The laminar burning velocity is a fundamental property of combustible mixtures important for kinetic model validation as well as for practical applications. Many efforts are directed towards its accurate determination. The heat flux method is one of the commonly recognized methods for measuring laminar burning velocity, however, the information on the accuracy of the method is scattered in the literature. In the present work, an attempt was made to summarize and extend the available information on the different factors contributing to the experimental uncertainty of the heat flux method. Experimental setup of the Lund University group, typical for the heat flux community, and the procedures used to determine the burning velocity are described. Furthermore, the influence of different uncertainty factors, originating from each part of the setup, is analyzed. Asymmetric heat fluxes and the method for determining flame surface area were found to give an important contribution to the total error. As a result of this, some of the previously published data have been re-evaluated. Finally, recommendations are presented on how to control or reduce the uncertainties in the heat flux measurements, and possible directions for future research, aimed at improvement of the accuracy and understanding of the method, are outlined.