An independent information publicationDirective (EU) 2022/2555 · The NIS2 DirectiveNot an official publication
The Independent NIS2 Desk
NIS2 Compliance
Independent information on the NIS2 Directive, for the organisations it covers.
HomeArticlesSecurity Control Testing Under NIS2: A Practical Assessment Method
Security Testing Audit / 2026-10-09

Security Control Testing Under NIS2: A Practical Assessment Method

Security control testing examines whether particular safeguards are designed appropriately and operate as expected. Under NIS2, a practical testing programme connects each measure to an identified risk, chooses a method that can test the intended outcome and follows up failures.

Start with the control objective, not the test tool

Article 21(2)(f) of Directive (EU) 2022/2555 requires policies and procedures to assess the effectiveness of cybersecurity risk-management measures. Before testing, state the objective of each measure. For instance, an access-control system should prevent unauthorised access to a service, not merely produce a configuration document.

Use the organisation's risk assessment and system inventory to decide which controls merit priority. A high-impact service with repeated incidents may need deeper testing than a low-risk environment already supported by reliable evidence.

Distinguish design and operating effectiveness

A design review asks whether the proposed measure is capable of meeting its objective. An operating test examines evidence that it actually works. A security policy may require rapid account removal, while access logs show that former contractors retained credentials for several weeks.

Choose representative evidence from the review period. For automatic safeguards, inspect settings and event records. For procedural measures, sample completed cases, approvals and exceptions. Document the tested population and any limits that prevent wider conclusions.

Match the method to the safeguard

Backup controls call for restoration exercises. Access governance can be tested through entitlement sampling and joiner, mover and leaver records. Detection controls may require approved event simulations and examination of alert handling. Supplier controls can be tested through risk reviews, contractual commitments and monitoring evidence.

A single penetration test cannot replace all these activities. Similarly, a successful tabletop exercise does not establish that a backup system can restore production data. Use a mixture of technical and operational assessment methods appropriate to the stated objective.

Plan testing around operational safety

Define test permissions, owners, change windows, stop conditions and escalation contacts before any potentially disruptive exercise. Production systems, critical infrastructure and systems containing sensitive personal information require safeguards against unplanned service interruption or data exposure.

Record who approved the procedure, the test environment and any deviations. If a test cannot safely be performed, document an alternative assurance method and the uncertainty that remains. Do not claim full effectiveness where only partial evidence exists.

Turn test outcomes into risk decisions

Record the observed result against a defined acceptance criterion. For failures, identify affected services, probable causes, immediate mitigations and accountable owners. Distinguish a control malfunction from missing evidence; both may require action, but their implications are different.

Repeat relevant tests after correction and reassess the scope when infrastructure changes. Provide management with a clear view of significant failures, recurring weaknesses and any risks accepted pending remediation. Review national and sector-specific NIS2 requirements when determining mandatory testing detail.

Points of clarification

Does NIS2 prescribe one universal assessment frequency?

No. Article 21 does not set a single frequency for every entity and technique. The programme should be risk-based and reflect applicable national or sector-specific requirements.

Can evidence of completed activity prove a control is effective?

Not by itself. Effectiveness requires evidence that the measure achieves its intended security outcome in the relevant environment.

What should management receive?

Management should receive the scope, material findings, exposure, limitations, corrective owners and evidence of verified closure.

Further entries on this subject

The texts relied on

General editorial guidance only: it does not determine your organisation's legal obligations. See our editorial policy for how this site works and corrects errors.