Skip to content

Universal Defence Framework

While each malware type requires specific countermeasures, the following seven principles form a universal foundation for organisational cyber defence.


The majority of successful attacks exploit known, patched vulnerabilities.

PrincipleAction
SpeedCritical patches within 72h (internet-facing: 24h)
PriorityInternet-facing → internal servers → endpoints
AutomationWSUS/SCCM/Intune for endpoints; Ansible/Terraform for infra
TestingStaged rollout: canary → pilot → production
ExceptionsDocumented, time-bounded, compensating controls

Metric: % of critical patches applied within SLA


Every user, system, and process should have only the minimum permissions necessary.

LayerPrinciple
UsersStandard accounts; admin via PAM/JIT
Service AccountsDedicated, scoped, rotated
ApplicationsNo admin; constrained language modes
NetworkMicro-segmentation; default deny
CloudIAM least privilege; boundary policies

Metric: % of accounts with admin rights; % of overly permissive roles


Layer multiple, complementary controls — no single control is infallible.

LayerControls
PerimeterFirewall, WAF, DDoS protection, DNS filtering
NetworkSegmentation, IDS/IPS, Zeek, NetFlow, TLS inspection
EndpointEDR, AV, application control, memory protection
IdentityMFA (FIDO2), PAM, JIT access, tiered admin
ApplicationSAST/DAST/IAST, WAF, RASP, SBOM
DataEncryption (at rest/transit), DLP, classification
HumanAwareness training, phishing simulations, reporting
RecoveryBackups, IR plan, tabletop exercises

Principle: Failure of one layer ≠ total compromise


Offline, tested, geographically distributed backups are the ultimate insurance.

RuleDescription
3-2-1 Rule3 copies, 2 media types, 1 offsite/offline
ImmutabilityWORM, Object Lock, legal hold
TestingQuarterly full restore test; monthly partial
RTO/RPODefined per tier; documented
Air-gapPhysical or logical (immutable bucket)
EncryptionAES-256; keys separate from backups

Metric: Restore test success rate; RTO/RPO compliance


Human beings remain the most targeted entry point for malware.

ProgrammeFrequencyContent
OnboardingDay 1Policy, reporting, phishing basics
Quarterly4x/yearPhishing, social engineering, safe computing
Phishing SimsMonthlyRealistic templates; no punishment
Role-basedAnnualDev: secure coding; Admin: PAM; Finance: BEC
ReportingContinuous”See something, say something” culture

Metric: Phish click rate; reporting rate; training completion


Never trust, always verify — threats exist both inside and outside.

PillarImplementation
Verify IdentityFIDO2 MFA, certificate-based auth
Verify DeviceCompliance posture, health attestation
Verify NetworkMicro-segmentation, encryption everywhere
Least PrivilegeJIT, PAM, tiered admin, zero standing access
Assume BreachMonitor everything; encrypt all traffic
Continuous VerificationContinuous auth, risk-based auth

Maturity Model: Traditional → Advanced → Optimal (CISA ZTMM)


The ability to rapidly detect, contain, eradicate, and recover minimises damage.

PhaseKey Actions
PrepareIR plan, playbooks, tooling, team, comms plan
DetectSIEM, EDR, NDR, threat intel, hunt team
ContainShort-term (isolate), long-term (rebuild)
EradicateRemove artefacts, rotate creds, patch root cause
RecoverRestore from immutable backup, validate, monitor
Lessons LearnedRoot cause, gap analysis, update playbooks
ExerciseFrequency
TabletopQuarterly
Red Team / Purple TeamSemi-annual
Full SimulationAnnual
Ransomware-specificSemi-annual

Metric: MTTR (Mean Time To Respond); Containment time; Recovery time


PhaseFocusTimeline
0 - BaselineAsset inventory, risk assessment, gap analysisMonth 1
1 - Quick WinsMFA, patching, backups, AV/EDR, awarenessMonths 2-3
2 - HardenNetwork seg, least priv, app control, monitoringMonths 4-6
3 - MatureZero Trust, hunt team, purple team, automationMonths 7-12
4 - OptimiseAutomation, AI/ML detection, continuous improvementOngoing