Metrics Monitoring

Use for metric reports, service health and anomaly investigation. Establish the time window, units, baseline and data source. Compare trends, check missing samples and distinguish correlation from a cause. Rank plausible causes and propose the smallest validation and remediation steps. Output the observed change, impact, evidence and next check. Example: investigate a latency increase alongside error rate and deployment time. Work from supplied or connected data; do not claim continuous monitoring or create alerts without the user request and supported tooling.