Serious, organised crime is an endemic social problem. Despite the extensive literature on organised criminality, the evidence is primarily qualitative, based on selective samples, or both. Population-level data and new sources of information on this phenomenon are needed to overcome these issues. Moreover, the interventions currently available to prevent organised crime from materialising in the first place have shown little success in curbing this multibillion dollar industry. Social control apparatuses such as law enforcement are often understudied and, when such research evidence exists, it is often based on weak causal estimates of treatment effects due to methodological shortcomings. This dissertation attempts to fill these gaps by answering three interrelated primary research questions. First, whom should the police target to prevent them entering organised crime groups, rather than pursuing them once they’re in? Analysis of population-level intelligence and records on organised crime show that prolific offenders who have yet to join an organised crime group provide a suitable yet under-investigated group of offenders. Second, where are serious and organised crime offenders’ haunts or hangouts? By identifying the “convergence points” of gangs and organised crime group members, new locations can be identified and targeted with Protect and Prevent police tactics. Third, do prevention programmes effectively reduce crime originating from organised crime offenders? Using a cluster randomised control trial methodology, I test the deterrent effect of a ‘carrot and stick’ approach on recruits and measure the vicarious effect on their recruiters. The test shows significant impacts that are potentially replicable in other serious and organised crime areas and in other policing fields. This project's practical and theoretical implications are discussed, with a call for future replications of this approach.