Purpose: This paper contends that data generated by research on supervision are often taken as authentic data. Through an examination of studies that use audio/visual recordings to investigate supervision, the paper both promotes and problematises the recording of supervision meetings as a useful technique for doctoral supervision research. This paper aims to encourage a critical evaluation of methodological choices in research on supervision, and both promotes and problematises the practice of recording supervision meetings to enhance nuance in research on supervision practices. Design/methodology/approach: This paper reviews how prior studies have adopted different research methods to construct the space of supervision, and how the chosen methods have been justified. The paper draws on data from an empirical study which included interviews with supervisors in China, based on recordings of their supervision meetings. Findings: Presenting a single case with one participant to explore the recording and interview process in detail, this study demonstrates how hearing the supervision meeting can present a multi-faceted picture of supervision practice. This multi-faceted picture underpins the alternative understanding of authentic data that this study unpacks. Originality/value: Drawing on the tradition of poststructuralist critiques of traditional research methodology, this study is presented as a methodological paper, with a core aim of interrogating and problematising methodological decisions taken in studies of doctoral supervision. This study reviews research methods that were used in prior studies on supervision, investigating how the chosen methods were justified and how these methods affect the resultant construction of supervision.