Every time a practitioner in children's services offers a child and their family an intervention a decision or decisions has to be made. However the decision making process in children's services has rarely been studied. Decision making has been extensively studied in other disciplines in both laboratory and real life situations. A dual process model has been proposed consisting of a fast, automatic, intuitive system and a slower reflective system. The two systems are deemed to work best in different situations and have their own strengths and weaknesses. Decision making in complex situations, such as those involving children and families, involves both types of processes but checks and balances help to ensure that the process is optimal. Expertise can develop over time through reflection on the process.This study explores the decision making process in a Targeted Mental Health in Schools Team (TaMHS) in one Local Authority. TaMHS is a three year Department for Children, Schools and Families' pathfinder programme aiming 'to improve mental health outcomes for children and young people via interventions delivered through school' (DCSF, 2008d). Substantial changes have taken place within children's services over recent years and research has explored the facilitators of and barriers to effective multi-agency working. However lack of clarity in terminology and detail has prevented an evaluation of the causal links between the facilitators and better outcomes for children and young people. I have used a case study approach with a multi-agency team which has practitioners from six professional backgrounds. Interview data from the manager and six practitioners and an observation of one of their cluster meetings has been collected and analysed using thematic analysis. I have developed a rich picture of the decision making process (DMP) in this team. The DMP is a complex, iterative process which is facilitated by a predetermined organisational structure and continues throughout the assessment and intervention stages. Diversity of views is welcomed and different perspectives are merged leading to shared decisions. Families and school staff are fully involved. Practitioners seem to use processes below conscious awareness as well as a more explicit process which links explanatory models, chiefly risk and resilience, with the choices of interventions. I have identified that many of the known facilitators for effective multi-agency working exist within this team and I propose that these could be the mechanisms which trigger effective decision making. I suggest that the group process involved in this team could be useful for other teams in children's services. I also discuss ways to improve decision making and I have created a DMP Attributes Model which I have described and then discussed as a tool to aid professional development through the supervision process for practitioners within children's services. I explore a possible role for educational psychologists in this process. Future research could study the usefulness of this tool with practitioners.