Overview#
Analytics that flag unusual behavior in systems, users, or network traffic. Relies on baselines, statistical models, or machine learning to highlight potential threats or misconfigurations.
Core objectives#
- Establish shared definitions of Anomaly Detection for security, engineering, and leadership teams.
- Connect Anomaly Detection activities to measurable risk reduction and resilience goals.
- Provide onboarding notes so new team members can quickly understand how Anomaly Detection works here.
Implementation notes#
- Identify the primary owner for Anomaly Detection, the data sources involved, and the systems affected.
- Document the minimum viable process, tooling, and runbooks that keep Anomaly Detection healthy.
- Map Anomaly Detection practices to standards such as ISO/IEC 27001, NIST CSF, or CIS Controls.
Operational signals#
- Leading indicators: early warnings that Anomaly Detection might degrade (e.g., backlog growth, noisy alerts, or missed SLAs).
- Lagging indicators: realized impact that shows Anomaly Detection failed or needs investment (e.g., incidents, audit findings).
- Feedback loops: retrospectives and metrics reviews that tune Anomaly Detection continuously.
Related practices#
- Align Anomaly Detection with defense-in-depth planning, threat modeling, and disaster recovery tests.
- Communicate updates to stakeholders through concise briefs, dashboards, and internal FAQs.
- Pair Anomaly Detection improvements with tabletop exercises to validate expectations.