Quantifying the impact of digital technology on enterprise processes using LLM-based causal claim networks

The adoption of digital technologies (DTs) has emerged as a key driver of enterprise digital transformation. While prior research has primarily examined the effects of individual DTs on specific enterprise processes, a comprehensive understanding of how diverse DTs uniquely influence the broader transformation landscape remains limited. To address this gap, this study proposes a Large Language Model (LLM)-based causal analysis framework for constructing knowledge graphs that map DTs, enterprise processes, and their causal relationships. Leveraging custom language models, we extract causal narratives in the form of cause-effect pairs from large-scale academic publications, identifying five common variable types: independent, dependent, effect, mediator, and moderator. Based on these, we construct causal claim networks that link various DTs to seven core enterprise processes. We further investigate how DTs shape these processes over time. The findings reveal significant disparities in the impact of various DTs on enterprise processes, correlating with the technological advancements inherent to each technology. Furthermore, the application of DTs within enterprise processes follows a progressive expansion, evolving from isolated processes to more comprehensive organizational functions. Throughout this progression, the adoption of DTs has fostered cross-process collaboration and interaction. This study proposes a novel representation approach based on a hypothesis-driven semantic causal network to elucidate how various DTs drive the complex process of enterprise digital transformation.