Abstract Exploratory Data Analysis (EDA), originally framed by John Tukey as a discipline of visual reasoning and judgement, is increasingly vital to managerial decision-making in environments characterised by high-dimensional and irregular data. This paper reframes EDA as a contemporary analytical methodology that enhances managerial insight. We do this by combining modern transformation techniques with visual diagnostics to reveal structure, surface anomalies, and challenge assumptions. Drawing on case studies in environmental science, digital forensics, and public policy, we show how EDA enables managers to act as numerical detectives — interrogating uncertainty, validating assumptions, and identifying operational risks earlier in the decision cycle. Across these cases, EDA exposes concealed variability and distributional irregularities that conventional reporting pipelines overlook. We conclude by discussing implications for analytical reproducibility, auditability, and the expanding role of EDA in organisational diagnostics and risk-informed strategic decision-making.
Cooke et al. (Fri,) studied this question.