Artificial Intelligence (AI)-enhanced decision-support systems (DSS) in the military domain can improve the speed, quality, and geographical coverage of critical decisions, resulting in operational advantages. Development and deployment of such DSS bring interdisciplinary challenges that must be systematically addressed. Technical challenges focus on determining and achieving the required performance levels of a DSS (e.g., accuracy, resolution, and speed), how to allocate functions between automation and humans, and how to avoid the cognitive costs of automation. Legal and ethical challenges regarding the use of AI-enhanced DSS focus on compliance with targeting law, control, agency and responsibility. The key to successfully tackling these challenges is understanding the decision making within the context in which a DSS would be used. Our solution to tackling these challenges is a contextual analysis of decision-making processes and the operational environment in which they take place. Operational, technical, legal, and ethical experts should understand the end-users’ Decision-Making Process (DMP) in various contexts and at different abstraction levels, the contextual factors influencing the decisions, their impact in the real world, and potential risks involved. Knowledge about the operational and contextual settings can be used to: (i) extract technical requirements; (ii) systematically study the potential impact of a DSS in the DMP and safe onboarding into the larger operational system; and (iii) analyse the potential benefits and negative side effects of the DSS under relevant conditions from the operational, legal, and ethical perspectives. The Contextualised AI Onboarding Method (COAIM) provides various tools to extract the context-based knowledge needed to tackle challenges related to AI-enhanced DSS development and onboarding.