Boards and steering committees in the Kingdom share a recurring problem: everyone agrees that cybersecurity decisions need clear owners, yet the charters that should say who decides what often stop at phrases such as "significant matters are escalated". This article sets out how to address SAMA ITGF cybersecurity decision rights between the steering committee and the board, and how to turn vague wording into a matrix that a committee can apply on an ordinary Tuesday.
What the framework expects
The SAMA IT Governance Framework (ITGF) expects IT and cybersecurity to be governed through defined structures: a steering committee, assigned roles and responsibilities, and reporting that reaches senior management and the board. The SAMA Cyber Security Framework (CSF) expects the same clarity for the cybersecurity function itself, including its independence and its accountability to senior leadership.
Neither framework gives you a ready-made table of monetary limits or risk scores. Setting thresholds is left to each institution, which is why the work falls to the people drafting charters. A regulator reviewing your governance will look for evidence that the thresholds exist, are approved, and are actually followed in committee minutes.
The usual gaps
When we see charters reviewed against these expectations, the same weaknesses recur:
- Escalation language with no measurable trigger, so each chair interprets it differently.
- Risk acceptance that has no named approver tiers, so decisions drift to whoever is available.
- Exceptions to policy that are granted without an expiry date or a review owner.
- Budget authority defined for IT in general but not for security spend specifically.
- Incident escalation that describes technical severity but not when the board must be told.
- A steering committee whose decisions are recorded, but whose recommendations to the board are not.
Principles before numbers
Before choosing thresholds, agree on a few principles. They keep the matrix defensible.
- Decide at the lowest level with the authority and the information. The board should see matters of strategy, appetite and material exposure, not routine operations.
- Tie each threshold to the institution's own risk appetite. A number copied from another organization has no logic behind it.
- Separate the decider from the requester. The person who proposes a risk acceptance should not be the person who approves it.
- Keep the function independent. The cybersecurity function should not report through IT operations, so its escalation path must reach the committee and the board without passing through the function it may be challenging.
A simple decision-rights matrix
Build the matrix with one row per decision type and one column per decision level: function owner, steering committee, executive management and board. For each cell use one of four roles: proposes, recommends, approves, informed. Each row then carries a measurable threshold.
The thresholds below are illustrations only. Calibrate them to your own risk appetite, financial scale and risk rating method.
Risk acceptance
- Low residual risk: approved by the CISO with the risk owner, reported to the committee in the periodic risk summary.
- Medium residual risk: approved by the steering committee, with a stated review date.
- High residual risk: recommended by the steering committee and approved by executive management, with the board informed.
- Critical residual risk, or any risk that falls outside stated appetite: approved by the board, or by its risk committee where one exists.
Policy exceptions
- Time-bound exceptions of up to a short period (for example 90 days) on non-critical systems: approved by the CISO.
- Longer exceptions, or any on a critical system: approved by the steering committee, with a named owner and expiry date.
- Exceptions that affect a regulatory obligation: raised to the board with the compliance function's view.
- Any exception renewed more than once returns to the committee, regardless of its original level.
Budget
- Spend within the approved security budget: approved by the function owner under delegated limits.
- Unplanned spend above a defined amount: approved by the steering committee.
- Spend above a higher defined amount, or a change to the approved annual security budget: approved by the board.
- Multi-year commitments are assessed on total contract value, not annual value, so they are not split to stay beneath a limit.
Incident escalation
- Incidents contained within normal operations: handled and logged, with a summary to the committee.
- Incidents affecting critical services, customer data or requiring regulator notification: the committee chair is informed within a stated number of hours and the board chair is informed by the executive sponsor.
- Major incidents: a board briefing is held, and the post-incident review returns to the committee and the board.
Notification duties to SAMA or to other authorities follow the regulator's own requirements. The matrix should reference them rather than restate them, so it does not go out of date.
What changes for the organization
Once the matrix is approved, several things change in practice. Committee papers carry a field stating which threshold a decision meets. Risk registers record the approving tier. Exception logs show owners and expiry dates. The company secretary can check each board paper against the matrix, and minutes show a clear trail from recommendation to approval.
A sensible order of work
- Collect the current charters, terms of reference and any delegation of authority documents.
- List the decisions that actually reach the committee and the board over the past year, and note how each was routed.
- Agree the decision types and the risk appetite statements the thresholds will rest on.
- Draft thresholds with the finance, risk and compliance functions, so they are consistent with wider delegations.
- Test the draft against real past decisions and adjust where it would have produced the wrong routing.
- Approve the charters, then run one full committee cycle before reviewing the thresholds again.
How we support this work
Our Cybersecurity Governance & Organization service covers the function, its mandate, committees, roles and responsibilities, and how decisions reach the board. The work follows four stages: review, gap, design and embed. We review the current structure, mandates, committees and reporting, then compare them with regulatory expectations and good practice. We then design the charter, committees, RACI and role descriptions, and finish with approval support, the first committee cycle and handover.
Deliverables include a committee charter and agenda templates, a RACI matrix across business, IT and security, and reporting lines and escalation paths. The service typically runs 4–6 weeks as a project, and the frameworks it draws on include SAMA ITGF, SAMA CSF and NCA ECC.
If you are not sure where your institution stands today, a compliance assessment can show the gaps against SAMA and NCA expectations before you redraft any charter.
Next step
If your charters still rely on phrases like "significant" and "material", the matrix above is a workable starting point. To have it built, tested and embedded within your own committee cycle, speak to us about Cybersecurity Governance & Organization.