Most ISMS projects get stuck in the same place: policies and procedures are written for three months, then the auditor asks, “Can I see the record showing this rule is applied?” and there’s no record in the folder. An information security management system (ISMS) isn’t a set of documents; it’s an operating model that defines who makes the organization’s information security decisions, by what criteria and how often. ISO/IEC 27001:2022 doesn’t dictate how these decisions are made; it requires the decisions made to be systematic, measurable and improvable. The sequence below shows the path an organization with no documents at all will follow up to the certification audit, and a mistake that can come up at each stage.
Three things to know before you start
- The structure of the standard: Clauses 4–10 of ISO/IEC 27001:2022 contain the management system requirements (context, leadership, planning, support, operation, performance evaluation, improvement), and Annex A contains the list of controls. The 93 controls in Annex A are grouped into four themes: organizational, people, physical and technological.
- The 2024 amendment: Amd 1, published in 2024, adds an assessment of whether climate change is a relevant issue for the organization in the context analysis, and requires interested parties’ expectations on this issue to be taken into account. For details, see our guide to the 2022 version and the 2024 amendment.
- Whose job certification is: Certificates are issued by accredited certification bodies. A consultant sets up and prepares the system; they can’t guarantee the audit result.
A fourth point can be added: clauses 4–10 aren’t optional. Annex A controls can be excluded with justification; management system clauses can’t.
Stage 1 — Context, interested parties and scope
The first step is to put in writing the internal and external issues that affect the organization’s information security (market, legislation, technology, organizational structure, outsourcing) and the expectations of interested parties (customers, employees, regulators, suppliers). The scope statement is the result of this analysis: which locations, processes and systems are inside the ISMS, and how will the interface with what’s left out be managed?
A common mistake is narrowing the scope artificially to reduce certification costs. If the process that delivers the service the customer expects is left out of scope, the certificate loses its commercial value. The second common mistake goes the other way: “the whole company” is written down, and then not a single control record can be produced for the factory’s OT network, the warehouse or the sales office. The scope should be as wide as you can actually operate.
The place of IT infrastructure within the scope must also be drawn separately. Cloud services (Microsoft 365, hosting, backup provider) are usually part of the processes in scope; they’re managed as suppliers but can’t be treated as out of scope.
Stage 2 — Leadership, policy and roles
Top management approves the information security policy, allocates resources and assigns ISMS roles. In practice, an ISMS manager (internal or outsourced), process-based control owners and a management representative with decision authority are appointed. Showing roles in a RACI table reduces the audit question “who’s responsible for this control?” to a single answer.
Auditors measure leadership by behavior, not by the policy text. A general manager who attends the management review meeting, an approved budget line and a manager whose signature is on risk acceptance decisions are this evidence. Having the IT manager sign the policy isn’t enough.
Stage 3 — Risk assessment methodology and risk register
The standard expects risk assessment to produce consistent, comparable results when repeated, risk owners to be identified and risk acceptance criteria to be defined. For this:
- An asset-based or scenario-based approach is chosen.
- A 3-point or 5-point scale is defined for likelihood and impact; the impact scale covers financial, operational, legal and reputational dimensions.
- The risk acceptance threshold is set with management (e.g. 8 or below is acceptable in a 5×5 matrix).
- Risks are identified and scored in workshops with process owners.
Scale definitions must be concrete. If instead of “high impact” you write “production stoppage longer than one working day” or “an incident requiring a personal data breach notification”, two different people will give the same risk the same score. That’s consistency. Having someone from IT in the workshop is also essential; otherwise technical risks such as “backups being on the same network” never make it onto the list.
For the method and sample records, see the ISMS risk analysis guide.
Stage 4 — Risk treatment and the Statement of Applicability
A decision is made for every risk above the acceptance threshold: reduce, avoid, transfer or accept with justification. Controls are selected for reduction decisions, and these controls are compared with Annex A to confirm that no necessary control has been missed. The result is the Statement of Applicability (SoA): for each Annex A control, whether it’s applied, the justification and the implementation status. The risk treatment plan and residual risks are approved by the risk owners. Detailed guide: What is an SoA?
The risk treatment plan is an action list: an owner, target date and budget for each control. If the line “MFA will be rolled out” doesn’t say who will do it, on which systems and by when, that line is a wish, not a plan.
Stage 5 — Documentation and putting controls in place
The document set is written according to the selected controls. Each procedure must define its owner, the implementation steps, the system used and the record to be produced. The sentence “Access rights are reviewed every three months” only means something together with the review list, approval and correction record. Details on which documents are needed are in the ISO 27001 documents guide.
For technical controls, the record often comes from the system itself: the backup software’s job report, the patch management tool’s compliance report, the list of privileged groups taken from Active Directory, the firewall rule review note. Saving these outputs once a month in the same folder with the same naming keeps audit week calm. No record, no control.
Stage 6 — Measurement, internal audit and management review
After controls have been operated for a few months, performance is measured (e.g. the rate of critical patches applied on time, access reviews completed on time), an independent internal audit is carried out and the results are presented to management review. Certification bodies expect at least one internal audit and management review cycle to have been completed by the Stage 2 audit.
An internal auditor can’t audit their own work. If the ISMS manager also runs access management, someone else must audit that area. In small organizations, this is solved with an outsourced internal audit. Findings in the internal audit aren’t a problem either; a record showing findings were closed with corrective action is the best evidence that the system works.
Stage 7 — The certification audit
The Stage 1 audit assesses scope, documentation and readiness; Stage 2 assesses the implementation and effectiveness of controls. Findings are closed with corrective actions; after certification, the cycle of surveillance audits and recertification begins.
In Stage 2, the auditor samples: a random employee’s onboarding record, a random server’s patch status, a random month’s log review note. Use the same method in your preparation; pick your own samples and confirm the record really exists.
What does a realistic project plan look like?
| Stage | Main output | Who takes part? |
|---|---|---|
| Context and scope | Context analysis, scope statement | Top management, ISMS manager |
| Risk and SoA | Methodology, risk register, SoA, risk treatment plan | Process owners, IT |
| Documentation | Policies, procedures, forms and records | Document owners |
| Operation | Control records, training | All units |
| Audit cycle | Internal audit report, management review minutes | Internal auditor, management |
The duration depends on the breadth of the scope, the number of locations and the organization’s current maturity. A realistic timeline can only be drawn up after a gap analysis.
Keeping the system alive after setup
The most common problem after certification is that record production drops. The project team disbands, monthly controls are skipped, and two weeks before the first surveillance audit someone tries to produce records retroactively. The auditor can tell from the dates. An annual ISMS calendar (monthly controls, quarterly risk and KPI reports, an annual internal audit and management review) reduces this risk. If there’s no internal resource, outsourced ISMS responsibility is an option.
Frequently asked questions
Do you need a full-time information security specialist to set up an ISMS?
Not necessarily. The standard expects roles and responsibilities to be assigned and competence to be ensured. In medium-sized organizations, the ISMS manager role can be given to an internal employee as an additional duty or covered by an outsourced specialist; what matters is that the role is owned and enough time is set aside for it.
Can the scope be limited to a single location?
Yes, the scope can be limited by location, process or service. However, how interfaces with units outside the scope (e.g. shared IT infrastructure, central HR) are managed must be defined, and the scope must include the service the customer expects.
How long should controls be operated before certification?
The standard doesn’t give a specific period; however, the Stage 2 audit expects records showing controls are applied, along with at least one internal audit and management review. That’s why, in practice, controls need to run long enough to produce several months of records.
If you want to draw up a scope and timeline for your organization, take a look at our ISO 27001 ISMS setup service or request a proposal directly.