Define an outcome for every important control
Begin with the risk that a measure is supposed to address. Multifactor authentication is intended to reduce the chance of account compromise through stolen passwords. A useful review therefore considers the coverage of relevant accounts, bypass routes and authentication exceptions, rather than just reporting the number of users enrolled.
For each critical control, record an owner, expected outcome, measurement method and evidence source. Identify the services in scope and any environments excluded. This prevents a positive result in one business unit being presented as assurance for the whole entity.
Separate activity metrics from outcome evidence
The percentage of patch tickets closed is an activity metric. It is useful, but it does not tell management whether critical vulnerabilities remain exposed. Combine process measures with results such as the time high-risk vulnerabilities remained exploitable and whether retests confirmed remediation.
An incident-response team may report the number of exercises held. To evaluate effectiveness, examine detection time, escalation quality, decision records and whether actions from previous exercises improved real responses. Quantitative scores should be explained rather than treated as absolute proof.
Use a compact set of defensible indicators
Practical indicators might cover privileged accounts reviewed on time, critical assets included in vulnerability assessments, backup restores completed against internally defined recovery objectives and high-priority corrective actions verified before closure. These are examples of management measures, not universal NIS2 legal thresholds.
For each indicator, define the numerator, denominator, review period and data source. Note missing evidence and data-quality concerns. If an asset inventory expands, a temporary fall in scanning coverage may reflect improved discovery rather than deteriorating scanner performance.
Investigate trends and exceptions
A stable monthly percentage can conceal a persistent failure in one critical system. Segment results by risk, service and system type where possible. Look for repeated exceptions, overdue corrective actions and controls that pass routine checks but fail during incidents.
Do not reward teams for suppressing findings or excluding difficult systems. Keep coverage and unresolved uncertainty visible. Interpret measures alongside configuration reviews, incident lessons, audit observations and targeted exercises.
Report measurements as decisions for management
Summarise which controls are effective, which are unproven and where failure could affect critical services. Recommend a response for each material gap, with owners, due dates and a plan to verify correction. Management oversight is more useful when reports distinguish acceptable residual risk from unresolved exposure.
Keep historic definitions and evidence so trends remain comparable after systems or methodologies change. NIS2 is implemented in national law; confirm whether the competent authority or sector-specific requirements prescribe particular measures or reporting arrangements before claiming a mandatory score.
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.