The next part of the process evaluation consisted of qualitative semi-structured interviews with participants of the MAW structures. The original intention was to include three cities per country and five interviews per city (3*15) in order to achieve a representative sample. Due to staff turnover involving our contacts in one Belgian city, this city was replaced by another one. Two interview had already been carried out prior to that. Furthermore, due to the difficulties in motivating a third city from the Netherlands to take part in the process evaluation, this third Dutch city was replaced by an additional Belgian city. This resulted in a sample of three German cities (n=3×5), two cities in the Netherlands (n=2×5) and five in Belgium (n=(4×5)+2), giving a total of 47 (15+10+22) interviews, two more than intended. The in-depth interviews were carried out both face-to-face and online via Microsoft Teams, depending on the Covid-19 pandemic measures that were in place at the time. We interviewed the mayors of the included cities, different local security actors and the local socio-preventive actors. Our partners VVSG, VPN and RadarAdvies facilitated the first contacts with candidates for the scientific evaluation, by explaining the framework of the research design, and indicating the importance of participation, etc. After this introduction, we were able to get started and make contact. The interviews in Dutch (i.e., Belgium and the Netherlands) were conducted by researchers from Ghent University. The German interviews were carried out by VPN, assisted by Ghent University. The interviews were based on a semi-structured interview guide and an informed consent form was signed each time. The semi-structured interview guide was created using the process indicators (listed in Appendix 3) that had been derived from the literature review. This chapter outlines the findings from the interviews. It is structured according to the broad categories used to group the process indicators, and this grouping also forms the basis for the structure of the self-evaluation tool. It must be noted that this list of process indicators is not exhaustive. The semi-structured interview guide can be found in Appendix 5. All interviews were conducted in Dutch, German or English and were fully transcribed. The interviews in German were transcribed in German and then translated for analysis. The interviews were coded using the program NVivo by means of a codebook. The codebook was based on the process indicators and gradually supplemented throughout the analysis. In total, 51 MAW actors were interviewed (47 interviews, of which four had two interviewees present) in Belgium, the Netherlands and Germany. Some features of the interviewees are summarised here. One in four respondents had either a coordinating function (approximately 25%) or a participatory role (the majority, almost 80%) at the MAW meeting. On average, respondents had been in the MAW structure for three years. They worked in the following sectors: police/security, municipality, judiciary, social/wellbeing, academia/experts or youth and education. The most common organisations within these sectors were the police, social services and civil services. Twenty-one female and 28 male respondents were interviewed