GRC, Decoded: Governance, Risk & Compliance as One System

Three words that get used as one acronym and run as three silos. Here is how they fit together — the standard behind each layer, the ten steps to build a programme, the arithmetic, and the vocabulary.

3disciplines: direct, manage uncertainty, meet obligations
5layers once cyber & privacy join
10steps from charter to continuous improvement
8risk areas every register should cover
30terms, explained in one line each

Governance decides where the organisation is going and who is accountable. Risk management works out what could knock it off course and how much of that it is willing to carry. Compliance proves that the obligations it has signed up to — laws, contracts, standards, its own policies — are actually met. Run separately, the three produce three registers, three sets of evidence and three different answers to the board. Run as one system, a single risk drives a single control, tested once, reported once, and mapped to every obligation it satisfies.

1 · Five Layers and the Standard Behind Each

Most GRC diagrams stack the same five layers. What they rarely say is that each layer already has an international reference — you do not need to invent the method, only to connect them.

LayerIts jobReference standardsWhat it produces
GovernanceDirects and oversees; sets accountabilityISO 37000:2021 (governance of organisations); COBIT 2019 (governance of enterprise I&T, 40 objectives)GRC charter, RACI, policy hierarchy, board reporting cadence
RiskManages uncertainty against objectivesISO 31000:2018 (principles & process); COSO ERM 2017 (5 components, 20 principles)Risk appetite statement, risk register, treatment plans, KRIs
ComplianceMeets obligations and proves itISO 37301:2021 (compliance management systems — certifiable)Obligations register, compliance tracker, audit reports
CybersecurityManages cyber risk and resilienceISO/IEC 27001:2022 (93 Annex A controls, 4 themes); NIST CSF 2.0 (Govern, Identify, Protect, Detect, Respond, Recover); NIST SP 800-53 Rev. 5 (control catalogue)ISMS, Statement of Applicability, control matrix
PrivacyProtects personal data and data-subject rightsISO/IEC 27701 (privacy information management); laws: India’s DPDP Act 2023, EU GDPR, Saudi PDPLRecords of processing, consent & notice records, breach log

▸ The tell-tale of a mature programme: CSF 2.0 added “Govern” as a sixth function in 2024. Cybersecurity standards now expect the governance layer to exist above them — they no longer pretend security can be run as a purely technical function.

Regional overlays sit on the cyber layer: in Saudi Arabia the NCA’s ECC, CCC (cloud), DCC (data), CSCC (critical systems) and OTCC (operational technology); in India, CERT-In directions and sector regulators (RBI, SEBI, IRDAI).

2 · The 10-Step Build

The order matters. Teams that start at step 7 (buying a tool, writing controls) end up with controls that protect nothing in particular, because nobody decided at steps 1–2 what mattered or how much risk was acceptable.

  1. Governance structure & charter — who owns GRC, which committee it reports to, decision rights. Adopt the IIA three lines model: management owns risk, a second line oversees it, internal audit assures it independently.
  2. Risk appetite — a board-approved statement of how much risk, of which kinds, is acceptable. Without it, every risk decision is an argument.
  3. Policy framework — a hierarchy: policy (what and why) → standard (measurable rule) → procedure (steps) → guideline (advice).
  4. Regulatory mapping — an obligations register: every law, contract clause and standard, mapped to an owner and a control.
  5. Risk identification — assets, threats, vulnerabilities and business processes, captured in one register.
  6. Risk assessment — likelihood × impact on a defined scale; rank against appetite.
  7. Control implementation — one control, mapped to many obligations (“test once, comply many”).
  8. Monitoring & KRIs — key risk indicators with thresholds that trigger escalation before a risk becomes an incident.
  9. Reporting & dashboards — the board sees appetite vs actual exposure, not control counts.
  10. Continuous improvement — audit findings, incidents and near misses feed back into steps 5–7.
Draft→Review→Approve→Implement→Monitor→Improve

Every GRC document — charter, policy, risk register, control matrix — goes round this loop, with a named approver and a review date.

3 · The Arithmetic

GRC has very little maths, and the little it has is often misused. Three formulas carry most of the weight:

Risk score = Likelihood (1–5) × Impact (1–5)  →  1–25
Control effectiveness % = (Controls tested effective ÷ Controls tested) × 100
Compliance rate % = (Obligations met ÷ Obligations in scope) × 100
ScoreBandTypical treatment
15–25HighAbove appetite — treat now, owner at executive level, report to board
8–14MediumTreat within an agreed plan; monitor with a KRI
1–7LowAccept and record; review annually

▸ Two traps. A 5×1 and a 1×5 both score 5 — but a near-certain nuisance and a rare catastrophe need different treatment, so keep both numbers visible. And an untested control is not an effective one; count only controls whose operation was tested in the period.

Treatment options are the same in every standard: avoid (stop the activity), mitigate (add controls), transfer (insurance, contract) or accept (within appetite, signed by the owner).

4 · Eight Risk Areas and the Controls That Answer Them

GRC fails where accountability is unclear. A register that covers only cyber risk is a security register, not a GRC one. These eight areas, each paired with the assurance mechanism that tests it, are the minimum coverage:

AreaWhat goes wrongControls & assurance
GovernanceWeak board oversight, undefined accountability, policy gapsBoard governance reviews, RACI, policy lifecycle management
StrategicMisaligned objectives, no risk appetite, KPIs nobody watchesEnterprise risk assessments, appetite statements, performance reviews
OperationalProcess failures, control breakdowns, vendor disruptionStandard operating procedures, internal control testing, third-party monitoring
ComplianceRegulatory violations, missed filings, licensing gapsRegulatory monitoring, compliance audits, training
FinancialFraud, misstatement, budget overrunsFinancial audits, anti-fraud controls, segregation of duties
Cyber & ITData breaches, ransomware, downtimeIT security audits, vulnerability assessments, incident-response testing
Data governancePoor data quality, unclassified data, privacy violationsData governance framework, classification reviews, privacy checks
PhysicalAsset theft, intrusion, safety hazardsPhysical security audits, surveillance, safety inspections

▸ The four mistakes that sink programmes: weak tone at the top, an undefined risk appetite, policies nobody enforces, and monitoring that happens once a year at audit time.

5 · 30 Terms, One Line Each

TermPlain meaning
GovernanceHow the organisation is directed, controlled and held to account
Risk managementFinding what could cause harm and deciding what to do about it
ComplianceMeeting the laws, contracts and rules the organisation must obey
PolicyA high-level rule stating what must happen and why
ProcedureThe step-by-step way a policy is carried out
Internal controlA check built into a process to keep it safe, accurate and honest
Control testingChecking that a control actually operated as designed
AssuranceIndependent confidence that controls and processes work
AuditA formal, evidence-based examination against criteria
Risk assessmentRating what could go wrong by likelihood and impact
Risk registerThe list of identified risks, owners, scores and treatment status
Risk appetiteHow much risk the organisation is willing to pursue or retain
Risk toleranceThe acceptable variation around that appetite for a specific risk
Risk mitigationActions that reduce a risk’s likelihood or impact
Residual riskThe risk that remains after controls are applied
Risk monitoringWatching risks and indicators for change over time
KRIKey risk indicator — a metric that warns a risk is growing
Third-party riskRisk that comes from vendors, suppliers and partners
Due diligenceCareful checking before a decision, deal or onboarding
RegulationA legally binding rule made by a government or regulator
Compliance monitoringOngoing checks that obligations are still being met
SanctionsPenalties for breaking rules or laws
IncidentAn unexpected event that causes, or could cause, harm
Incident responseThe planned way to contain and recover from an incident
Data breachPersonal or confidential data reached people it should not have
Business continuityKeeping critical services running through a disruption
EscalationRaising an issue to a higher level of authority
AccountabilityBeing answerable for a decision or an outcome
WhistleblowerA person who reports wrongdoing inside an organisation
EthicsDoing the right thing when no rule forces it

Bottom line

  • One system, not three silos — one risk, one control, one test, mapped to every obligation it meets.
  • Every layer already has a standard — ISO 37000/COBIT, ISO 31000/COSO, ISO 37301, ISO 27001/NIST CSF, ISO 27701.
  • Appetite before controls — steps 1 and 2 decide whether steps 7–10 protect anything that matters.
  • Keep likelihood and impact visible — a single score hides the rare catastrophe.
  • Cover all eight risk areas — a register that is only cyber is not GRC.

Sources & method. Layer-to-standard mapping from the published scopes of ISO 37000:2021, COBIT 2019 (ISACA), ISO 31000:2018, COSO ERM (2017), ISO 37301:2021, ISO/IEC 27001:2022, NIST CSF 2.0 (February 2024), NIST SP 800-53 Rev. 5 and ISO/IEC 27701; three lines model from the IIA (2020). Score bands are an illustrative 5×5 convention, not a standard requirement. Unsourced statistics that circulate with GRC infographics are deliberately not repeated. Original analysis; no third-party graphics or text reproduced.

← Back to Cyber security