The question of what constitutes quality in interpreting has been answered in different ways by linguists, psychologists and interpreters themselves. Measurements were initially based on transcripts and found insufficient, but obtaining authentic audio and video data that can be analysed is difficult. Even surveys among users appear to yield inconclusive results. If interpreting is regarded as a service whose quality needs to be assessed, it is necessary to choose a process perspective which includes all phases of an interpreting assignment. For the purposes of quality assurance, a checklist that includes the different phases helps document how an assignment has been carried out. This article attempts to develop profiles of speeches made in conferences and their versions in other languages as produced by conference interpreters. Within this framework, an interpreting product can be assessed as a function of the original speech and its characteristics. Thus, a degree of measurability should be achieved, and the results of measurements should reflect not only source and target texts and their presentation but also the conditions and factors that influence their perceived quality. The profiles were tested with professional interpreters and may serve as a component of quality assurance in the field of conference interpreting, and hopefully in the future for all types of interpreting.