The study combines the student data obtained by Jääskeläinen with new data from four professional translators and four “educated laypersons” (true non-professionals in translation). Contrary to the initial plan (“identify professional behaviour”), the aim is to identify process features which seem to go together with successful task performance, as not all professionals did well in the experiment. The explanations relate to differences in the amount of time and effort invested in the task, as well as differences in the types of knowledge used: the least successful processes, irrespective of the professionality of the translator, seemed to be characterised by treating the task as a mechanistic code-switching operation.