This privileged access management case study for Saudi Arabia describes a composite situation we see often in financial institutions: access reviews exist on paper, yet leaver accounts and shared administrator credentials linger. It is written so that you can compare your own baseline against it. No client is identified, and no figures are claimed. The aim is to show what changed in the way controls were operated and which measures to track.
The starting position
The organization had an access control policy aligned to ISO 27001 Annex A and an annual access review that was signed off every year. The paperwork was in order. The operating reality was different:
- Leaver lists from HR were sent by email and were not reconciled against directories, applications or cloud identity.
- Administrator credentials for servers, databases and network devices were shared among team members, with passwords that rarely changed.
- Reviewers approved spreadsheets of access they could not interpret, often in bulk.
- Mover cases kept their old access when staff changed roles, so entitlements accumulated.
- No one could say how long it took to remove access after someone left.
None of this is unusual. It is what happens when a control is designed once and then not run. Regulators such as SAMA, through its Cyber Security Framework, and the NCA, through the Essential Cybersecurity Controls, expect access to be restricted to need, privileged access to be tightly managed, and evidence to exist that this happens. ISO 27001 Annex A expects the same: access rights provisioned, reviewed and removed in a controlled way, and privileged access restricted and managed.
Signs it is time to operate these controls
If several of these apply to you, the gap is probably operational rather than documentary:
- Your last review found accounts nobody could attribute to a person or a system.
- Leaver removal depends on a manager remembering to raise a ticket.
- Administrators use shared or generic accounts, so logs cannot name an individual.
- Review sign-off is fast, with few or no access removals recorded.
- An auditor has asked for evidence of when access was revoked and you needed days to assemble it.
Step one: baseline and discovery
Before changing anything, the team established what existed. This meant listing accounts across directories and key applications, mapping roles, and identifying the paths through which privileged access was obtained. The output was a first register of privileged accounts, each with an owner, a purpose and a platform.
Two baseline measures were fixed at this point, because without them improvement cannot be shown:
- The count of orphaned accounts, meaning active accounts with no current employee or documented system owner.
- The time to revoke, measured from the HR leaving date to confirmed removal of access.
Step two: joiner-mover-leaver tied to HR events
The central change was to stop treating HR and IT as separate processes. Joiner, mover and leaver events in the HR system became the trigger for access changes, and each event was checked against what was actually provisioned.
- Joiners received access by role, with approval recorded by a named owner.
- Movers had previous-role access reviewed and removed, not simply added to.
- Leavers triggered revocation across directory, applications and remote access, with confirmation recorded.
Reconciliation was run on a schedule: the HR leaver list compared with active accounts. Anything that did not match became an exception with an owner and a deadline. This is where most orphaned accounts were found and closed, including service and contractor accounts that HR never knew about.
Step three: quarterly access reviews with named reviewers
The annual review was replaced by quarterly recertification campaigns. Three design choices made the difference:
- Each review had a named reviewer who understood the access, usually the line manager for user access and the application owner for entitlements.
- Reviewers saw current entitlements in readable form, not raw exports, and could approve, remove or query each item.
- Unreviewed items were escalated rather than auto-approved, and every removal was logged.
Segregation-of-duties conflicts were examined in the same cycle, for example one person able both to create and to approve a payment beneficiary. Conflicts that had a business reason were accepted formally and recorded in an exceptions log with an approver and an expiry date.
Step four: vaulting and session recording
Shared administrator credentials were the highest-risk finding. They were moved into a vault so that administrators no longer knew the passwords. Access was requested, approved, time-bound and tied to an individual. Passwords were rotated after use and on a schedule.
Sessions on critical systems were recorded, so that privileged activity could be reviewed and attributed. This matters for investigations and for fraud and insider-risk controls, because a shared account makes it impossible to say who did what. Vault ownership stayed with the organization, and any access by external support staff was limited, logged and approved.
What moved
The measures to watch are simple, and you can calculate them from your own data:
- Orphaned account count: should fall after the first reconciliation and stay low, with new cases appearing only as short-lived exceptions.
- Time to revoke: should shorten once HR events drive removal, and should be reported as a trend, not a single figure.
- Shared privileged accounts: should trend to zero, with each exception recorded.
- Review outcomes: a review that removes access is evidence that someone looked. Zero removals across several cycles deserves scrutiny.
- Open exceptions past due: shows whether approvals are being managed or forgotten.
We have deliberately not quoted numbers here. The useful comparison is your own movement from baseline, quarter by quarter.
What a good outcome looks like
A good outcome is not a clean report once a year. It is a state in which leaver access is removed promptly and provably, privileged access is attributable to individuals, reviewers make real decisions, and exceptions are known and time-limited. When an auditor or a regulator asks, the evidence already exists.
How we run this
Our Managed Identity and Privileged Access service is the operation of identity and privileged access controls: joiner-mover-leaver, access reviews, privileged account vaulting and session oversight. We work through four stages: discover accounts, roles and privileged paths; define review cycles and approval owners; operate reviews, rotation and monitoring; and produce evidence for audit and regulators.
The service includes access recertification campaigns, joiner, mover and leaver checks, privileged account vaulting and rotation, session monitoring for administrators, and segregation-of-duties conflict review. Deliverables are access review results, a privileged account register, a monthly identity risk report and an exceptions and approvals log.
It is delivered as an ongoing managed service on 12-month terms. Coverage includes common directories, cloud identity and privileged access tools, with the platform list confirmed in scoping. Credentials stay in a vault under your ownership, and access by our staff is limited, logged and approved. The service maps to SAMA CSF, NCA ECC, PCI DSS and ISO 27001.
Next step
If your reviews are signed off but leaver and privileged accounts still linger, start by measuring your orphaned account count and time to revoke. Those two numbers will show whether the controls are running. If you would like to discuss scoping a managed identity and privileged access service, speak with our team about Managed Identity and Privileged Access.
