The use of video to assist with ethnographical-based research is not a new phenomenon. Recent advances in technology have reduced the costs and technical expertise required to use videos for gathering research data. Audio-visual records of learning activities as they take place, allow for many non-vocal and inter-personal communication interchanges to be collected. However, the use of video does add greater complexity to the research process. In this article, the use of video to collect data of trade-skill learners and the accompanying multimodal discourse analysis method used to derive meaning from these data are discussed. The adoption of video along with multimodal discourse analysis may facilitate future research into learning in a vocational context.