A tabletop exercise that ends with a hot wash and no written after action report usually changes nothing: the same gaps reappear in the next exercise, or in a real incident. The after action report (AAR) and its improvement plan are where observations become owned, dated corrective actions. This template follows the incident response phases your plan uses, so each finding lands against the part of the plan it tests. It draws on the evaluation approach in the US Homeland Security Exercise and Evaluation Program (HSEEP) and CISA's tabletop exercise packages, adapted for cyber incidents. It is for exercise planners, facilitators, evaluators and the security leaders who must close the actions.
How to use this template
- Plan the report before the exercise: agree objectives and the evaluation criteria for each one during planning, because HSEEP treats evaluation as starting in planning and running through improvement planning. See the Tabletop Exercise Guide for design and facilitation.
- Assign evaluators and a note taker: evaluators observe without playing. They capture plan activation, roles and authorities, key decisions and information sharing, under a non-attribution rule.
- Collect three inputs: evaluator notes, the hot wash held immediately after play, and participant feedback forms. CISA's packages include a feedback form and an AAR template you can use alongside this one.
- Write findings, not minutes: each observation follows the format below. Do not record who said what.
- Issue a draft within [10] working days: hold an after action meeting with the sponsor to agree findings and owners, then publish the final AAR.
- Replace bracketed text: [Organisation], [Exercise sponsor] and other placeholders. Mark the report with your classification, for example [Internal: restricted], because it describes your weaknesses.
Exercise summary (complete first)
| Field | Entry |
|---|---|
| Exercise name and date | [Name], [date], [duration] |
| Sponsor and facilitator | [Exercise sponsor], [Facilitator] |
| Scenario summary | [Two or three sentences, for example "Ransomware on file servers after VPN credential theft; attacker also holds a cloud API key"] |
| Participating functions | [Security operations, IT, legal, privacy, communications, executives, suppliers] |
| Plans and playbooks tested | [IR plan v3.2, ransomware playbook v1.4, communication plan v2.0] |
| Objectives | [3 to 5 objectives, each measurable, for example "Decide on regulator notification within the 72-hour window using the plan's criteria"] |
| Overall assessment | [One paragraph: top strengths, top three gaps] |
Inject and decision record
Reconstruct the exercise as a short timeline. HSEEP lists event reconstruction among its analysis methods, and a timeline makes it obvious where decisions stalled.
| Exercise time | Inject | Expected action (per plan) | Actual action | Time to decide | Finding ID |
|---|---|---|---|---|---|
| T+00 | EDR alert: encryption activity on two file servers | SOC triages, notifies incident commander within [15] minutes | [Actual] | [Minutes] | [D-01] |
| T+30 | Threat actor email demands payment, claims data theft | Declare incident at [severity], engage legal and IR retainer | [Actual] | [Minutes] | [R-02] |
| T+60 | Cloud audit log shows API key used from unknown network | Revoke key, identify all systems using it | [Actual] | [Minutes] | [C-02] |
| T+90 | Journalist asks for comment | Holding statement issued by communications lead | [Actual] | [Minutes] | [N-01] |
Observation format (use for every finding)
- ID and phase: for example C-03 (Containment, finding 3).
- Objective: which objective it relates to.
- Type: strength or area for improvement.
- Observation: what happened, in one or two factual sentences.
- Analysis: what should have happened under the plan, whether there was a difference, its impact, and whether the plan or procedure supports the activity at all. These are the questions HSEEP evaluators ask.
- Root cause category: plan, people (training, roles), process, technology or third party.
- Rating: on a four-point scale, for example performed without challenges, with some challenges, with major challenges, unable to be performed.
- Recommendation: the change that would close it.
1. Scope and triggers
Assess how the exercise team recognised that the scenario was an incident and invoked the plan.
- Was the incident declared against documented criteria, and by someone with authority to declare?
- Was severity set using the matrix, and re-assessed as injects escalated? See the Incident Severity Classification Matrix Template.
- Did participants know which plan or playbook applied, and could they find the current version?
- Were the systems, data and identities in scope (including service accounts, API keys and third-party integrations) identified early?
[Findings S-01 to S-nn in the observation format.]
2. Roles and contacts
- Was a single incident commander named, and was that clear to everyone? See the Incident Commander Role Guide.
- Were decision makers and deputies available, and were their delegated authorities clear (for example who can take a revenue-generating service offline)?
- Did contact details work, with at least two methods for key contacts, including out-of-hours and out-of-band options? See the Incident Response Contact and Call Tree Template.
- Were legal, privacy, communications, HR, insurer and IR retainer engaged at the right time?
[Findings R-01 to R-nn.]
3. Detection and triage
- Could the team say which logs and telemetry would confirm the scenario, and were they confident those sources exist and are retained long enough?
- Was evidence preservation considered before containment? See the Forensic Readiness Guide.
- Was a decision log started, with times and rationale? See the Incident Log and Decision Record Template.
[Findings D-01 to D-nn.]
4. Containment
- Were containment options and their business impact understood (isolate hosts, disable accounts, revoke tokens, rotate keys, block egress)?
- Could the team actually revoke machine credentials quickly, and did they know where those credentials are used? A gap here is common in scenarios involving a stolen API key or OAuth token.
- Who approved containment that affected customers, and how long did the decision take?
[Findings C-01 to C-nn.]
5. Eradication
- Did the team plan how to find and remove persistence, and how to confirm the environment was clean?
- Was a credential reset scope agreed for people and non-human identities, including privileged and directory service accounts?
- Were dependencies on third parties or the IR retainer for this phase realistic?
[Findings E-01 to E-nn.]
6. Recovery
- Did the team agree restoration order, and does it match the business impact analysis and recovery time objectives?
- Were backups assumed to be clean, or did the team describe how to verify them?
- Was there a decision point and approver for returning to normal operations?
- Did the team plan heightened monitoring after restoration, and say for how long, in case the attacker kept access through an account or key that was missed?
- Were manual workarounds for critical business processes known, tested and owned by the business rather than by IT?
[Findings V-01 to V-nn.]
7. Communication and notification
- Were regulatory clocks identified and timed correctly (for example UK GDPR 72 hours from awareness, NIS2 early warning within 24 hours of becoming aware of a significant incident, SEC Item 1.05 four business days after a materiality determination, where they apply)? See the Cyber Incident Reporting Requirements Map.
- Were internal updates regular, with a set cadence, and did staff hear before the media?
- Were holding statements ready and approved, and was there a channel if email or chat was compromised?
- Were customers, suppliers and insurers informed in line with contracts?
[Findings N-01 to N-nn.]
8. Post-incident review
This section is the improvement plan. Every area for improvement becomes at least one corrective action, written to be specific, measurable, achievable, relevant and time-bound (HSEEP's SMART guidance).
| Action ID | Finding | Corrective action | Capability element | Owner | Due | Evidence of completion | Status |
|---|---|---|---|---|---|---|---|
| IP-01 | C-02 | Document and test API key revocation for the top 20 integrations | Process | [IAM lead] | [Date] | Runbook link, test record | Open |
| IP-02 | R-01 | Add deputy incident commanders to the rota and train them | People | [Head of SOC] | [Date] | Rota, training log | Open |
| IP-03 | N-03 | Pre-approve three holding statements with legal | Plan | [Comms lead] | [Date] | Approved statements | Open |
- Strengths to keep: list what worked, so it is not lost in the next plan revision.
- Re-test: name the next exercise that will check the high-priority actions. See the Tabletop Exercise Scenario Pack.
- Sign-off: [Exercise sponsor], [date].
Implementation guidance
- Agree the template with the sponsor first: the sponsor commits to owning the improvement plan before the exercise is scheduled.
- Use it for every exercise: including short scenario drills, so findings are comparable over time.
- Track actions where other risk work is tracked: the risk register, plan of action and milestones, or the security programme backlog, not a standalone document.
- Escalate overdue actions: report open and overdue corrective actions to the risk or audit committee quarterly.
- Feed real incidents in: the same observation format works for real incident reviews. See the Post-Incident Review and Root Cause Analysis Template.
Scaling the report
- Small organisations: one facilitator can also evaluate. Keep the summary, the observation list and the improvement plan; skip the inject table if the exercise ran under an hour. Five well-owned actions beat twenty vague ones.
- Large or regulated organisations: use dedicated evaluators per function, keep evaluator notes as working papers, and retain the AAR as evidence of testing for auditors and regulators. Exercises with suppliers should share the relevant findings with them under agreed terms.
- Executive exercises: focus findings on decision authority, notification judgement and communications rather than technical steps.
Where identity fits
Many cyber scenarios hinge on identities: a compromised administrator, a help desk reset, a leaked API key or an over-privileged OAuth integration. Score separately whether the team could identify which human and non-human identities were affected, who owns them, and how fast they could be disabled or rotated. See the Identity Incident Response Playbook.
Measures that show it is working
- Time from exercise to final AAR (target under 15 working days).
- Percentage of areas for improvement with an owned, dated corrective action (target 100 per cent).
- Percentage of corrective actions closed by due date.
- Repeat findings: the same gap recorded in two consecutive exercises (target zero).
- Share of exercise objectives rated as performed without challenges, trended year on year.
Common failure patterns
- Minutes instead of findings: long narratives with no analysis or recommendation.
- Blame in the report: naming individuals ends candid participation in future exercises.
- Actions with no owner: "improve communication" assigned to "IT".
- No re-test: actions marked closed without evidence that they work.
Framework mapping
| Framework | Reference | How this template supports it |
|---|---|---|
| NIST CSF 2.0 | ID.IM-02 | Improvements identified from security tests and exercises |
| NIST CSF 2.0 | ID.IM-04 | Incident response plans maintained and improved |
| NIST SP 800-53 Rev. 5 | IR-3, CA-5 | Incident response testing; plan of action and milestones for corrective actions |
| ISO/IEC 27001:2022 Annex A | 5.24, 5.27 | Incident management planning and preparation; learning from incidents |
| CIS Controls v8.1 | Control 17 | Incident response management, including routine exercises |
Standards and references
- FEMA: Homeland Security Exercise and Evaluation Program (HSEEP) 2020
- CISA: Tabletop Exercise Packages
- NIST: SP 800-84, Guide to Test, Training, and Exercise Programs for IT Plans and Capabilities
- NIST: SP 800-61 Rev. 3, Incident Response Recommendations and Considerations
- NIST: Cybersecurity Framework 2.0
- NCSC: Your cyber incident response processes
Related NHI Mgmt Group resources: Tabletop Exercise Guide · Incident Response Plan Template · Incident Response Maturity Model · Top 10 Incident Response Mistakes