Data Governance for Big Data Business Intelligence: A Conceptual Framework and Research Agenda

Organizations increasingly depend on big data business intelligence (BI) to support operational and strategic decisions. Yet the realized value of BI remains uneven. Practitioners frequently attribute shortfalls not to data scarcity or analytical capability but to governance failures: unclear data ownership, inconsistent metric definitions, weak lineage, slow access approvals, and uncontrolled proliferation of shadow analytics. Big data amplifies these problems through continuous pipeline change, semantic heterogeneity, decentralized ownership, and the embedding of algorithmic outputs into decision workflows. Prior research offers governance principles and reference models, yet the field lacks an outcome-centered conceptual framework that specifies how governance mechanisms translate into BI performance under big data conditions. This paper addresses that gap. We propose a mechanism-based conceptual framework linking governance mechanisms (structural, process, technical) to BI outcomes (trustworthiness, decision quality, agility, risk/compliance) through four intermediate governance capabilities: data quality assurance, lineage and traceability, access agility with control, and analytical accountability. We develop six testable research propositions (P1–P6) specifying directional relationships and theorized boundary conditions including decentralization and regulatory intensity. We conclude with a focused research agenda comprising operationalization guidance, study designs, and units of analysis to support cumulative empirical inquiry. The framework contributes a diagnostic lens for practitioners and a theoretically grounded foundation for hypothesis testing in big data BI governance research.