Abstract
Digital transformation has become a central construct in information systems research, yet what it refers to remains hard to pin down. Responding to concerns about construct clarity and validity, this paper shifts attention from definitional debates within academia to how digital transformation is articulated in practice. Drawing on a longitudinal analysis of earnings conference call transcripts from publicly listed firms between 2014 and 2024, we examine how digital discourse varies across industries and evolves over time. We show that digital transformation is neither a unified nor stable phenomenon. Instead, it exhibits systematic industry-level heterogeneity and temporal reconfiguration. The meaning of “digital” and the framing of “transformation” shift unevenly across sectors, with periods of convergence followed by renewed fragmentation. We conceptualize this pattern as discursive drift: a structured but ongoing movement in how organizations construct technological change. These findings suggest that digital transformation cannot be treated as a fixed analytical category. The challenge is empirical: the phenomenon itself does not hold still, and definitional refinement cannot stabilize what continues to reorganize across sectors and over time. We discuss what this means for construct validity, theoretical accumulation, and field maturation in IS research.
Recommended Citation
Svahn, Fredrik and Magnusson, Johan, "Digital Transformation as Discursive Drift: The Empirical Roots of the Construct Validity Problem" (2026). 17th Scandinavian Conference on Information Systems. 19.
https://aisel.aisnet.org/scis2026/19