Objective: Observational studies of practices used in clinical supervision-as-usual can be leveraged to advance the limited research on workplace-based supervision as an evidence-based treatment (EBT) implementation strategy. This exploratory observational study examined the presence of supervision approaches (comprised of supervision techniques) and whether these predicted clinicians' EBT technique delivery. Methods: Participants included 28 supervisors, 70 clinician supervisees, and 60 youth clients and guardians from 17 public mental health organizations. Data included audio recorded supervision-as-usual sessions over 1 year, audio recorded Trauma-focused Cognitive Behavioral Therapy (TF-CBT) treatment sessions with youth for 6 months, and youth-reported post-traumatic stress severity scores. Audio recordings of 438 supervision sessions were coded for session duration and the presence of 13 supervision techniques and intensity of their coverage. Audio recordings of 465 treatment sessions were coded for presence and intensity of coverage of TF-CBT practice elements. Agglomerative hierarchical cluster analysis examined the presence of clusters of supervision technique use, termed supervision approaches. Generalized estimating equations estimated the relation between supervision approaches and delivery of TF-CBT elements. Results: Two supervision approaches were identified- Supportive-Directive and Supportive- that discriminated between use of five supervision techniques. Clinicians who received a higher proportion of supportive-directive supervision sessions had greater odds of delivering the trauma narrative with a client. Conclusion: Findings suggest that patterns of supervision techniques can be identified and may shape EBT delivery. Supervision approaches show some evidence of being tailored to individual clinicians. Implications for the development of supervision implementation strategies and future directions are discussed.