By international standards, the Italian judicial system is performing poorly (OECD, 2013a; Esposito et al., 2014; Ministero della Giustizia (Italian Ministry of Justice), 2015). Scholars and practitioners have advanced a number of possible reasons for this critical condition, such as the high level of litigiousness in the Italian society (Buonanno and Galizzi, 2014; CEPEJ, 2014); the limited availability of alternative ways of resolving disputes (Castelli, 2013); and tensions between political and judicial powers (Guarnieri, 2015). In addition, however, deficiencies in the management of Italian judicial offices are also blamed; despite long-running debate, efforts to modernise and reform managerial practices began only in the late 2000s (Steelman and Fabri, 2008; Vecchi, 2013a). The present paper investigates the implementation of a national programme for improving the management of judicial offices and describes an in-depth case study of the Law Court of the Milan District (LCM). Amid patchy implementation of this programme, Milan stands out as a success story and therefore as a possible source of insight into how modern management tools can be introduced to judicial offices. For present purposes, we adopt the perspective of programme designers and policy makers at the national level. Confronted with the poor implementation results across the country, a prospective designer might look to the LCM in envisaging a more effective intervention. Learning from the LCM success requires an understanding of whether and how the programme helped to initiate change in the LCM and – given the likely influence of local factors – how these can be integrated into a new national programme and reproduced in different contexts. To begin, the analysis develops a three-step theory of how the programme may have contributed to the LCM success by engaging leaders, providing previously inaccessible knowledge, and securing knowledge transfer. This theory is tested using process tracing (PT) (Beach and Pedersen, 2013; Bennett and Checkel, 2015), which is an innovative method for making within-case inferences from a single case study. PT is one of the most promising developments to emerge from recent debate concerning qualitative methodology (see Brady and Collier, 2010), and the public management context represents fertile ground for the deployment of this method. The paper proceeds as follows. Section 2 provides an overview of both the national programme and the LCM case. Section 3 introduces the method of PT, and Section 4 elaborates the three-step programme theory referred to above. Based on PT prescriptions, Section 5 discusses the prior likelihood of the hypothesis and sets out the relevant evidence. Finally, Sections 6-8 describe the case material and discuss the probative value of the evidence collected.