Problem-Oriented Process Mining for Auditable Marketing Automation Lifecycle Control

摘要

Problem-oriented process mining (POPM) introduces a systematic framework for diagnosing, monitoring, and optimizing marketing automation workflows through data-driven process intelligence. Unlike conventional automation analytics that emphasize campaign performance metrics, POPM emphasizes the detectability and traceability of procedural deviations across the marketing lifecycle—from lead acquisition and nurturing to conversion and retention. By integrating event logs, control-flow discovery, and conformance checking, POPM enables organizations to identify compliance breaches, detect process bottlenecks, and validate automation decisions against predefined business rules. The auditable nature of POPM allows for transparency in algorithmic actions, ensuring that automated marketing decisions remain explainable and aligned with regulatory and ethical requirements. This paper reviews state-of-the-art applications of POPM in marketing automation systems, examining frameworks that combine process mining with customer journey analytics, robotic process automation (RPA), and explainable AI. It further proposes an auditable lifecycle control model that integrates root-cause analytics, rule-based auditing, and feedback mechanisms for continuous improvement. The study concludes that problem-oriented process mining not only strengthens accountability and governance in marketing automation but also establishes a foundation for trustworthy, adaptive, and regulation-compliant automation ecosystems.