Article Summary

Access control is the mechanism by which an organisation translates its information security principles into operational reality. This article covers the four core principles — least privilege, segregation of duties, authentication, and accountability — how access control requirements differ across data centres, banks, government facilities, hospitals, and small businesses, what the policy must contain in each area, and a sample policy structure.

Why This Policy Exists

Access control is the mechanism by which an organisation translates its information security principles into operational reality. It is the difference between an organisation that says only authorised people can access sensitive information and an organisation that can demonstrate it. Every breach investigation that has ever been conducted finds access control failures at or near the root cause.

Accounts with excessive privileges. Credentials that were not revoked when an employee left. Systems accessible from the internet that should have been internal only. Shared accounts that make accountability impossible. Access control failures are not exotic. They are common, predictable, and preventable.

The Access Control Policy exists to establish the organisation's requirements for managing who can access what, under what conditions, through what mechanisms, and with what level of accountability. CISM's risk management domain is explicit that access control is one of the primary mechanisms for reducing information security risk to an acceptable level.

How the Policy Is Derived

The Access Control Policy is derived from the Information Security Policy's principle that access to information must be restricted to authorised individuals based on a legitimate business need. The specific requirements are informed by the organisation's risk assessment, its data classification scheme, its regulatory obligations, and the specific access risks present in its technology environment.

An organisation that processes payment card data derives specific access control requirements from PCI DSS, which mandates least privilege access, individual accountability for all access to cardholder data, and strict controls on remote access. An organisation subject to GDPR derives access control requirements from the regulation's requirement to implement appropriate technical measures to protect personal data. These regulatory requirements do not create separate access control policies. They add specific requirements to the single Access Control Policy the organisation maintains.

The Core Principles of Access Control

01
Least Privilege
Every user receives only the access rights required to perform their specific job function. A finance manager needs access to financial systems. They do not need access to HR databases, production systems, or security monitoring tools. Granting access beyond what is required does not make anyone's work easier in a meaningful way. It does create risk every time a credential is compromised, every time an employee makes an error, and every time someone acts with malicious intent.
02
Segregation of Duties
Critical functions are divided between multiple people so that no single individual can complete a high-risk transaction without the involvement of at least one other person. A payments system should require one person to initiate a payment and a different person to approve it. A system administrator should not also be the person who reviews and approves administrator access rights. Segregation of duties is simultaneously an access control principle and an internal audit principle.
03
Authentication
The organisation can verify that the person requesting access is who they claim to be. Strong authentication requires more than a password. Passwords are stolen, guessed, phished, and reused across multiple systems. Multi-factor authentication adds a second verification step that an attacker who has stolen a password cannot complete without also compromising the user's second factor. The policy must establish authentication requirements for different system types, with more stringent requirements for systems holding more sensitive data.
04
Accountability
Every access event can be traced to a specific individual. Shared accounts, generic accounts, and service accounts without individual ownership make accountability impossible. When a security incident involves unauthorised access, the investigation requires the ability to determine exactly who accessed what and when. Without individual accountability built into the access control architecture, that determination is often impossible.

How Access Control Differs by Organisation and Environment

Data Centre

Access control applies at multiple layers simultaneously. Physical access to server rooms is controlled through card readers, biometric scanners, and mantrap entry. Logical access uses privileged access management solutions and jump servers. Remote access to management systems requires MFA over encrypted connections. The policy must address both physical and logical layers explicitly.

Bank or Financial Institution

Teller systems, trading platforms, payment processing, and customer databases all require granular role-based access with individual accountability for every transaction. Financial sector regulators mandate strict controls on privileged access, comprehensive audit logging of all access to customer financial data, and formal access review processes. Access to trading systems requires additional controls to prevent insider trading.

Government Facility

Personnel security clearance levels determine access to classified information, and the access control policy must integrate with the national clearance scheme. Physical access to secure areas requires clearance verification as well as authentication. The principle of need-to-know is applied more rigorously than in commercial environments — a cleared individual can only access classified information they have a specific, documented need to access for their current assignment.

Hospital

Must balance strict access control for patient data with the operational reality that clinical staff need rapid, reliable access in time-critical situations. An emergency physician treating an unconscious patient cannot wait for an access provisioning process to complete. The policy must address break-glass procedures with full audit logging and mandatory post-access review — a requirement unique to the healthcare environment.

Small Business

Cannot implement a full privileged access management platform or dedicated identity governance solution. The policy must reflect realistic implementation. Strong passwords enforced through the directory service, MFA for remote access and email, regular access reviews by managers, and prompt revocation on employee departure are achievable and represent a significant risk reduction without requiring enterprise tooling.

Cloud-First Organisation

Traditional network perimeter no longer provides meaningful access control. Identity becomes the primary perimeter. The policy must address cloud identity and access management, federation with on-premises directories, service account governance, and the management of API keys and secrets. Just-in-time access models — where privileged access is granted for a defined task window and automatically revoked — are particularly important in cloud environments.

What the Policy Must Contain and Why

Requirement Area What the Policy Establishes Why It Matters
Access Provisioning All access requests must be formally submitted with business justification, approved by the relevant system or data owner, and granted in accordance with least privilege. Informal provisioning creates access rights that are not documented, not reviewed, and often not revoked when the employee moves to a different role.
Access Review Access rights reviewed at least annually for standard users, at least quarterly for privileged accounts. Conducted by system owners and line managers, not the IT function alone. Access rights accumulate over time as employees change roles. Without periodic review, the least privilege principle erodes continuously from the day access is first granted.
Access Revocation Access revoked on the user's last day of employment. Covers all systems including those managed by third parties. Confirmed through a formal offboarding checklist. For high-risk departures, revocation completes before the employee is notified. Access remaining active after an employee departs is a significant security risk, particularly for employees who left under difficult circumstances.
Privileged Access Privileged accounts separate from regular user accounts. Granted only when operationally required. All sessions logged. Reviewed quarterly. Shared privileged accounts prohibited. Privileged accounts can access, modify, or delete vast amounts of organisational data. They are high-value targets for attackers and high-risk accounts from an insider threat perspective.
Authentication MFA mandatory for all remote access, all access to Confidential or Restricted data, and all privileged access. Specific requirements defined in the Authentication Standard. Passwords alone are insufficient against modern attack methods. MFA prevents credential-only attacks from succeeding even when credentials have been compromised.
Remote Access All remote access uses encrypted connections over approved solutions. MFA mandatory. Access from unmanaged personal devices to sensitive systems prohibited unless specifically approved. Remote access has become standard operationally but significantly expands the attack surface. Controls must expand with it.
"Least privilege is the single most effective access control principle in practice. It reduces the blast radius of every credential compromise, every error, and every insider threat — simultaneously."

Sample Access Control Policy Structure

Sample Structure — Illustrative, Not a Template to Be Copied
Organisation: [Name]  |  Version: 1.0  |  Owner: CISO  |  Approved By: Information Security Steering Committee  |  Classification: Internal

1. Purpose

This policy establishes [Organisation Name]'s requirements for controlling access to information systems, applications, and data. It defines the principles, mechanisms, and processes by which access rights are granted, maintained, reviewed, and revoked, with the objective of ensuring that access to information is restricted to authorised individuals with a legitimate business need, in proportion to their role and responsibilities.

2. Scope

This policy applies to all information systems, applications, databases, and physical facilities owned or operated by [Organisation Name] or processing [Organisation Name] data on its behalf. It applies to all employees, contractors, consultants, and third parties who require access, regardless of location or access method.

3. Access Control Principles

Least Privilege: users receive only the access rights required for their assigned job functions. Segregation of Duties: critical functions are divided between individuals to prevent any single person completing a high-risk transaction without oversight. Individual Accountability: all access activity is attributable to a specific named individual. Shared or generic accounts are not permitted for systems processing Confidential or Restricted information. Need to Know: access to sensitive information is further restricted to individuals with a specific, documented business need, even within their authorised role.

4–13. Provisioning, Review, Revocation, Privileged Access, Authentication, Remote Access, Physical Access, Regulatory Alignment, Exceptions, Related Documents

All access requests submitted formally with business justification and approved by system or data owner. Access rights reviewed annually for standard users and quarterly for privileged accounts. Access revoked on last day of employment across all systems. Privileged accounts separate, logged, and reviewed quarterly — shared privileged accounts prohibited. MFA mandatory for remote access, sensitive data access, and all privileged access. Remote access encrypted; unmanaged devices prohibited for sensitive systems. Physical access controlled through approved mechanisms with logs retained per regulatory requirements. Regulatory alignment with PCI DSS, GDPR, and applicable sector regulations. Exceptions require CISO approval, documented risk assessment, and annual review.

Revision History: Version 1.0 — [Date] — Initial issue — Author: CISO — Approved by: Information Security Steering Committee

Key Takeaways

Access control failures are the most common breach root cause
Excessive privileges, unrevoked credentials, shared accounts, and missing MFA are not exotic vulnerabilities. They are common, predictable, and preventable with a well-enforced access control policy.
Least privilege is the highest-impact single principle
It reduces the blast radius of every credential compromise, every insider threat, and every unintentional error simultaneously. It should be the baseline assumption, not a special control applied to sensitive systems only.
Access reviews must be owned by the business, not IT
The IT function cannot determine whether access is still justified without business context. System owners and line managers must conduct reviews because they know whether the access still matches the role.
Revocation on departure is the most time-critical access control event
Active credentials for departed employees represent an immediate risk that grows with each day they remain. For involuntary departures, revocation must complete before the employee is notified — not after.
MFA is no longer optional for sensitive access
Passwords alone cannot withstand modern credential attack methods. MFA for remote access, privileged access, and access to sensitive data is the current minimum viable standard, not an advanced control.

Previous in this series

Data Classification Policy

Next in this series

More policy articles coming