CMDB Data Governance and Business Continuity: Identifying Research Gaps in AI-Driven IT Operations

Abstract

The Configuration Management Database (CMDB) is widely positioned as the authoritative record of IT infrastructure and its relationships to business services, yet industry surveys consistently report accuracy levels around 60 percent and find that a majority of CMDB implementations fail to meet their intended objectives. This gap between the CMDB's intended role and its practical reliability has been studied from several angles: general data governance frameworks, machine learning approaches to configuration data quality, and, increasingly, artificial intelligence integration within IT service management platforms. This paper reviews that literature and argues that one consequential intersection remains comparatively under-examined: the relationship between CMDB data governance and Business Continuity and Disaster Recovery (BCDR) outcomes, particularly as AI-driven and increasingly agentic IT operations begin to act on CMDB data with reduced human review. Drawing on recent work in CMDB-focused machine learning, AI integration in IT service management, general data governance frameworks, business continuity and cyber-resilience research, and the longer-standing literature on trust in automation, this review identifies a structural gap: existing work addresses CMDB data quality in general terms, or business continuity in general terms, but rarely connects specific CMDB governance dimensions to specific continuity and recovery outcomes. This is a targeted, narrative literature review rather than a systematic one, and its contribution is a research agenda rather than a new empirical study: an explicit characterization of this gap, expressed as a set of governance dimensions and framework requirements capable of addressing it, intended as the basis for continued research.