Scaling data management capabilities for enterprise AI: a maturity model for the banking industry

Abstract

Integrating artificial intelligence (AI) into financial institutions requires more than deploying technical solutions. It requires data management capabilities (DMCs) aligned with regulatory constraints and strategic priorities. This clinical study investigates how a German bank evolved from fragmented, Excel-based processes to a cloud-enabled data environment supporting AI enablement. Using a longitudinal embedded clinical case study, we trace the development of DMCs across three transformation phases and show how executives institutionalized interdependent DMCs through iterative cycles of experimentation, negotiation, and learning. We identify five mutually reinforcing capability domains: (1) Technology and Infrastructure, (2) Data Governance and Quality, (3) Cultural and Organizational Shifts, (4) Regulatory Compliance and Risk Management, and (5) AI Enablement. We contribute to IS research by (1) developing a grounded DMC maturity model that translates these insights into a practical diagnostic tool, (2) providing a process-based lens on cumulative DMC development that highlights sociocultural readiness and recurring managerial missteps as central mechanisms, explaining how readiness builds toward scaled AI enablement in regulated environments, and (3) extending clinical IS research by demonstrating how embedded inquiry can generate context-aware artifacts to support AI transformation. For practitioners, the maturity model provides diagnostic signals, minimum viable actions, and readiness evidence for scaling trustworthy AI.