Background: Clinical aphasiologists have long attended to repetition in aphasia classification and used it in treatment. Within traditional approaches, repetition has been conceptualised narrowly as the ability to produce relatively immediate, verbatim reproductions of target behaviours; treatment protocols have relied heavily on drill, eliciting client repetition of targets. In sharp contrast, sociocultural theories conceptualise repetition as a fundamental, pervasive feature operating at every level of language use. Repetition thus includes partial and paraphrased as well as verbatim repetitions, across time as well as immediate. These theories also stress the communicative functions of repetition. With respect to learning, sociocultural theories emphasise the way such loosely structured, diverse patterns of repetition emerge in, and are prompted by, repeated engagement in meaningful activities. Aims: This study (1) presents a sociocultural approach to repetition in conversation; (2) illustrates that approach through analysis of a clinician-client pair's repeated productions of labels for 30 target cards during a 10-session pilot treatment; and (3) offers detailed examples of how the pair's repeated engagement with target cards across sessions might support learning. Methods and Procedures: This study utilises situated discourse analysis of a pilot barrier task treatment (10 sessions) in which the clinician and client (a 67-year-old man, with mild anomia and severe amnesia) worked together as partners to identify and place target cards. At the end of each session researchers interviewed the clinician-client pair to identify their agreed-upon target labels (ATL) for the cards. Analysis of card label repetition included: (1) identification of all verbal labels used for target cards during game play; and (2) coding labels as matching or not matching the ATL, and as either a first or repeated use of that label during that sequence. Outcomes and Results: Analysis confirmed that repetition was pervasive. The client-clinician pair routinely repeated their own or each other's referencing expressions during the task, collaboratively developing specific, meaningful, and increasingly succinct labels from chains of conversational repetition (within, between, and across trials). Critically, this repetition occurred without clinician-directed repetition of isolated treatment targets. Conclusions: This examination of repetition suggests that marshalling conversational repetition through repeated engagement offers a theoretically and empirically grounded framework for reconceptualising language intervention. Furthermore, memory research offers useful guidance in understanding the role of repetition across multiple types of learning, which we propose can guide SLPs in when to deploy drill-based and/or conversational repetition to best achieve specific treatment goals.