At a glance
- Owner
- StratQuad
- Version
- November 2026
- Applies to
- Any organisation responding to suspected or confirmed ransomware
- Mandatory
- Guidance
Key points
- Contain without destroying evidence: isolate affected systems from the network; do not power them off or wipe them.
- Assume identity is compromised. Most ransomware operators hold domain or cloud administrator access before they encrypt.
- Protect backups and the identity platform first, because they are what you will recover with.
- Move coordination to communication channels the attacker cannot see.
- Start the regulatory, insurance and law enforcement clocks early; the data protection 72 hours often begins on day one.
Hour 0 to 1: contain and mobilise
- Declare an incident and name an incident lead with authority to take systems offline.
- Isolate, do not power off. Disconnect affected hosts from the network, or isolate them through endpoint detection and response. Memory and running processes hold evidence that is lost on shutdown.
- Cut attacker access paths: disable VPN and remote access for all but named responders, block known malicious infrastructure, and suspend accounts showing malicious activity.
- Protect backups: confirm backup systems and their credentials are separate from the compromised domain; take an offline or immutable copy out of reach if one is not already.
- Open an out of band channel: a phone bridge and messaging on devices and accounts not joined to the compromised environment.
- Call your insurer and incident response retainer. Many policies require their panel firm.
Practitioner note
The single most expensive early mistake is rebuilding or restoring before you know how the attacker got in and what they still hold. Restored systems rejoin a domain the attacker controls and are encrypted again. Contain first, scope second, recover third.
Hour 1 to 4: scope
- Identify patient zero and the access route: exposed remote access, exploited edge device, phishing, or stolen credentials, often bought from an initial access broker.
- Map encryption: which systems, which data, which business services.
- Look for exfiltration: large outbound transfers, archive tools, and file transfer utilities in the days before encryption. Most groups steal data before encrypting it.
- Identify the group from the ransom note, file extension and leak site. Known groups have documented tactics, decryptor availability and sanctions status.
- Assess identity: domain controllers, Entra ID or other identity providers, privileged accounts, and newly created accounts or persistence such as scheduled tasks and remote management tools.
Hour 4 to 24: decide and notify
- Notification decisions
Personal data breach assessment (72 hour clock); NIS or sector regulator duties; NCSC; Report Fraud or Police Scotland; DSPT for health and care; contractual notifications. See Reporting a cyber incident in the UK.
- Identity recovery
Reset privileged credentials from a clean device, reset the Kerberos KRBTGT account password twice, waiting at least the maximum ticket lifetime (10 hours by default) between resets, revoke cloud sessions and tokens, and review conditional access and app registrations.
- Recovery plan
Prioritise systems by business impact, rebuild from known good images, restore data only from backups confirmed to predate the compromise, and monitor rebuilt systems closely.
- Communications
A holding statement for staff, customers and the public; one spokesperson; no speculation about attribution or data loss until confirmed.
- Payment position
Record the board's position early. Check sanctions exposure with legal advisers; paying a sanctioned person is an offence. The NCSC and National Crime Agency advise against payment, and payment does not guarantee deletion of stolen data.
Evidence expected
Preserve: memory captures and disk images from representative affected hosts; firewall, VPN, identity provider, EDR and email logs, exported before retention expires; the ransom note and any communications; and a time stamped decision log. Insurers, regulators and any later legal proceedings rely on these.
Common mistakes
Common mistakes
- Powering off servers, losing volatile evidence.
- Discussing the response on email or chat inside the compromised tenant.
- Restoring before the access route is closed, leading to re encryption.
- Resetting user passwords while leaving attacker controlled privileged accounts, tokens and persistence in place.
- Treating "no data leaked yet" as "no data stolen".
Sources
- NCSC. Mitigating malware and ransomware attacks (opens in a new tab)
- NCSC. Incident management (opens in a new tab)
- CISA. #StopRansomware Guide (opens in a new tab)
- Office of Financial Sanctions Implementation. Financial sanctions guidance for ransomware (opens in a new tab)