An incident response plan is judged on the morning it is used, not on the day it is approved. The document can be complete and still fail because nobody knows who may shut down the payment gateway, the supplier contact is a general support line, or the evidence the insurer needs was never collected. This checklist is designed to expose those gaps before an incident does. It is written for an SME where every task has to be assigned to someone who already has another job.
How to use the checklist
Mark each line Ready, Partial or Gap. Ready means the item exists, has a named owner and was tested in the last twelve months. Partial means it is written down but untested, out of date or owned by a vacant role. Gap means nothing usable. Record the artefact, its location, the last check and the accountable person.
Give each section a sign-off. The plan owner is usually the managing director or chief operating officer; the incident commander should be someone else with authority to act when the owner is unreachable. Review contacts quarterly, the full plan annually, and every section after an incident or a material change.
1. Authority and scope
- The plan names an owner, a deputy and an incident commander.
- In scope: systems, data, premises, cloud accounts and suppliers covered by the plan.
- Out of scope: systems and entities not covered, stated explicitly so nobody assumes protection that does not exist.
- A pre-approved emergency spend limit, and the named person who can approve spending above it.
- Written authority to isolate a system, disable an account, block a payment or take a site offline, usable at 2am without a committee.
- An insurance notification owner, with the policy number and incident line stored offline.
- A named legal or privacy adviser for decisions that affect customer data or regulatory notification.
Section sign-off: managing director. Evidence: signed authority page and current contact card. Frequency: annual, or when a role changes.
2. Severity and decision rights
Severity should describe the consequence to the business, not the technology label. Test the examples against three real scenarios; a missing laptop and a compromised payment gateway are not the same event.
| Level | Example | Who decides | Update rhythm |
|---|---|---|---|
| Critical | Payments, fulfilment or customer data unavailable or exposed across the business. | Incident commander, with the managing director informed. | Every hour until contained. |
| High | A single site, key supplier or large customer segment is affected. | Incident commander. | Twice daily. |
| Medium | A limited group of users or a non-critical system is affected. | Functional lead, reported to the commander. | Daily. |
| Low | A single device, account or user, with no evidence of spread. | IT or operations lead. | At closure. |
The plan should also state who may communicate externally, who may instruct staff to stop work, and which decisions must wait for the insurer or legal adviser.
3. Contact and dependency register
A contact list stored only inside the systems under attack is useless. Keep a printed copy and an offline copy with the incident commander.
- Internal tree with at least one alternate for every role.
- Critical suppliers, with account or merchant identifiers, contract references and the escalation path beyond first-line support.
- Bank, card acquirer, mobile financial service provider and payment gateway contacts for each channel you accept.
- Hosting, domain registrar, email provider and website support.
- Insurer, legal counsel, forensic provider and public relations contact.
- Regulator or industry body, where notification is required.
- Landlord, building security and utility contacts for premises incidents.
Test the list by asking each owner to confirm the details. A quarterly call to the payment provider is more useful than a spreadsheet row that has never been dialled.
4. Playbooks that fit on one page
Each playbook should state the first actions, what not to do, the evidence to preserve and the person in charge. Keep it short enough to use under pressure.
| Scenario | First actions | Evidence to preserve |
|---|---|---|
| Ransomware or file-encrypting malware | Isolate affected devices from the network, disable shared accounts, do not pay, contact the incident commander and insurer. | Affected device images, ransom note, login history, backup logs and the timeline of first symptoms. |
| Business email compromise | Reset credentials, revoke sessions, check mailbox rules and payment changes, warn the finance team. | Email headers, mailbox audit logs, forwarding rules and any changed bank details. |
| Customer data leak | Stop the exposure, preserve logs, involve legal and privacy advisers before public statements. | Access logs, export records, affected data fields and the scope of exposure. |
| Payment fraud or chargeback spike | Pause the affected channel if authorised, contact the acquirer, pull transaction and delivery evidence. | Order records, delivery proof, customer communications and provider case numbers. |
| Website defacement or malware | Take a snapshot, restrict access, remove the page from service and check for persistence. | Web server logs, file hashes, deployment history and the original notice. |
| Lost or stolen device | Remote wipe if possible, revoke tokens, change passwords used on the device. | Device serial, last login, data classification and wipe confirmation. |
5. Evidence, communication and recovery
- An evidence log records time, source, handler, storage location and a hash where practical.
- A communications lead drafts holding statements in Bangla and English, with an approval chain and a single spokesperson.
- Customer, regulator and supplier notification criteria are written down, with the person who decides whether a threshold has been met.
- Recovery follows a documented order: identity, network, critical applications, data restore and customer-facing services.
- Return to service requires a technical check, a business check and a named approver.
- Credential rotation covers passwords, API keys, tokens, certificates and shared service accounts.
- A blameless review happens within five working days, with corrective actions, owners and due dates.
6. Exercise and maintenance schedule
Run a 90-minute tabletop every six months: confirm the scenario, walk through the first hour, test the contact tree, review evidence needs and record the gaps. Do not score the team on knowing every answer. Score the plan on whether it gave them a next step.
Test contacts and a backup restore quarterly, walk through one playbook annually, and run a fuller simulation annually where a critical system is at stake. Each exercise should produce a short record with participants, scenario, findings, actions and a sign-off.
One-page sign-off
| Section | Owner | Last reviewed | Status | Approver |
|---|---|---|---|---|
| Authority and scope | Managing director | Date | Ready / Partial / Gap | Board or owner |
| Severity and decisions | Incident commander | Date | Ready / Partial / Gap | Managing director |
| Contacts and dependencies | Operations lead | Date | Ready / Partial / Gap | Incident commander |
| Playbooks | IT lead | Date | Ready / Partial / Gap | Incident commander |
| Evidence and recovery | IT and finance | Date | Ready / Partial / Gap | Managing director |
| Exercise record | Plan owner | Date | Ready / Partial / Gap | Managing director |
A plan with three honest gaps and named owners is safer than a polished document with no evidence. If there are more gaps than the team can close, start with authority, contacts and one tested playbook. Those decide whether the first hour is controlled or chaotic.
If you need help turning the checklist into a tested plan, start with Managed Technology & Cybersecurity. The pricing page explains how assessment, plan development and exercises are scoped separately. For the budget side of the same decision, read the incident response plan cost guide.
Sources
- NIST, SP 800-61r3: Incident Response Recommendations and Considerations for Cybersecurity Risk Management, accessed 17 November 2026.
- NIST, Cybersecurity Framework 2.0, accessed 17 November 2026.
- CISA, Federal Incident and Vulnerability Response Playbooks, accessed 17 November 2026.
- UK National Cyber Security Centre, Incident Management, accessed 17 November 2026.
Responses (0)
No comments yet. Add the first one below.
Leave a response
Your email address will not be published. Required fields are marked *.