This project explores improving Database Monitoring within Google Cloud Platform by redesigning how metrics are structured, surfaced, and understood inside a Metrics Dashboard. Observability is central to cloud infrastructure reliability, helping teams maintain availability, performance, and stability across complex systems. However, existing dashboards often bury critical signals, require excessive navigation, and increase cognitive load during troubleshooting. Using a structured design process, the project combined desk research, stakeholder interviews, user archetype mapping, and competitor analysis across AWS, Azure, and IBM Cloud. Key insights emphasized reducing clicks, improving metric discoverability, and supporting scalability as observability needs grow. Two design directions were evaluated: suggestive dashboards that proactively highlight anomalies, and a bucketing based model that groups metrics for flexible exploration. The bucketing approach was selected for its clarity, scalability, and balance between consistency and personalization. The final low fidelity prototype introduces hierarchical categories, collapsible sections, and vertical navigation for faster anomaly detection and health insights.