WHY CHANGE FAILS IN IT: A MULTIDIMENSIONAL DIAGNOSIS OF SYSTEMIC FAILURE PATTERNS AND ROOT CAUSES

Organizational change in IT companies continues to fail at alarming rates, with reported failure frequencies ranging from 60% to 80% – despite decades of evolving change management methodologies. This study argues that the core problem is not methodological insufficiency, but categorical misdiagnosis. Most frameworks – both traditional (e.g., Lewin, Kotter, ADKAR) and modern (e.g., Agile, SAFe, DevOps) – treat transformation as a technical process that can be managed through planning, communication, and tool deployment. In contrast, this paper presents change as an adaptive challenge that threatens identity, disrupts culture, and exposes unresolved organizational paradoxes. Using a hybrid methodology, the study combines meta-analysis of empirical failure data from leading industry sources (McKinsey, Gartner, Standish Group, Scrum Inc., Radixweb) with theoretical triangulation from identity theory (Kegan & Lahey), paradox theory (Smith & Lewis), and the Adaptive Change Model (Heifetz et al.). The research identifies five primary causal clusters that drive failure: cognitive misalignment, identity-based resistance, cultural-systemic misfit, structural incoherence, and execution breakdown. Failures are further analyzed across five dimensions: organizational level (executive, middle, team), type of change (process, cultural, structural, digital), implementation phase (initiation, transition, sustainment), failure visibility (overt, covert, latent), and root-cause clustering. Findings show that change most often fails when it is treated as a finite project rather than a sustained shift in organizational identity, norms, and rhythms. The study concludes by calling for a new generation of transformation models – ones that embed change into lived identity, tolerate paradox, and reinforce new behaviors through long-cycle rhythm. This paper lays the empirical and conceptual groundwork for such a model, currently under development, which translates adaptive theory into operational, identity-centered architecture.