The current study investigated the effects of segmentation design and drawing on college students' video learning. Participants were 158 college students randomly assigned to view either a segmented or continuous video lecture (video type: segmented vs continuous) and who either received instructed to draw while learning or no instructions at all (learning strategy: drawing vs passive viewing). Participants' eye movements were recorded as they viewed the video, and data was collected regarding their learning satisfaction, cognitive load, both immediate and 7-day delayed learning outcomes, and their perceptions regarding the instructional efficiency of the lectures. The results showed that the drawing activity moderated the segmentation effect in that students did not benefit from the segmented video design when viewing passively, but did when required to draw while viewing. Furthermore, the positive effect of segmentation was mediated by drawing accuracy.