":"https://cdn.loom.com/sessions/thumbnails/f3484d6f211d490ba3a502867482ed77-1715626696741.gif","duration":396.833,"title":"Bayesian Model Updates for GrowthBook 3.0","description":"Hi everyone, in this Loom, I'm Luke Sonnett, the lead data scientist at GrowthBook. I'll be discussing the new Bayesian model we're rolling out as part of GrowthBook 3.0. The model offers three key advantages: the ability to specify your own priors, the inclusion of Cupid for variance reduction, and improved accuracy when the baseline average is near zero. I'll explain these changes and how they benefit your experimentation process. No action is required from you, but I encourage you to watch the video to understand the updates and their impact on your work."}