An initial evaluation of the utility of designing an intervention to address neuroscience-based subtyping of children who have conduct problems was undertaken in this pilot study. Drawing on the literature on callous-unemotional traits, a novel intervention programme, "Let's Get Smart", was implemented in a school for children with social emotional and behavioural difficulties. A mixed-methods design was used to investigate the perspectives of staff participant-observers in the change process, alongside standardised scores on measures of pupil performance and behaviour. Both qualitative and quantitative results showed reductions in externalising behaviour and improvements in measures of hypothesised underlying cognitive and affective processes. While externalising behaviour improved across subtypes, associated changes in underlying processes differed by subtype, supporting the potential value of neuroscience-informed contributions to intervention planning. (Contains 4 tables.)