Interactive tasks have been claimed to be an effective tool for the classroom by studies within interactionist perspectives (Mackey, 2012) as well as by those from the field of task-based language teaching (Van den Branden, Bygate and Norris, 2009). Among the variables that contribute to this effectiveness, the present study concentrates on task repetition and, more specifically, on procedural repetition, the repetition of the same task type with different content. Procedural repetition has great interest for the classroom, where tasks are often repeated more than once with greater or lesser modifications. To date, this variable has only received a modest amount of attention and researchers have mainly focused on its effects on the overall performance (usually measured in terms of complexity, fluency and accuracy) of learners working individually (Kim and Tracey-Ventura, 2013) while only few studies have combined task repetition with the use of collaborative tasks in order to analyse its effects on learners’ amount of negotiation (Mackey, Kanganas and Oliver, 2007). The objective of the present study is to measure the effects of procedural repetition on the oral production of a group of 20 11-year-olds learning English in a Content and Language Integrated (CLIL) school in Spain while carrying out an interactive task in pairs. The learners had to resolve the same task type (a picture placement task) with different content three times over a period of three weeks. Their oral interactions were analyzed to search for any effects of repetition on the following aspects: amount and type of interaction strategies, accuracy, complexity, fluency and L1 use. The results obtained show some effects of task repetition in the case of interactional features and accuracy: The amount of interaction strategies produced by the participants as well as the number of errors per clause significantly decrease in the third repetition. On the contrary, complexity and fluency remain clearly stable throughout the three tasks. As for the results regarding L1 use, they present a complex scenario. Although the learners resorted to Spanish only scarcely at all times, the number of L1 terms increased significantly in the second repetition only to drop again in the third one. Also, the L1 terms used by our learners often referred to simple words that the learners could have easily paraphrased. All this suggests that amount of L1 use, which has been commonly linked to task difficulty, motivation or students’ level of proficiency, also seems to be used randomly on some occasions. In light of these results some pedagogical implications of the use of procedural repetition of interactive tasks in the language classroom will be discussed.