People communicate through language as well as visual embodied actions like gestures, yet audio remains the default recording technology in interview-based writing research. Given that texts and writing processes are understood to involve semiotic resources beyond language, interview talk should receive similar treatment. In this article, I synthesize research that examines how visual embodied actions reveal and construct embodied knowledge and stance, and I apply these lenses to my own study, showing how visual embodied actions are essential to understanding three writers' experiences with particular writing styles. I conclude by discussing the benefits of videorecording for writing research, offering guidance on how video can help researchers explore the interview as a social practice, and suggesting ways to design the consent process with transparency and democratic practice in mind. Ultimately, this article serves as a guide for writing researchers who wish to challenge the audio default when conducting interviews.