In recent years, video-based lectures have been increasingly used in education as part of flipped classroom approaches, adult education, distance learning, and other applications. Though approaches in video-based lectures have been widely studied, less has been reported in the development of approaches to teach relevant technical skills, such as laboratory techniques. This type of skill acquisition is made more complicated by the need to understand what a skilled person would be looking at, what they might be doing (while monitoring something else visually), and what subtle interplay these two might have. Here, we report a multiperspective video approach that allows a viewer to see both what a skilled person would be watching and doing while guiding the viewer's attention between these domains. We compare two cohorts of students (one faculty-trained in class, the other trained via this video-based approach) and examine how well each group performed a task with a clearly defined numerical "right answer." We find that both approaches were equivalent in fostering skill acquisition and that similar approaches should be valid in various space-separated teaching approaches (e.g., distance learning, flipped classrooms).