This paper outlines the design and conclusions, mostly methodological, drawn from a pilot study conducted among trainee translators to measure the use and usefulness of Knowledge-Rich Contexts (KRCs) in the translation process. After discussing the issue of context and KRCs in translation, it reviews the literature on previous observation protocols and tools designed for the study of the translation process. It then prevents the customized software (Argos) designed for the experiment to record the translator's activity. Finally, it describes the various parameters involved in the pilot study, some preliminary results and the methodological changes that are planned for the subsequent final experiment(s).