In this article, the authors share some of the methodological lessons learned from the Third International Mathematics and Science Study (TIMSS) 1995 and TIMSS 1999 Video Studies. They focus on their experiences overseeing these large-scale, video-based, cross-national surveys of teaching. The article documents some of the progress the authors made, as well as some of the obstacles they encountered. The authors hope that the ideas shared here help prevent possible missteps, enhance the quality of projects, and encourage researchers to advance the use of video as a methodology.