Introduction. The quality and reliability of data regarding changes in regional labour markets, as well as the quantitative and qualitative characteristics of the demand for workers and mid-level specialists, alongside an assessment of the potential for modernising secondary vocational education and the vocational training system, represent significant challenges for medium- and long-term planning in this field. Addressing these challenges largely relies on forecasting techniques that integrate data on labour market development prospects with information on the capabilities of personnel training systems. Aim. The present research aimed to explore the concept of “predictive analytics” as the foundation of a data-driven management methodology and to examine the potential application of this methodology in vocational education. Methodology and research methods. The applied research was conducted using a comprehensive scientific methodology. Various methods were employed, including generalisation, theoretical analysis, empirical analysis, cluster analysis, synthesis, and conceptualisation. Results and their scientific novelty. The authors view predictive analytics as a tool for implementing data-driven management methodologies in vocational education and they substantiate the concept of delayed educational outcomes as the central focus of predictive analytics within the education management system. Practical significance. Four groups of parameters are proposed to facilitate the development of various predictive analytics models, applicable at both the level of an educational organisation and within the regional vocational education system.