Quick answer

A risk register is the structured, living list of identified risks an organisation is tracking – with severity, likelihood, owner, treatment plan and review date. It is the central artefact of operational risk management and the input to almost every risk-related conversation: board reporting, audit, insurance, due diligence and incident response.

Definition and purpose

A risk register is a structured catalogue of risks the organisation has identified, scored, assigned and is actively managing. It is not a list of incidents (those are an incident register), and it is not a list of issues (issues are realised risks). It is the live picture of what could go wrong, how bad it would be, how likely, who’s responsible, and what’s being done.

What good registers contain

Inherent vs residual risk

Inherent risk is the risk score before any controls are applied – what the risk would be in a world with no mitigations. Residual risk is the score after the controls in place are accounted for. The gap between the two demonstrates the value of the control framework. Reporting both is standard practice; reporting only one misleads.

Likelihood and severity scoring

Qualitative

Low / Medium / High or 1-5 scales. Simple, fast, intuitive. Limited precision but adequate for most organisations.

Quantitative

Expected loss in money terms, expected frequency in events per year. More rigorous but requires data that’s often missing.

Hybrid

Qualitative scales backed by quantitative anchors (‘High = likely to occur once every 5 years’). Most practical for mid-sized organisations.

Register hygiene

Three rules:

Common pitfalls

How PrivIQ supports risk registers

PrivIQ provides configurable risk registers across privacy, AI, third-party, operational and tailored domains. Each risk links to controls and assessments. Inherent and residual scoring is captured. Owners, treatment plans and review dates are first-class fields. Risks roll up into board-facing reporting.

Key takeaways
  • A risk register is the live catalogue of what could go wrong and what’s being done about it.
  • Inherent and residual scoring should both be reported.
  • Every risk needs an owner, a review date and a treatment plan. Otherwise it’s documentation, not management.
  • Risks expire. Reassessment cycles matter as much as the initial entries.
PrivIQ

PrivIQ helps organisations and consultants put this into practice — with policies, controls, evidence, tasks, registers and reporting that survive audit.

Frequently asked questions

More on GRC & Operational.

What’s the difference between a risk register and an issue log?

Risks are things that could go wrong. Issues are risks that have materialised – already happening. Most organisations maintain both, with issues feeding back into the risk register to inform residual scoring.

How many risks should be in a register?

For an operational risk register, typically 30-150 in a mid-sized organisation. Fewer suggests the register is incomplete; many more suggests inconsistent granularity. Quality matters more than count.

Should AI risks be in the main risk register?

Yes – and often as a dedicated section. AI risks have distinct dynamics (model drift, hallucination, behavioural changes) but they should live alongside other operational risks for board reporting.

Who should own the risk register?

The risk function (Chief Risk Officer or equivalent) typically owns the register as a whole. Individual risks are owned by accountable business owners. The risk function curates, the business owns each entry.

How often should the risk register be reviewed?

Continuously, but with formal review cycles: tier-1 risks monthly or quarterly, tier-2 risks twice a year, tier-3 annually. Plus event triggers – incidents, material changes, new regulation.