Alert Correlation
Connect related alerts to understand the full attack story.
Alert Correlation
Correlation is the process of connecting related alerts to see the complete picture of an attack. A single alert rarely tells the full story.
Why Correlation Matters
Attackers don't operate in isolation:
Phase 1: Phishing email → User clicks link
Phase 2: Malware downloads → XDR detects process
Phase 3: C2 beacon → Firewall sees connection
Phase 4: Lateral movement → SIEM logs authentication
Phase 5: Data theft → Firewall sees large uploadEach phase generates alerts. Correlation connects them.
Correlation Indicators
IP Address
Same source or destination IP across alerts:
- External IP contacting multiple hosts
- Internal IP making unusual connections
- Known bad IP appearing in multiple logs
User Account
Same user in multiple suspicious events:
- Failed logins → successful login → privilege escalation
- Normal user suddenly accessing sensitive systems
- Service account used interactively
Hostname/Endpoint
Multiple alerts from the same system:
- Process execution → file creation → network connection
- Login → process → data access chain
- Endpoint showing signs of compromise
Time Window
Events occurring close together:
- Alerts within minutes of each other
- Sequence suggesting attack progression
- Burst of activity followed by quiet
IOCs (Indicators of Compromise)
Shared technical indicators:
- Same file hash across endpoints
- Common domain in network logs
- Matching process names or paths
Correlation Groups
When alerts are correlated, they form a correlation group:
| Field | Description |
|---|---|
| Group ID | Unique identifier |
| Alert Count | Number of related alerts |
| Severity | Highest severity in group |
| Timeline | First to last alert |
| Common IOCs | Shared indicators |
Using Correlation in SOC Simulator
SIEM View
- Click on an alert
- Check the "Related Alerts" section
- View the correlation timeline
- See shared IOCs
Correlation Timeline
Visual representation showing:
- When each alert occurred
- How alerts relate to each other
- The attack progression
Correlation Graph
Network-style view showing:
- Alerts as nodes
- Connections between them
- Central indicators (IPs, users, etc.)
Tips for Better Correlation
When you see one suspicious alert, always ask: "What else should I see if this is real?"
- Pivot on IOCs - Search for IPs, users, hashes across all logs
- Expand timeframe - Look 30 minutes before and after
- Check all tools - SIEM might miss what XDR catches
- Follow the chain - Initial access → execution → persistence → etc.
Correlation Scoring
SOC Simulator scores your correlation ability:
| Metric | Description |
|---|---|
| Groups Found | Correlation groups you identified |
| Completeness | Did you find all related alerts? |
| Speed | How quickly you connected the dots |
| Accuracy | Were your correlations correct? |
Practice Exercise
When you see an alert:
- Note the key IOCs (IP, user, hostname, hash)
- Search for each IOC in other alerts
- Build a timeline of events
- Determine if it's isolated or part of a chain
- Assess the full scope before making a triage decision