The focus of this paper is on self-recording, a method for collecting empirical data. The context for the method is a study of the professional identity development of novice primary school mathematics teachers as seen from the perspective of the teachers themselves. The background for using the method is based on theoretical and methodological issues as well as problems regarding data collection in this study. Self-recordings were tried as a complement to observations and interviews and contributed with data of original character, e.g. data regarding current happenings and data giving insights into day-to-day work from the perspective of the informants. In this paper, the theoretical and methodological starting points, the method as such and examples of the empirical outcome will be presented and discussed.