Global transitions to sustainable development: A cross-country clustering of SDG progress

Achieving the Sustainable Development Goals (SDGs) demands systemic transformations rather than isolated interventions. This paper applies an unsupervised neural network clustering technique (self-organizing maps) to cluster 184 countries based on SDG indicator data, employing multiple imputation to address missing values. It examines the extent and patterns of transformative shifts across six transition areas: food systems, energy, digital connectivity, education, welfare and social protection, and environment. By clustering countries based on their performance in 2015 and 2025 separately, the analysis reveals divergent development trajectories and dynamic shifts over the decade. Digital connectivity emerges as the transition with the strongest positive momentum globally, while environmental sustainability shows the most concerning setback. Notably, income level does not predict a country's development trajectory, though it remains a strong driver when channeled through building foundations for change. A domino-like sequence of transitions is evident, with education and digital connectivity serving as foundational catalysts. Investing in education can drive digital transformation, which in turn stimulates progress in environment, energy, and welfare and social protection, ultimately cascading into food system improvements. These findings offer actionable insights for policymakers seeking to identify high-leverage intervention points within the integrated architecture of sustainable development.