Should organisations prioritise preparation or response in cyber resilience programmes?

Why This Matters for Security Teams

Preparation is the part of cyber resilience that determines whether an event becomes a controlled disruption or a full-blown business incident. Response teams can contain damage, but only if identity, logging, backup, segmentation, and recovery paths were designed before pressure arrives. This is especially true when adversaries use stolen credentials, supplier access, or automation to move faster than human response cycles. NIST control guidance in NIST SP 800-53 Rev 5 Security and Privacy Controls remains a useful reference point for that preparation layer.

Teams often overrate incident playbooks and underrate the operational decisions that happen earlier, such as hardening privileged access, segmenting critical services, and making restoration possible without reintroducing the same compromise. Preparation also reduces confusion during an event because ownership, escalation, and recovery criteria are already agreed. In practice, many security teams encounter the limits of response only after identity compromise or lateral movement has already spread across the environment, rather than through intentional resilience testing.

How It Works in Practice

Effective resilience programmes treat preparation and response as a sequence, not a choice. Preparation establishes the conditions that make response viable: asset visibility, attack surface reduction, tested backups, crisis communications, and access controls that limit blast radius. Response then executes against those conditions through containment, triage, eradication, and recovery. Where identity is involved, that means tightening privileged access, rotating exposed secrets, and knowing which human and non-human identities can reach critical systems.

A useful operating model is to prioritise the following preparation activities:

• Map business-critical services to recovery time and recovery point objectives.
• Maintain current asset, identity, and dependency inventories.
• Test backup restore paths, not just backup completion.
• Align threat intelligence with exposure management and access reviews.
• Pre-stage containment actions for high-risk scenarios such as credential theft or ransomware.

That preparation should be informed by live threat analysis, not static annual planning. Sources such as CISA cyber threat advisories and the ENISA Threat Landscape help security teams translate current attacker behavior into concrete control priorities. If the question involves AI-enabled operations, the same logic applies to model and agent governance: the Anthropic report on AI-orchestrated cyber espionage shows why automated misuse can compress response windows. These controls tend to break down when environments rely on sprawling legacy estates, unmanaged identities, and manual restoration steps because response actions cannot outrun the attacker’s access path.

Common Variations and Edge Cases

Tighter preparation often increases operational cost and change overhead, requiring organisations to balance resilience gains against delivery speed and budget constraints. That tradeoff becomes sharper in regulated environments, mergers, and complex cloud estates where ownership is fragmented and dependencies are poorly documented.

There is no universal standard for this yet, but current guidance suggests three common edge cases. First, in small teams, response may appear more urgent because staffing is limited; even then, the highest-value investment is usually a few preparation controls that prevent recurring incidents. Second, in high-availability environments, aggressive prevention can be misread as reduced resilience, so recovery testing must prove that protection does not create brittle failure modes. Third, where AI systems or autonomous agents participate in operational workflows, preparation must include prompt, tool, and access governance because malicious or malformed instructions can become an attack path. In that context, MITRE ATLAS adversarial AI threat matrix is useful for mapping how model abuse may affect resilience planning.

For most organisations, the practical answer is not to choose one over the other. Preparation should receive the larger share of investment, while response remains continuously exercised so that controls, people, and recovery paths work under pressure.

Related resources from NHI Mgmt Group

• Why do organisations need stronger incident response planning when cyber resilience regulation raises the bar?
• How should organisations design identity recovery for cyber incident response?
• What should organisations prioritise first in identity governance programmes?
• What should organisations prioritise first in IoT security programmes?