A holistic model for evaluating innovation in agro-industries through the intersection of internal and external factors

Purpose In an increasingly competitive and technology-driven environment, innovation has become essential for agro-industries seeking to maintain competitiveness and adapt to dynamic markets. However, innovation assessment in this sector remains fragmented and often limited to isolated indicators. This study aims to analyze internal and external factors of innovation and to propose a holistic model for evaluating the degree of innovation in agro-industries. Design/methodology/approach A systematic literature review (SLR) was conducted following a structured and transparent protocol to identify innovation factors relevant to agro-industrial contexts. The identified factors were classified into internal and external dimensions and subsequently validated by sector experts to ensure practical relevance. A mathematical measurement model was then developed using the best–worst method (BWM), a multi criteria decision-making technique applied to weight the innovation factors and structure the evaluation framework. Findings The results identify innovation factors organized into internal and external dimensions that together form a holistic innovation evaluation model. The weighting process highlights the multidimensional nature of innovation in agri-industries, emphasizing the combined influence of strategic, organizational, technological, market-related and institutional factors. The proposed framework integrates these elements into a hierarchical structure linked to key performance indicators, enabling a strategic diagnostic of innovation performance. Originality/value This study develops a holistic and structured model for innovation evaluation in agri-industries, grounded in systematic evidence and expert knowledge. The application of the Best–Worst Method represents a methodological advancement in innovation measurement for the agri-food sector. The model was applied in an exploratory assessment involving seven agri-industrial firms, demonstrating its feasibility and diagnostic potential, while large-scale validation remains a future research stage.