A Framework for Evaluating Potential Return On Investment in Artificial Intelligence Based on Implementation Risks

Artificial intelligence (AI) is increasingly integrated into organisational practice, yet a substantial share of implementation projects fail to achieve their intended outcomes. Gartner’s 2024 forecast indicates that up to 30% of AI initiatives may be cancelled due to unclear value creation, poor data quality, and underestimated costs. Despite the proliferation of maturity models, risk frameworks, and ROI calculators, these tools are typically applied in isolation, leaving organisations without a coherent structure that links technological readiness with financial outcomes. This paper presents a conceptual decision-support framework for evaluating the potential return on investment (ROI) of AI initiatives, integrating three interdependent analytical dimensions: organisational maturity, solution complexity, and data–process health. The framework consolidates previously fragmented approaches and translates AI-related risks into financial indicators, enabling more structured investment planning. The contribution lies not only in integrating existing frameworks but also in positioning organisational readiness as a precondition for meaningful ROI assessment. Future research will focus on empirical validation across industrial and service-sector enterprises.