Organisations are rapidly experimenting with agentic artificial intelligence systems that can plan, act and learn autonomously. However, organisations often lack concrete guidance on how to design, deploy and govern these systems in practice. This gap between high-level readiness models and day-to-day implementation decisions leaves managers unsure when their organisation is truly ready, which risks stalled pilots, wasted investment and unsafe deployments. Based on a design science methodology, this thesis develops and validates a set of design principles and implementation guidelines for agentic AI deployment in organisations in the Netherlands and then operationalises them in a practical management readiness checklist. A surprising finding from interviews with practitioners and experts is that the main obstacles to effective agentic AI were not technical limitations in models or data infrastructure but organisational factors such as risk-averse leadership, fear of job loss among employees and unclear ownership of oversight. At the same time, interviewees converged on a concrete strategy for building trust, introducing agentic AI as a junior employee that gradually earns autonomy while human oversight shifts from close supervision to high-level monitoring. The resulting design principles and readiness checklist translate this insight into diagnostic questions and observable evidence that support go or no-go decisions. In doing so, the thesis turns agentic AI hype into operational value by giving managers a structured way to align culture, processes, data and governance with the capabilities and risks of autonomous AI agents.