The automotive sector increasingly relies on consumer-generated data, such as customer complaints and market claims, to support quality assurance and post-market decision-making. However, the growing volume, complexity, and unstructured nature of consumer responses pose significant challenges to sustainable data governance, particularly in ensuring timely analysis, consistent prioritization, and accountable decision-making. This study proposes a machine learning-based predictive analytics framework to support sustainable data governance of consumer responses in an automotive manufacturing context. The framework integrates data preprocessing, feature learning, and feature selection to enhance data quality and governance readiness. Unsupervised clustering techniques are applied to identify latent patterns within consumer response data and to distinguish crisis-level issues that require immediate attention, while supervised classification models are developed to automate the categorization of incoming claims. The predictive models achieve an accuracy exceeding 90%, demonstrating their effectiveness in supporting reliable and scalable analytics. To enhance transparency and operational usability, the analytical outcomes are visualized through a business intelligence dashboard developed using Power BI, enabling real-time monitoring and cross-functional decision support. The results indicate that the proposed approach improves responsiveness, accountability, and sustainability in managing consumer response data. This study contributes to the adoption of data-driven governance practices in the automotive sector and supports Industry 4.0 initiatives through the integration of predictive analytics and business intelligence.