The “what works” literature, also referred to as “evidence-based practice,” has led to a shift in the way corrections approaches treatment interventions. Generally, the “what works” literature pinpoints which types of treatment interventions work best to reduce recidivism rates among offenders. In the United Kingdom especially, the work of entire organizations is based on evidence-based practice. However, some argue that caution must be taken to avoid offering a unilateral approach treatment intervention model that is based strictly on data concerning reconviction rates as indicators of treatment success. This article questions the efficacy of applying the major findings of the “what works” literature to non-mainstream correctional groups, including female, adolescent, and Indigenous offenders. A meta-analysis of Andrews and Bonta (2003) and the “what works” literature reveals that successful rehabilitation depends on whether treatment is matched to the criminogenic and non-criminogenic needs of the offender. The importance of offender needs-assessments is underscored in choosing the appropriate treatment approach for non-mainstream offenders. Furthermore, the unilateral application of the “what works” literature to non-mainstream correctional groups is uncalled for given the lack of research and evaluation of the effectiveness of various types of interventions, particularly the popular group-based and cognitive-behavioral treatment, with non-mainstream offender groups. In fact, research concerning adolescent, female, and Indigenous offenders has been largely neglected, especially in Australia. Finally, it is argued that there is a lack of evidence that would specify the uncritical application of group-based and cognitive-behavioral treatments to groups of offenders outside of the adult mainstream. Future research concerning effective treatment processes for non-mainstream offenders is called for given the gap in the current literature. References