Emerging from projects that have involved working with primary school children in school-related research, this article offers suggestions of how drawing as a principal means of data gathering can be either constructive or of little value. The qualitative research projects discussed include investigations of school improvement and consideration of school design, in which freehand drawing was used in different contexts with young people. In many cases, the value of the visual data was high, contributing strongly to the research aims. In some, however, the work contributed little meaningful data to address the research questions. The usefulness of data derived from drawings to the research was assessed by observing factors such as the materials and time available as well as teacher and peer influence. This article discusses those variables and proposes recommendations to improve the likelihood of obtaining quality visual data when working with children.