Article Summary

Every organisation that handles information needs documented security policies — not as a compliance formality, but as the mechanism that makes decision-making consistent and risk-controlled at scale. This article explains where policies come from, how they relate to each other, what separates a policy from a standard or a procedure, and the governance structures that give them teeth. It is the reference foundation for a ten-policy series covering the most critical domains in information security.

Why an Organisation Needs Policies

Every organisation makes decisions every day about how information is handled, who can access what, how systems are configured, and how incidents are managed. In the absence of documented policies, those decisions get made individually, inconsistently, and often without any reference to the organisation's risk appetite or legal obligations.

Two engineers in the same team will configure systems differently. Two managers will respond to the same security incident differently. Two vendors will be onboarded with completely different levels of scrutiny. That inconsistency is where risk lives, and it compounds quietly until something goes wrong.

"Policies exist to make the organisation's intentions explicit and repeatable. A policy says: this is how we have decided to operate, this is what is acceptable, and this is what is required."

They remove ambiguity from decisions that would otherwise be made differently by different people on different days. They give the organisation a documented baseline against which behaviour, systems, and third parties can all be measured.

Policies also serve a legal and regulatory function. When a regulator, an auditor, or a client asks how the organisation manages data, controls access, or handles a security incident, the answer starts with a policy. A documented, approved, reviewed, and communicated policy creates the foundation for demonstrating compliance. Without one, the organisation has nothing to point to except individual judgment calls made in the moment.

Beyond compliance, policies signal organisational maturity. An organisation that has clearly defined, consistently enforced policies communicates that it takes information security seriously. An organisation that cannot produce them communicates the opposite — and in regulated industries, in procurement processes, and in post-incident reviews, that difference matters enormously.

Where Policies Come From — The Hierarchy from Business to Policy

Information security policies do not exist in isolation. They sit at a specific layer in a hierarchy that starts with the organisation's business objectives and flows down through strategy, programme, and policy. Understanding that hierarchy is essential to writing policies that are grounded in reality rather than copied from a template that was never designed for the organisation using it.

The Policy Hierarchy

01
Business Objectives
What the organisation exists to achieve — revenue, patient care, financial transactions, national security. Everything in information security ultimately exists to enable and protect these.
02
Information Security Strategy
A 3 to 5 year directional plan translating business objectives into a systematic approach to information risk. Sets priorities, identifies significant risks, determines which capabilities need to be built.
03
Information Security Programme
The operational translation of the strategy into specific initiatives, controls, and activities with assigned ownership, timelines, and resources. The programme is where strategy becomes action.
04
Policy Framework
Documented rules, principles, and requirements governing how the programme's controls and activities are implemented consistently across the organisation.

The relationship flows in one direction. Business objectives inform the strategy. The strategy informs the programme. The programme informs the policy framework. A policy that has no connection to a business objective, a strategic priority, or a programme control is a policy that exists for its own sake and will never be taken seriously by anyone who has to follow it.

A hospital's business objective is patient care, and its information security programme must reflect the risks that could disrupt that care — from ransomware encrypting clinical systems to unauthorised access to patient records. A fintech's business objective is financial transactions, and its programme must reflect the risks that could disrupt or compromise those transactions. Generic policies copied from the internet rarely survive contact with reality precisely because they were not derived from this chain.

How the Information Security Policy Sits at the Top of the Framework

Within the policy framework itself, there is also a hierarchy. The Information Security Policy — sometimes called the overarching or master information security policy — sits at the top. It expresses the organisation's overall commitment to information security, defines the scope of the framework, establishes the governance structure, assigns accountability at the senior level, and sets the principles that all subordinate policies must follow.

Every other policy in the framework is derived from the overarching Information Security Policy. The Access Control Policy derives from the principle that access to information should be restricted to authorised individuals. The Third Party Risk Management Policy derives from the principle that the organisation's information security obligations extend to parties acting on its behalf. The Incident Management Policy derives from the principle that security events must be identified, managed, and learned from systematically.

The overarching Information Security Policy should be a short document. It is not the place for procedural detail. It states intent, scope, principles, governance, and accountability. The detail lives in the subordinate policies, standards, guidelines, and procedures that sit beneath it.

The Difference Between a Policy, a Standard, a Guideline, and a Procedure

These four terms are frequently confused and frequently misused. Getting the distinction right matters because it determines how binding each document is, how often it changes, who owns it, and how it is governed.

Policy

States what the organisation requires. Mandatory, high level, approved by senior leadership. Changes infrequently because it reflects organisational intent, not implementation detail.

Example: "All access to sensitive systems must be authenticated using multi-factor authentication."

Standard

States how the policy requirement is met in specific, mandatory terms. May be technology-specific. Changes when technology or regulatory requirements change, without requiring the policy to change.

Example: "MFA must use TOTP or a hardware token. SMS-based authentication is not acceptable for sensitive systems."

Guideline

Provides recommended approaches and best practices. Advisory, not mandatory. Supports the policy and standard, giving people discretion to apply judgment where the standard alone is not sufficient.

Example: "When selecting a TOTP app, prefer applications that support encrypted backup and biometric unlock."

Procedure

Provides step-by-step instructions for completing a specific task in compliance with policy and standard. Operational and role-specific. Changes most frequently as systems and processes evolve.

Example: "Step 1: Navigate to the MFA enrollment portal. Step 2: Select your authenticator application…"

Policies must not contain the level of detail that belongs in standards or procedures. When a policy becomes too prescriptive, it changes too often, becomes difficult to approve through governance, and conflates intent with implementation in a way that creates confusion about what is actually mandatory.

The Balance a Policy Must Strike

A policy that is too permissive fails to protect the organisation. A policy that is too restrictive fails the business. Both are real failure modes and both are common.

Over-controlled policies create friction that people work around. If the access control policy is so restrictive that developers cannot do their jobs, they will find unofficial ways to bypass it. A policy that nobody follows is worse than no policy at all because it creates the illusion of control without the substance, and it undermines the credibility of every other policy in the framework.

Under-controlled policies leave gaps that attackers, accidents, and auditors will find. A data classification policy that does not address cloud storage, personal devices, or third-party sharing in an organisation that routinely uses all three is a policy that does not reflect how the organisation actually operates.

"A useful test for any policy requirement: if this requirement were violated, what is the realistic harm to the organisation? If the answer is significant, the requirement belongs in the policy. If minimal, it belongs in a guideline — or nowhere at all."

How Often a Policy Should Change

Policies should be stable. They reflect the organisation's intent and principles, which do not change with every new technology, regulation, or threat. A policy that is rewritten every year signals it was written at the wrong level of abstraction.

The standard practice is to review policies annually and to update them when there is a material change in the organisation's risk environment, regulatory obligations, business model, or governance structure. What changes more frequently is the standards and procedures beneath the policy. When a new technology is adopted, the relevant standard is updated. The policy remains stable while the implementation layer evolves around it.

Every policy should carry a version number, a review date, an owner, and an approval authority. These are the mechanisms by which the organisation knows whether its policies are current, who is accountable for them, and when they were last examined by someone with authority to change them.

One Policy, Multiple Regulations

A common mistake organisations make is to create separate policies for each regulation they are subject to — a GDPR policy, a PCI DSS policy, an ISO 27001 policy. This approach creates duplication, contradiction, and maintenance overhead that grows with every new regulation.

The right approach is to maintain one policy per domain, written to reflect the organisation's own requirements and best practice, and then to incorporate regulatory obligations where they add specific requirements beyond the baseline. The mapping between policy requirements and regulatory obligations is maintained separately, typically in a controls matrix. This allows the organisation to demonstrate to any regulator exactly where in its policy framework their requirements are addressed.

Policy Ownership — Who Owns a Policy and What That Means

Every policy must have a named owner. Ownership is not a formality — it is accountability. The policy owner is responsible for ensuring the policy exists, reflects current requirements, is reviewed on schedule, and is communicated to those who need to follow it. A policy without a named owner belongs to nobody and will drift out of currency the moment the person who wrote it moves on.

Ownership means the policy owner understands the policy thoroughly, can explain and defend every requirement in it, monitors whether the organisation is actually complying, escalates violations, and drives the review process when the review date arrives.

Policy Reviewer, Steering Committee, and Segregation of Duties

A policy owner writing, reviewing, and approving their own policy is a governance failure. The same person who has operational responsibility for a domain cannot be the final authority on whether the policy governing that domain is adequate. That is a conflict of interest and a segregation of duties violation.

RoleResponsibilityWho Typically Fills This
Policy Author Drafts the policy with deep domain knowledge. Policy owner or subject matter expert under their direction.
Policy Reviewer Reviews for technical accuracy, regulatory completeness, feasibility, and alignment with the overarching policy. Must be independent of the author. Legal Counsel, IT Director, Internal Audit — multiple reviewers from different functions.
Policy Approver Formally approves the policy, giving it organisational standing. Authority level should match the scope and risk significance of the policy. CISO, Chief Risk Officer, Steering Committee, or Board depending on scope.
Compliance Assessor Independently assesses whether the organisation is actually complying with the policy over time. Internal Audit function.

Policy Versioning

Every policy carries a version number and a revision history. Versioning is the mechanism by which the organisation tracks what the policy said at any point in time, what changed, why, and who approved the change. Without it, the policy framework cannot support any regulatory or legal process that requires the organisation to demonstrate what its requirements were at a specific historical point.

Major version numbers — 1.0, 2.0 — indicate significant changes requiring full re-approval. Minor version numbers — 1.1, 1.2 — indicate clarifications or additions that do not change fundamental requirements. The revision history table records version number, date, description of change, author, and approver. This is the audit trail for the policy's lifecycle.

The Ten Policies This Series Will Cover

The following ten policies form the backbone of any information security management system. The series covers each in depth, including base requirements applicable to all organisations and industry-specific variations across fintech, healthcare, defence, energy, and consulting environments.

The Ten Core Policies — Policy Series
01 Information Security Policy Read article →
02 Access Control Policy Read article →
03 Data Classification Policy Read article →
04 Change Management Policy Coming soon
05 Incident Handling and Management Policy Coming soon
06 Third Party Risk Management Policy Coming soon
07 Human Resource Security Policy Coming soon
08 Business Continuity Policy Coming soon
09 Network Security Policy Coming soon
10 Cloud Computing Policy Coming soon

Key Takeaways

Policies prevent inconsistency at scale
Without documented policies, the same decision gets made differently by different people on different days. That inconsistency accumulates until it produces an incident, a compliance finding, or an audit failure.
Policies derive from business objectives
A policy with no traceable connection to a strategic priority or programme control will never be taken seriously. Derivation from the hierarchy is not a formality — it is what makes a policy legitimate.
Policy, standard, guideline, and procedure are distinct instruments
Conflating them produces documents that change too often, confuse intent with implementation, and lose credibility with both their authors and their audience.
One policy per domain, not one policy per regulation
Maintaining separate policies for GDPR, PCI DSS, and ISO 27001 creates duplication and contradiction. One policy per domain with regulatory obligations incorporated as clauses, plus a separate compliance mapping document.
Segregation of duties applies to the policy lifecycle
The author, reviewer, approver, and compliance assessor should never be the same person or function. A policy owner who writes, reviews, and approves their own policy has no independent check on adequacy.
Version everything and maintain the audit trail
Without a revision history, the organisation cannot demonstrate what its requirements were at a specific point in the past — critical in regulatory investigations, post-incident reviews, and contract disputes.