Vulnerability Management RACI Matrix Template one signed operating model for identification, triage, routing, remediation, exception, and governance accountability
A free, copy-ready vulnerability management RACI matrix template. Twelve structured sections covering header and document control, four-letter legend with the single-A rule, sixteen role column definitions, identification lifecycle rows, triage and validation rows, routing and ownership rows, remediation and verification rows, SLA breach and stop-the-clock rows, exception lifecycle rows with a full residual-band approver ladder, reporting and review rows, governance review with operational realism tests, and signatures with stakeholder acknowledgement. Aligned with ISO/IEC 27001 Clause 5.3 and Annex A 5.2, NIST SP 800-53 PM-2 and RA-5, NIST SP 800-40 Rev. 4 Section 2, PCI DSS Requirement 12.4 and 6.3, SOC 2 CC1.3 and CC7.1, HIPAA 45 CFR 164.308(a)(2), NIS2 Article 21, and DORA Article 5.
Run the matrix against the live finding record, not against a slide deck
SecPortal carries the routed Remediation Owner, the named Retest Owner, the residual-band exception approver, and the timestamped state changes on one workspace so RACI accountability operates as a byproduct of the work. Free plan available.
No credit card required. Free plan available forever.
Twelve sections that turn a vulnerability management programme into a defensible RACI matrix
A vulnerability management RACI matrix is the operating-model document a security or GRC function publishes to name, for each step of the vulnerability lifecycle, who is Responsible (does the work), Accountable (carries the outcome), Consulted (must be heard before the work proceeds), and Informed (has to be told once a decision lands). It sits one level below the umbrella vulnerability management policy and one level above any per-finding workflow. The twelve sections below cover the durable shape of the artefact across ISO/IEC 27001 Clause 5.3 and Annex A 5.2, NIST SP 800-53 PM-2 and RA-5, NIST SP 800-40 Rev. 4 Section 2, PCI DSS Requirement 12.4 and 6.3, SOC 2 CC1.3 and CC7.1, HIPAA 45 CFR 164.308(a)(2), NIS2 Article 21, and DORA Article 5. Copy the section that fits your stage and paste the rest as you go.
Copy the full matrix (all twelve sections) as one block.
1. Matrix header and document control
Open the RACI with version, effective date, owner, and the policy hierarchy it sits inside. The header is the artefact the audit reads first to confirm the matrix is a controlled document rather than a working draft. ISO/IEC 27001 Clause 7.5 expects documented information to carry version, owner, and review cadence; the RACI matrix is no exception.
Document title: Vulnerability Management RACI Matrix
Version: {{RACI_VERSION}}
Effective date: {{EFFECTIVE_DATE}}
Last review date: {{LAST_REVIEW_DATE}}
Next scheduled review date: {{NEXT_REVIEW_DATE}}
Document owner (maintains the matrix and schedules review):
- Name: {{DOCUMENT_OWNER_NAME}}
- Role: {{DOCUMENT_OWNER_ROLE}}
- Function: {{DOCUMENT_OWNER_FUNCTION}}
Approving authority (signs the matrix and material revisions):
- Name: {{APPROVING_AUTHORITY_NAME}}
- Role: {{APPROVING_AUTHORITY_ROLE}}
- Approval date: {{APPROVAL_DATE}}
Parent policy (referenced from this matrix):
- Vulnerability Management Policy, version {{VM_POLICY_VERSION}}, Section 3 (Roles and responsibilities).
Sibling artefacts (operate alongside this matrix):
- Vulnerability Remediation SLA Policy, version {{SLA_POLICY_VERSION}}.
- Security Exception Register Policy, version {{EXCEPTION_POLICY_VERSION}}.
- Risk Acceptance Policy, version {{RISK_ACCEPTANCE_POLICY_VERSION}}.
- Asset Criticality Classification Policy, version {{ASSET_CRITICALITY_POLICY_VERSION}}.
- Vulnerability Disclosure Policy, version {{VULN_DISCLOSURE_POLICY_VERSION}}.
Frameworks the matrix evidences (the audit reads accountability per lifecycle step against these references):
ISO/IEC 27001 Clause 5.3 and Annex A 5.2, NIST SP 800-53 PM-2 and RA-5, NIST SP 800-40 Rev. 4 Section 2, PCI DSS Requirement 12.4 and 6.3, SOC 2 CC1.3 and CC7.1, HIPAA 45 CFR 164.308(a)(2), NIS2 Article 21, DORA Article 5, internal vulnerability management policy.
Distribution at publication and revision:
- Heads of business units carrying remediation responsibility.
- Heads of platform and infrastructure teams.
- Engineering leadership.
- GRC, compliance, and internal audit functions.
- Procurement and vendor management (for supply-chain coverage).
- {{ADDITIONAL_STAKEHOLDERS}}
2. Legend and single-A rule
Publish the four-letter legend explicitly so reviewers do not have to assume what each letter means. State the single-A rule (exactly one Accountable role per row) at the top of the document because that single constraint is what separates a defensible RACI from a wish list. The audit reads the legend to test whether the matrix obeys the convention rather than reinventing it.
Legend:
R - Responsible
The role that does the work for the lifecycle step. R may be carried by multiple roles when the work is genuinely distributed.
A - Accountable
The role that carries the outcome of the lifecycle step. A is unique on every row; exactly one Accountable role per task. A must have the authority to escalate when an R role does not deliver. A may also be R, but A-only is the more durable pattern because it forces the work to flow to a delivery role.
C - Consulted
The role whose input has to be sought before the lifecycle step proceeds. C is two-way: C must be asked, and C must respond inside a documented window. C is reserved for roles whose absence would compromise the decision.
I - Informed
The role that has to be told once the lifecycle step lands. I is one-way; no response is required. I is reserved for roles that will act later based on the outcome (audit committee reading the operational record, downstream business unit absorbing a routed finding).
Rules:
1. Exactly one Accountable role per row. Multiple A entries on a row are a defect and must be resolved before publication.
2. R, C, and I may be empty on a row when no role applies; A must always be filled.
3. A role may be both A and R on the same row; A-only is preferred where the Accountable role does not personally deliver.
4. Empty cells are explicit (the role has no role on that step) rather than implied; an empty cell signals deliberate exclusion.
5. The matrix is reviewed at the cadence in Section 11; ad-hoc edits to A assignments are not permitted between revisions.
6. Role transitions (joiner, mover, leaver) trigger an explicit re-assignment of inflight rows rather than silent inheritance.
Conflict resolution:
When two or more roles dispute Accountability for a row, the dispute is escalated to the document owner, who consults the parent policy and the role definitions. The dispute is logged on the document control history with the resolution and rationale. A persistent pattern of disputes on a row is a sign the lifecycle step has been described too coarsely; the row is split into smaller steps until each carries one unambiguous Accountable role.
3. Role column definitions
Define each role column before the matrix table so reviewers read accountability against named definitions rather than against guesses. The matrix runs to 8 to 14 columns in most enterprise programmes; more than 14 is a sign the operating model has not been simplified enough. Define roles, not individuals, so the matrix survives reorganisation.
Role: VM Programme Owner (VMP)
- Function: security or GRC.
- Authority: operational lead for the vulnerability management programme; runs the discipline against the policy and reports performance.
- Reports to: head of security or CISO.
Role: Finding Triage Owner (FTO)
- Function: security.
- Authority: confirms a finding is canonical, deduplicated, classified, and routed.
- Reports to: VM Programme Owner.
Role: Asset Owner (AO)
- Function: engineering, product, infrastructure, or platform team.
- Authority: operates the asset and is the primary remediation accountability for findings on that asset.
- Reports to: head of engineering or relevant business unit head.
Role: Remediation Owner (RO)
- Function: engineering team carrying the work.
- Authority: delivers the remediation inside the SLA window and records evidence on the finding record.
- Reports to: Asset Owner.
Role: Retest Owner (RTO)
- Function: security.
- Authority: confirms remediation closes the finding, the SLA was met, and evidence is recorded.
- Reports to: VM Programme Owner.
Role: Risk Owner (RKO)
- Function: business stakeholder for the affected business unit or product line.
- Authority: carries residual risk for any exception and signs the risk acceptance.
- Reports to: business unit head or executive sponsor.
Role: Security Manager (SM)
- Function: security.
- Authority: approves low-residual exceptions; first level of the exception approver ladder.
- Reports to: Head of Security.
Role: Head of Security (HOS)
- Function: security.
- Authority: approves medium-residual exceptions; manages programme-level escalations.
- Reports to: CISO.
Role: CISO
- Function: security executive.
- Authority: approves high-residual exceptions; signs policy revisions; carries programme outcome at the executive level.
- Reports to: executive sponsor or CEO.
Role: Executive Sponsor (ES)
- Function: executive committee or named executive (CTO, COO, CFO, CEO).
- Authority: co-approves critical-residual exceptions with the CISO; carries enterprise-level residual risk.
- Reports to: board.
Role: GRC and Compliance (GRC)
- Function: GRC, compliance, and audit liaison.
- Authority: maps findings and exceptions to framework controls; runs evidence collection for external audits.
- Reports to: CISO or CFO depending on programme structure.
Role: Internal Audit (IA)
- Function: internal audit (independent of security).
- Authority: tests programme operation against the policy and reports findings to audit committee.
- Reports to: audit committee.
Role: Engineering Leadership (EL)
- Function: head of engineering or VP engineering.
- Authority: ensures remediation owners are resourced for the inflow the policy generates.
- Reports to: CTO or CEO.
Role: Platform / Infrastructure Team (PT)
- Function: platform engineering or infrastructure operations.
- Authority: shared owner for assets without a clear product-team owner; carries fallback routing.
- Reports to: head of engineering or head of infrastructure.
Role: Vendor Manager (VM)
- Function: procurement or third-party risk management.
- Authority: carries vendor-side accountability for findings on third-party services and supply-chain dependencies.
- Reports to: CFO, COO, or head of procurement.
Role: Audit Committee (AC)
- Function: board-level committee.
- Authority: receives quarterly governance reports; tests programme performance against board expectation.
- Reports to: board.
Replace, add, or remove roles to match the local operating model; preserve the function-and-authority pattern so the matrix survives the next reorganisation.
4. Identification lifecycle (rows)
Identification is the inflow side of the lifecycle. Every source that surfaces findings has to feed the canonical record on a documented cadence; the RACI assigns accountability per source so a missed identification cycle has a named owner rather than a shared hope.
Row: Schedule and run authenticated scans
- R: Platform Team, VM Programme Owner.
- A: VM Programme Owner.
- C: Asset Owner, GRC.
- I: Engineering Leadership.
Row: Schedule and run external (unauthenticated) scans
- R: VM Programme Owner.
- A: VM Programme Owner.
- C: Asset Owner.
- I: GRC, Engineering Leadership.
Row: Schedule and run code scans (SAST, SCA, secret scanning)
- R: Engineering Leadership, VM Programme Owner.
- A: VM Programme Owner.
- C: Asset Owner.
- I: GRC.
Row: Run threat modelling and manual security review
- R: Security Manager, Asset Owner.
- A: Head of Security.
- C: Engineering Leadership.
- I: VM Programme Owner.
Row: Commission third-party penetration testing
- R: VM Programme Owner.
- A: Head of Security.
- C: Asset Owner, GRC, Vendor Manager.
- I: CISO, Engineering Leadership.
Row: Run bug bounty and external researcher disclosure programme
- R: VM Programme Owner.
- A: Head of Security.
- C: Legal, GRC.
- I: Asset Owner, Engineering Leadership.
Row: Ingest threat intelligence (KEV listings, vendor advisories, EPSS feeds)
- R: VM Programme Owner.
- A: VM Programme Owner.
- C: Finding Triage Owner.
- I: CISO, GRC.
Row: Receive internal disclosure (engineer-reported, support-channel signals)
- R: Finding Triage Owner.
- A: VM Programme Owner.
- C: Asset Owner.
- I: Engineering Leadership.
Row: Receive regulator and supply-chain notification
- R: GRC, Vendor Manager.
- A: CISO.
- C: VM Programme Owner, Legal.
- I: Audit Committee, Executive Sponsor.
Row: Maintain identification coverage SLOs (authenticated scan coverage percent, external scan coverage percent, code scan coverage percent)
- R: VM Programme Owner.
- A: Head of Security.
- C: Engineering Leadership, Platform Team.
- I: CISO, GRC.
Row: Investigate and resolve scanner credential failure or coverage gap
- R: VM Programme Owner, Platform Team.
- A: VM Programme Owner.
- C: Asset Owner.
- I: Head of Security, GRC.
5. Triage and validation lifecycle (rows)
Triage and validation turn raw identification into a canonical queue. The audit reads triage accountability hardest because deduplication and classification decisions are where severity calibration succeeds or drifts. Name the Accountable role per row so the override rate, the dedup quality, and the routing latency have a named owner.
Row: Validate and confirm a finding (deduplication, asset attribution, false-positive review)
- R: Finding Triage Owner.
- A: Finding Triage Owner.
- C: Asset Owner, VM Programme Owner.
- I: GRC.
Row: Assign initial CVSS severity (base score recording)
- R: Finding Triage Owner.
- A: Finding Triage Owner.
- C: Asset Owner.
- I: VM Programme Owner.
Row: Apply severity adjustment (KEV listing escalation, public exploit availability)
- R: Finding Triage Owner.
- A: VM Programme Owner.
- C: Head of Security, GRC.
- I: Asset Owner.
Row: Apply environmental modifiers (modified attack vector, modified privileges, environmental impact)
- R: Finding Triage Owner, Asset Owner.
- A: VM Programme Owner.
- C: Head of Security.
- I: GRC.
Row: Manual severity override (increase or decrease against base score)
- R: Finding Triage Owner.
- A: Head of Security.
- C: VM Programme Owner, Asset Owner.
- I: GRC.
Row: Review severity override rate at programme level
- R: VM Programme Owner.
- A: Head of Security.
- C: GRC.
- I: CISO, Audit Committee.
Row: Tag finding to compliance scope (PCI DSS scope, HIPAA scope, SOC 2 scope)
- R: GRC.
- A: GRC.
- C: VM Programme Owner.
- I: Internal Audit.
Row: Run deduplication review at programme level (dedup accuracy)
- R: VM Programme Owner.
- A: VM Programme Owner.
- C: Finding Triage Owner.
- I: GRC.
6. Routing and ownership lifecycle (rows)
Routing is what makes ownership operationally real. A finding without a named Accountable role at routing time is a finding the programme cannot enforce a window against. Publish the routing rule in the parent policy; the matrix names the role that operates the rule.
Row: Read affected asset from the finding record
- R: Finding Triage Owner.
- A: Finding Triage Owner.
- C: Asset Owner.
- I: -.
Row: Look up asset ownership in the live asset register
- R: Finding Triage Owner.
- A: Finding Triage Owner.
- C: Platform Team.
- I: -.
Row: Route finding to named Remediation Owner
- R: Finding Triage Owner.
- A: VM Programme Owner.
- C: Asset Owner.
- I: Engineering Leadership.
Row: Notify Remediation Owner of the routed finding (acknowledgement SLA starts)
- R: Finding Triage Owner.
- A: VM Programme Owner.
- C: -.
- I: Engineering Leadership, Asset Owner.
Row: Re-route finding when asset record is incomplete or owner is unclear
- R: Finding Triage Owner, Platform Team.
- A: VM Programme Owner.
- C: Asset Owner, Engineering Leadership.
- I: GRC.
Row: Maintain asset ownership mapping (cadence and quality)
- R: Asset Owner, Platform Team.
- A: Head of Security.
- C: Engineering Leadership.
- I: GRC.
Row: Audit routing accuracy at programme level
- R: VM Programme Owner.
- A: Head of Security.
- C: Engineering Leadership.
- I: GRC, Internal Audit.
Row: Resolve disputed routing (two or more candidate owners)
- R: VM Programme Owner.
- A: Head of Security.
- C: Asset Owner, Engineering Leadership.
- I: CISO if recurring.
7. Remediation and verification lifecycle (rows)
Remediation is where the SLA window operates. Pair every row with the role that does the work and the role that carries the outcome. Retest accountability is its own row because verification is the discipline that prevents re-open rates from growing silently.
Row: Acknowledge routed finding (Remediation Owner accepts within acknowledgement SLA)
- R: Remediation Owner.
- A: Remediation Owner.
- C: Asset Owner.
- I: VM Programme Owner.
Row: Plan remediation (capacity reservation, dependency analysis, sequencing)
- R: Remediation Owner.
- A: Asset Owner.
- C: Engineering Leadership, Platform Team.
- I: VM Programme Owner.
Row: Apply remediation (patch, configuration change, code change, dependency upgrade)
- R: Remediation Owner.
- A: Asset Owner.
- C: Platform Team if shared infrastructure.
- I: VM Programme Owner.
Row: Deploy compensating control in lieu of patch (per exception path)
- R: Remediation Owner, Platform Team.
- A: Asset Owner.
- C: Head of Security, GRC.
- I: VM Programme Owner.
Row: Record remediation evidence on the finding (commit, change ticket, configuration export)
- R: Remediation Owner.
- A: Remediation Owner.
- C: Retest Owner.
- I: VM Programme Owner.
Row: Run retest (post-fix scan or manual confirmation)
- R: Retest Owner.
- A: Retest Owner.
- C: Remediation Owner.
- I: VM Programme Owner, Asset Owner.
Row: Verify retest passed and finding meets closure criteria
- R: Retest Owner.
- A: Retest Owner.
- C: VM Programme Owner.
- I: GRC.
Row: Close finding (state change, evidence retention)
- R: Retest Owner.
- A: VM Programme Owner.
- C: GRC.
- I: Asset Owner, Engineering Leadership.
Row: Investigate failed retest (finding remains open after declared remediation)
- R: Retest Owner, Remediation Owner.
- A: Retest Owner.
- C: VM Programme Owner.
- I: Engineering Leadership.
Row: Manage finding re-open (closed finding surfaced again by subsequent scan)
- R: Finding Triage Owner.
- A: VM Programme Owner.
- C: Retest Owner, Remediation Owner.
- I: Asset Owner, GRC.
8. SLA breach and stop-the-clock lifecycle (rows)
The SLA breach path and the stop-the-clock path are where governance pressure gets applied. Name accountability so a missed window or a paused clock does not silently absorb into the operating record.
Row: Detect approaching SLA breach (early warning at 70 percent, 90 percent of window)
- R: VM Programme Owner.
- A: VM Programme Owner.
- C: Remediation Owner.
- I: Engineering Leadership, Asset Owner.
Row: Escalate at SLA breach (window expired without closure or accepted exception)
- R: VM Programme Owner.
- A: Head of Security.
- C: Remediation Owner, Asset Owner, Engineering Leadership.
- I: CISO, GRC.
Row: Record stop-the-clock event (defensible condition met)
- R: VM Programme Owner, Remediation Owner.
- A: Head of Security.
- C: GRC.
- I: CISO if recurring.
Row: Resume the clock (stop-the-clock condition resolved)
- R: VM Programme Owner.
- A: Head of Security.
- C: Remediation Owner.
- I: GRC.
Row: Audit stop-the-clock pattern at programme level (rate, category, cumulative duration)
- R: VM Programme Owner.
- A: Head of Security.
- C: GRC.
- I: CISO, Audit Committee.
Row: Aggregate breach evidence for governance and audit reporting
- R: VM Programme Owner.
- A: Head of Security.
- C: GRC.
- I: CISO, Audit Committee.
Row: Resolve persistent breach concentration (one team, one asset class, one severity band)
- R: Head of Security, Engineering Leadership.
- A: CISO.
- C: VM Programme Owner.
- I: Audit Committee, Executive Sponsor.
9. Exception lifecycle (rows)
Exception governance is where the matrix has to be most explicit. The residual-rating approver ladder is what keeps the SLA from becoming a fiction. Name the Accountable role per residual band so the audit reads approval against named authority rather than against the security function as a single entity.
Row: File exception request (linked finding, risk summary, compensating control, residual rating, expiry)
- R: Remediation Owner.
- A: Asset Owner.
- C: Risk Owner, VM Programme Owner.
- I: GRC.
Row: Validate exception request (completeness, residual-rating discipline, expiry)
- R: Finding Triage Owner.
- A: VM Programme Owner.
- C: GRC.
- I: -.
Row: Approve exception with low residual rating
- R: VM Programme Owner.
- A: Security Manager.
- C: Asset Owner.
- I: Risk Owner, GRC.
Row: Approve exception with medium residual rating
- R: VM Programme Owner.
- A: Head of Security.
- C: Asset Owner, Risk Owner.
- I: GRC, CISO.
Row: Approve exception with high residual rating
- R: Head of Security.
- A: CISO.
- C: Asset Owner, Risk Owner, Executive Sponsor.
- I: GRC, Audit Committee.
Row: Approve exception with critical residual rating
- R: CISO.
- A: CISO and Executive Sponsor (joint).
- C: Asset Owner, Risk Owner, Audit Committee.
- I: Board if material.
Row: Track exception in the security exception register
- R: VM Programme Owner.
- A: GRC.
- C: Risk Owner.
- I: Audit Committee.
Row: Renew exception at expiry with fresh evidence and fresh approval
- R: Remediation Owner.
- A: original residual-band approver.
- C: VM Programme Owner, Risk Owner, GRC.
- I: -.
Row: Cancel exception when remediation lands or trigger condition invalidates it
- R: VM Programme Owner.
- A: VM Programme Owner.
- C: Remediation Owner, GRC.
- I: Risk Owner.
Row: Aggregate exception register at governance and board cadence
- R: VM Programme Owner.
- A: GRC.
- C: CISO.
- I: Audit Committee, Executive Sponsor.
Row: Investigate exception accumulation pattern (count growing past threshold)
- R: VM Programme Owner.
- A: CISO.
- C: GRC, Audit Committee.
- I: Executive Sponsor.
10. Reporting and review lifecycle (rows)
Reporting and review are where the matrix surfaces in front of audit and leadership. Name accountability for each reporting cadence and each review forum so the leadership read is anchored to a named owner rather than to a shared mailbox.
Row: Produce monthly operational report (closure rate, breach rate, exception count, coverage SLO)
- R: VM Programme Owner.
- A: VM Programme Owner.
- C: Finding Triage Owner, Retest Owner.
- I: Head of Security, Engineering Leadership.
Row: Produce quarterly governance report (operational metrics summarised, twelve-month trend lines, material breaches)
- R: VM Programme Owner.
- A: Head of Security.
- C: GRC.
- I: CISO, Audit Committee.
Row: Produce board-level report (headline performance, residual risk read, policy revisions)
- R: Head of Security.
- A: CISO.
- C: GRC, Executive Sponsor.
- I: Audit Committee, Board.
Row: Run monthly operational review (programme owner reads metrics with the operational team)
- R: VM Programme Owner.
- A: VM Programme Owner.
- C: Finding Triage Owner, Retest Owner, Remediation Owner sample.
- I: Head of Security.
Row: Run quarterly governance review (audit committee reads programme performance)
- R: VM Programme Owner.
- A: CISO.
- C: GRC, Audit Committee.
- I: Executive Sponsor.
Row: Run annual board review (full-year programme outcome, residual risk, programme direction)
- R: CISO.
- A: CISO.
- C: Audit Committee, Executive Sponsor.
- I: Board.
Row: Capture review outcomes and action register
- R: VM Programme Owner.
- A: Head of Security.
- C: GRC.
- I: CISO, Audit Committee.
Row: Track review action items to closure
- R: VM Programme Owner.
- A: Head of Security.
- C: action owners.
- I: GRC.
Row: Map findings, exceptions, and reviews to compliance frameworks (ISO 27001, SOC 2, PCI DSS, NIST, HIPAA, NIS2, DORA)
- R: GRC.
- A: GRC.
- C: VM Programme Owner.
- I: CISO, Internal Audit.
Row: Carry programme evidence into external audit
- R: GRC.
- A: GRC.
- C: VM Programme Owner, Internal Audit.
- I: CISO, Audit Committee.
11. Governance review and matrix maintenance
A RACI matrix without a review cadence drifts from the operating environment within twelve months. Pair the matrix with a review schedule, a material-change list that triggers off-cycle revision, and a documented archive of prior versions so the audit trail is reproducible.
Annual review:
- Frequency: at least every twelve months.
- Owner: document owner (Section 1).
- Review participants: Head of Security, VM Programme Owner, GRC, Engineering Leadership, Asset Owners (sample).
- Review inputs: programme performance trend, organisational change record, framework drift assessment, role transition log.
- Review outputs: revision draft (redline against current version), revision rationale, archived prior version.
Material-change-triggered revision:
- Organisational restructure that moves a role.
- New business unit, geography, product line, or acquisition that adds a role to the column set.
- New finding source that adds a row to the lifecycle.
- New framework adoption (PCI DSS scope, SOC 2 attestation, regulated geography).
- New tooling that changes who runs identification, triage, or retest.
- New regulatory expectation (CISA directive, sector-specific rule, EU regulation update).
- Policy revision that changes Section 3 of the parent VM policy.
Revision process:
1. Document owner drafts revision; redline against current version.
2. VM Programme Owner reviews operational impact.
3. Approving authority signs the revision.
4. Revised matrix distributed to full distribution list (Section 1).
5. Effective date set with appropriate notice (default 30 days; emergency revisions effective immediately with documented rationale).
6. Revision logged in document control history with date, summary of change, approver, rationale.
7. Prior version archived per the retention policy.
Document control history (maintained inline or as an appendix):
- Version, effective date, summary of change, approver, rationale.
- Previous version archived per Audit Evidence Retention Policy.
Operational tests of matrix realism (run at every governance review):
1. Sample five recent findings and check the routed remediation owner matches the matrix accountability.
2. Sample three exceptions across residual ratings and check the approver matches the residual-band rule.
3. Read the breach evidence for the period and check breach reasons map to a matrix row.
4. Read the override rate and confirm the Accountable role for severity calibration is operating.
5. Read the role transition log and confirm inflight rows were re-assigned at each move.
Failure to operate the tests is a sign the matrix is being maintained on paper but not at the finding-record level; the corrective action is to bind matrix accountability into routing, exception, and reporting workflows rather than into the document alone.
12. Signatures and acknowledgement
Sign the matrix at publication and at every material revision. The signature trail is what makes the rule defensible at audit; an unsigned operating model is treated as a draft regardless of how widely it is followed in practice. Acknowledgement from key stakeholders documents that the rule is known to the people who carry it.
Approval signatures (required at publication and at material revision):
Document owner
- Name: {{DOCUMENT_OWNER_SIGNATURE_NAME}}
- Role: {{DOCUMENT_OWNER_SIGNATURE_ROLE}}
- Date: {{DOCUMENT_OWNER_SIGNATURE_DATE}}
VM Programme Owner
- Name: {{VM_PROGRAMME_OWNER_SIGNATURE_NAME}}
- Role: {{VM_PROGRAMME_OWNER_SIGNATURE_ROLE}}
- Date: {{VM_PROGRAMME_OWNER_SIGNATURE_DATE}}
Head of Security
- Name: {{HEAD_OF_SECURITY_SIGNATURE_NAME}}
- Role: {{HEAD_OF_SECURITY_SIGNATURE_ROLE}}
- Date: {{HEAD_OF_SECURITY_SIGNATURE_DATE}}
CISO
- Name: {{CISO_SIGNATURE_NAME}}
- Role: {{CISO_SIGNATURE_ROLE}}
- Date: {{CISO_SIGNATURE_DATE}}
Executive Sponsor
- Name: {{EXECUTIVE_SPONSOR_SIGNATURE_NAME}}
- Role: {{EXECUTIVE_SPONSOR_SIGNATURE_ROLE}}
- Date: {{EXECUTIVE_SPONSOR_SIGNATURE_DATE}}
Acknowledgement (recorded for stakeholder groups; renewed at material revision):
- Heads of business units carrying remediation responsibility: {{BUSINESS_UNIT_ACKNOWLEDGEMENTS}}
- Heads of platform and infrastructure teams: {{PLATFORM_TEAM_ACKNOWLEDGEMENTS}}
- Engineering leadership: {{ENGINEERING_LEADERSHIP_ACKNOWLEDGEMENTS}}
- GRC and compliance function: {{GRC_ACKNOWLEDGEMENTS}}
- Internal audit function: {{INTERNAL_AUDIT_ACKNOWLEDGEMENTS}}
- Procurement and vendor management (for supply-chain coverage): {{PROCUREMENT_ACKNOWLEDGEMENTS}}
Effective date once all approval signatures are collected: {{EFFECTIVE_DATE_FINAL}}
Acknowledgement evidence:
- Signed acknowledgement records stored with the matrix version.
- Acknowledgement renewal cadence: at every material revision plus the annual training cycle.
- New stakeholder onboarding includes RACI acknowledgement alongside policy acknowledgement.
Six failure modes the matrix has to design against
The RACI matrix fails the audit read in recognisable patterns. Each failure has a structural fix the template above is designed to enforce. Read this list before you customise the template so the customisation does not weaken the discipline that makes the matrix defensible.
Multiple Accountable roles on a single row
Two or three columns carry an A on the same row because nobody could agree which leader owned the outcome. The audit reads the row and finds the accountability is diffuse; the operating record reads the same way. The fix is the single-A rule: split the row into smaller steps or escalate the dispute to the document owner until exactly one role carries the outcome.
Accountable role has no authority to escalate
A role is named Accountable but has no organisational authority to compel the Responsible role to deliver. When the Responsible role misses the window the Accountable role has no recourse, and the breach evidence cannot reach a leader who can act. The fix is to map A to RBAC; if the role does not control approval, escalation, or budget for the lifecycle step, it is not the right A.
Matrix lives in a slide rather than on the finding record
The RACI matrix is published once a year as a slide deck and never bound to the operational workflow. When findings are routed, exceptions are approved, or breaches are escalated, the role assignments come from memory rather than from the matrix. The fix is to bind matrix accountability into routing, exception, and reporting at the finding-record level so the role is captured at decision time rather than at reporting time.
Role transitions silently inherit accountability
A leader moves teams, leaves, or is hired into the role; their name is updated in the directory but inflight findings carry the old assignment. The new leader inherits accountability they have not formally accepted, and the audit trail records a name on findings the new leader did not approve. The fix is an explicit re-assignment process at every joiner, mover, leaver event, with the inflight rows re-routed and the role transition logged.
Exception approver ladder collapses to one role
All exceptions route to the head of security regardless of residual rating because the lower bands have no engaged approver. The leader becomes the bottleneck; exceptions accumulate; the residual-band discipline disappears. The fix is to staff and operate the ladder at every band; if a band is unstaffed the matrix has to record that and route to the next available band rather than absorb everything at one level.
GRC accountability is missing for evidence handover
The matrix names security accountability for triage, routing, and remediation but leaves GRC out of compliance scope tagging, evidence handover, and audit reporting. The audit reads compliance evidence reconstructed at audit week from operational records rather than compiled inline from a named accountable role. The fix is a GRC column with explicit A on framework mapping, evidence retention review, and external audit handover.
Ten questions the quarterly governance review has to answer
The matrix is not self-maintaining. Operational review keeps the programme on top of breaches, exception flow, and routing accuracy. Governance review answers whether the matrix is the operating model the programme is actually running against, or a paper artefact the audit will read as drift. Run these ten questions at every quarterly review and capture the answers in the governance record.
1.Does every row in the matrix carry exactly one Accountable role, with no row leaving A blank or carrying multiple A entries.
2.Does every Accountable role have the authority (budget, RBAC, escalation path) to compel delivery from the Responsible role.
3.Are role transitions (joiner, mover, leaver) in the period reflected in inflight findings, exceptions, and reviews, or did the previous role inherit silently.
4.Is the exception approver ladder operating at every residual band, or are all exceptions routing to a single approver because the lower bands are unstaffed.
5.Is GRC accountability explicit on framework mapping, evidence retention, and external audit handover, or is GRC consulted but not Accountable for any row.
6.Does the routing record on each finding capture the named Remediation Owner at routing time, or is the owner identified from memory at reporting time.
7.Is the matrix referenced from the parent vulnerability management policy, the SLA policy, the exception register, and the runbooks, or does it live in isolation.
8.Is the document control history maintained inline with version, date, approver, and revision rationale, or are prior versions discarded.
9.Has any organisational change, framework adoption, or tooling change in the period triggered a material-change revision, or did the matrix remain unchanged through a change cycle that should have moved it.
10.Does the operational test of matrix realism at the governance review (sample findings, sample exceptions, breach evidence, override rate, role transition log) confirm the matrix is operating, or does the test surface drift the next revision has to fix.
How the RACI matrix pairs with SecPortal
The template above is copy-ready as a standalone artefact. If your team already runs finding tracking, remediation, and compliance evidence on a workspace, the RACI accountability becomes the byproduct of the work rather than a separate maintenance project. SecPortal carries every finding on a versioned engagement record through findings management, so the routed Remediation Owner, the named Retest Owner, the recorded Risk Owner on an exception, and the timestamped state changes are one query against the same record rather than a reconstruction from spreadsheets at reporting time.
The team management feature carries role-based access control with five role levels (owner, admin, member, viewer, billing) so the people who can apply a severity override, record a stop-the-clock event, approve an exception at each residual band, and run governance review are gated by the platform rather than by convention. The activity log captures the timestamped chain of state changes by user across discovery, triage acceptance, ownership assignment, fix deployment, retest verification, and closure, so the elapsed time at each lifecycle step is observable rather than asserted. The matrix rows in Sections 4 through 10 above each map to a state transition the activity log already records.
The engagement management feature groups the per-finding work under the engagement record so a programme-level read of RACI operation is one query against the workspace; the routing latency, the override rate, and the breach concentration in Section 11 read against live data rather than against a separate metrics layer. The compliance tracking feature maps findings and controls to ISO 27001, SOC 2, Cyber Essentials, PCI DSS, and NIST frameworks with CSV export, so the matrix can evidence specific controls when an auditor asks for the Accountable role per control row.
The notifications and alerts feature drives the I-column of the matrix so Informed roles receive the events they need to absorb downstream, and the document management feature carries the matrix file and the revision history alongside the operational record, so the matrix version in force at any reporting cycle is reconstructible from one record. The AI report generation workflow produces leadership summaries from the same engagement data so the audit committee read of programme performance and the operational read are the same record rather than two independently-edited documents that diverge between reporting cycles.