This is a study in the field of Translation Studies, more specifically in the field of descriptive studies of the translation process. It belongs to a bigger research (Malta, 2015) in The Experimental Translation Laboratory (LETRA), whose goal has been to characterize the translation process in a (re)translation task from key-logging registration data and eye-tracking triangulated by retrospective protocols. The analysis looked for evidence of cognitive effort. The research participants were translation students and Spanish as a foreign language teachers, all of them with experience in translation. The participants’ translation process was captured in real time by a recording program of Translog II activation and by eye-tracking (Tobii T60). This article is the second of a trilogy about the (re)translation process, whose first paper about eye-tracking data was published in 2016. In this article, key-logging data will be presented and analyzed. As far as the translation process phases are concerned, the results revealed that the research participants spent the longest amount of time doing the drafting phase, followed by the orientation phase and, lastly, the revision phase, being the latter the most heterogeneous phase. Furthermore, it was recorded that the number of translation micro units impacted on the total time of the task performance.