YouTube details planned video A/B tests with retention curves for both edits
YouTube is moving toward A/B testing the video itself, giving creators a way to compare edits rather than just thumbnails. In a November 16 YouTube Official Blog discussion with Paddy Galloway, VP of Creator Product Amjad Hanif described a planned workflow that would show retention curves for both edits. The post doesn’t announce a launch date, eligibility or general availability.
The proposed process has a clear decision point: YouTube will run a test, provide results and then give the creator seven days to choose which edit to use. That seven-day window is for making the choice; the post doesn’t specify how long the test will run. Hanif expects intros to be the most common use, based on conversations with creators, rather than describing an intro-only restriction.
For a creator, the useful prospect is being able to investigate an opening that didn’t land. Galloway explains that thumbnail optimization has received so much attention partly because it’s one of the few things creators can change after uploading to affect performance. Testing edits would extend that opportunity into the storytelling itself.

Retention curves could make the comparison more useful than a single winning score. They would let you examine where viewers leave each version, rather than relying on thumbnail clickthrough rate to judge what happens inside the video. Hanif says that focus is deliberate: the aim is to keep creators looking at information that helps explain the edit’s performance.
Once testing becomes available to you, a useful experiment could be comparing two ways into the same story and examining how their retention curves differ. For now, this is a preview of the intended workflow—not a reason to assume you can replace or test an uploaded intro today.
Source: YouTube Official Blog · 2026-11-16
By Hookami




