Big data analytics through the TOE framework: evidence from an automated machine learning approach

Purpose This study aims to investigate the determinants of big data analytics (BDA) use in manufacturing firms by applying the Technology–Organization–Environment (TOE) framework and automated machine learning (AutoML) approach. Design/methodology/approach By relying on the data set of 400 Polish firms, the authors analyze 40 predictors that combine survey data with secondary indicators at firm, industry, region and country levels. AutoML is applied to identify the most important factors and uncover potential nonlinear relationships across the TOE framework. Findings The technological dimension highlights the predictive relevance of regional innovation capacity and sectoral technological level. Organizational factors such as internationalization, buyer dependence and talent availability demonstrate strong predictive contributions, reflecting the interplay between global pressures and internal capabilities. In the environmental dimension, legal barriers have the highest predictive importance, while government support shows a more moderate predictive association. Overall, the findings reveal that technological factors alone do not provide the strongest predictive performance; organizational and environmental conditions contribute substantially to explaining variation in BDA usage. Originality/value This study extends BDA and Industry 4.0 research by enriching the TOE framework through a multilevel and nonlinear perspective and demonstrating the predictive value of AutoML for BDA use.