liveBook · Manning

Severity

As explained in Azure Security

Alert severity is a classification indicating the urgency and assessed likelihood or impact of a security alert.

Alert severity classifies Microsoft Defender for Cloud alerts by investigation priority. The Security alerts dashboard uses severity to help decide what to investigate first: high alerts indicate high confidence of compromise, medium alerts indicate suspicious activity, low alerts can be benign positives or blocked attacks, and informational alerts may matter when considered with an incident’s other alerts. section 7.3.1

Use the severity levels as an initial triage signal:

  • High: investigate with the highest priority, such as when a known credential-theft tool is detected. section 7.3.1
  • Medium: assess suspicious activity that might indicate compromise, such as a sign-in from an unusual location. section 7.3.1
  • Low: review in context; it can represent a benign positive or blocked attack, including routine log clearing. section 7.3.1
  • Informational: consider alongside related alerts in a security incident before deciding whether further investigation is warranted. section 7.3.1

Severity is a prioritization aid, not the only signal. Where available, MITRE ATT&CK mappings identify associated tactics and techniques, helping clarify adversary intent and refine the investigation order. section 7.3.1

Route medium-or-higher alerts

Configure email notifications for the relevant subscription so Contributor-role recipients and specified addresses receive alerts at Medium severity or higher. The portal also supports High and Low thresholds; use the threshold that matches the team’s response capacity. section 7.2.9

Figure 7.6 The Security Alerts dashboard in Microsoft Defender for Cloud

Use the Security alerts dashboard to view alerts and begin triage according to their default severity-based priority. section 7.3.1

How Azure Security covers it

How deep it goes
Defines four alert levels with threat examples and relates them to dashboard triage; it does not prescribe full investigation procedures.
Examples
Uses credential-theft tooling, unusual-location sign-ins, routine log clearing, and medium-or-higher email routing.
What you can do after
Operate/configure alert notification thresholds and evaluate which Defender for Cloud alerts warrant investigation first.
How it connects
Connects alert triage to MITRE ATT&CK, security incidents, email notifications, and Azure RBAC roles.

Also explained in

Compare how 7 books explain severity →

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Azure Security. The section links point to the parts of the book it draws on.