The Incident Response Lifecycle
prepare, detect, contain, eradicate, recover, learn
An incident is an event that actually or potentially harms the confidentiality, integrity or availability of information or systems. Incident response is the planned, practiced way of dealing with it, so the worst moment of the year does not also become the most chaotic one.
The six phases
Frameworks word these slightly differently. NIST's guidance has recently been realigned with the CSF 2.0 functions (govern, identify, protect, detect, respond, recover).[1] This six-step version is a practical way to remember the flow:
- Prepare. Before anything happens: a written plan, roles and contacts, tools and logging in place, access for responders, playbooks, backups and regular practice exercises. Preparation decides how the rest goes.
- Detect and analyze. Spot signs of trouble (alerts, user reports, outside notification), decide whether it is a real incident, determine scope and severity, and start a timeline and evidence log.
- Contain. Stop it spreading and limit damage. Short-term actions such as isolating a host or disabling an account; longer-term ones such as network changes while you prepare to clean up. Balance speed against evidence: do not destroy the data you need to understand the attack.
- Eradicate. Remove the cause: malware, attacker access, stolen credentials, the exploited weakness.
- Recover. Restore systems from clean sources, verify they work, monitor closely for signs the attacker returns, and return to normal operation in a controlled way.
- Learn. A blameless post-incident review: what happened, what worked, what did not, and which fixes will happen, by whom and when.
In real incidents phases overlap and loop back: new findings during recovery can send you back to containment.
Roles that need to be clear
An incident commander who coordinates; technical leads; communications (internal, customers, media); legal and privacy (notification duties); HR when staff are involved; executives for decisions such as shutting down a service or paying costs. Know who can authorize drastic action at 3 a.m.
Good habits
- Keep a log of decisions and actions with times.
- Communicate on a separate channel if the attacker may be reading your email and chat.
- Preserve evidence and follow chain of custody if legal action is possible.
- Classify severity consistently so response matches impact.
- Meet notification deadlines for regulators and contracts. Know them in advance.
- Practice with tabletop exercises and simulated alerts.
Measures
Time to detect, time to contain, time to recover, and how many incidents repeat a known cause. Improvement here shows the program is working.
Sources
Test yourself
My notes
Private — saved only on this device, never sent anywhere.