Decision overload in data-driven procurement

Designing organisational decision architectures

Abstract

Increasing digitalisation has expanded the availability of data in operational procurement processes. Although data-driven systems are expected to improve decision-making, decision quality does not necessarily increase under conditions of high information density. Instead, decision-makers are confronted with multiple simultaneously plausible options, resulting in decision overload. This study analyses decision overload in data-driven ordering processes from an organisational perspective. Based on a structured literature review and theory-guided case analysis, the study identifies recurring mechanisms that impair decision-making capability. The findings show that decision overload does not primarily result from data availability, but from the absence of institutionalised selection mechanisms. Three core mechanisms are identified: lack of information selection, diffusion of decision responsibility, and absence of goal prioritisation. Building on these findings, a decision architecture framework is developed that translates these mechanisms into organisational design principles. The framework includes information filtering, clear assignment of responsibilities, temporal structuring, limitation of automated decision-making logic, and institutionalised goal prioritisation. The study shifts the focus of data-driven decision-making research from data and analytics to organisational decision structures, demonstrating that decision quality depends on the institutional design of selection mechanisms.