Join our Newsletter — 33% off our NHI Course
Home› Guides› ISO/IEC 27003 and 27004 Guide: Building an ISMS…
Guide Regulations & Frameworks

ISO/IEC 27003 and 27004 Guide: Building an ISMS Metrics Programme That Passes Audit

PDF for newsletter subscribers · free to read online

← All guides
By NHI Mgmt Group, edited by Lalit Choda Updated 14 November 2026 14 min read
On this page

ISO/IEC 27001 clause 9.1 requires you to measure whether your ISMS works, but it does not say how, and that gap is where many certification audits find weak evidence. Two companion standards fill it. ISO/IEC 27003 explains each management system requirement of ISO/IEC 27001 in plainer terms, and ISO/IEC 27004 gives a full method for monitoring, measurement, analysis and evaluation. This guide focuses on the measurement side: how to turn 27004's process into a small, defensible set of measures, who owns them, how auditors test them and where identity and non-human identity measures belong. It uses 27003 as implementation context rather than repeating ISO/IEC 27001 clause by clause; for certification scope, risk treatment and Annex A, start with the ISO 27001 Certification Guide. It is for ISMS managers, internal auditors and security teams preparing for certification or surveillance audits.

Key takeaways

  • Both standards are guidance, not certifiable. You are certified against ISO/IEC 27001; 27003 and 27004 help you meet its requirements and explain your choices to an auditor.
  • Both editions still reference ISO/IEC 27001:2013. ISO lists ISO/IEC 27003:2017 and ISO/IEC 27004:2016 as published and due for revision, so map their clause references to the 2022 edition yourself.
  • 27004 separates performance from effectiveness. Performance measures show planned activity was done; effectiveness measures show it changed security outcomes. Auditors look for both.
  • Start from information needs, not from available data. The 27004 process begins with what management needs to decide, then builds measures, procedures, analysis and evaluation around it.
  • Fewer measures, better evidence. Ten well-defined measures with owners, targets and retained results beat fifty dashboard tiles nobody reviews at management review.

What it is and who it applies to

Both standards come from ISO/IEC JTC 1/SC 27 and belong to the ISO/IEC 27000 family. Neither contains requirements you can be certified against; both are written for any organisation that operates or is building an ISMS, of any type or size.

ItemISO/IEC 27003ISO/IEC 27004
Full titleInformation technology, Security techniques, Information security management systems, GuidanceInformation technology, Security techniques, Information security management, Monitoring, measurement, analysis and evaluation
Current edition2017 (second edition, March 2017)2016 (second edition, December 2016)
Written againstISO/IEC 27001:2013ISO/IEC 27001:2013, clause 9.1
Status on iso.org (checked November 2026)Published, stage 90.92 (to be revised)Published, stage 90.92 (to be revised); a revision is at Draft International Standard (DIS) stage
What it gives youExplanation and guidance for every main-body requirement of ISO/IEC 27001 (clauses 4 to 10)A measurement rationale, measure types, a six-step process and example measurement constructs
Who uses itISMS implementers and managersISMS managers, control owners, internal audit, management

Revisions in progress: the ISO/IEC 27001 Security site reports that the third edition of 27003 has reached Draft International Standard (DIS) stage with publication planned for 2027, possibly 2028, and that a separate part on implementation guidance started in 2025. It reports the 27004 revision will align references with the 2022 editions of ISO/IEC 27001, 27002 and 27005 and add example measures for organisations of all sizes. Draft content can change, so plan on the published editions until revisions are issued.

The 2013 to 2022 gap is manageable. The ISO/IEC 27001 main-body clause numbers for performance evaluation did not move, so 27004's guidance on clause 9.1 still lands on clause 9.1. What changed is Annex A, now aligned with ISO/IEC 27002:2022, so any control-level measures and example constructs need control references updated. See the ISO 27001 guide on the 2022 control set.

Structure and key requirements

ISO/IEC 27003: how each clause is explained

ISO/IEC 27003:2017 mirrors ISO/IEC 27001 clauses 4 to 10. For each requirement it restates the required activity, explains what it implies, gives guidance on how to meet it, and adds other information such as examples and cross-references. An annex describes an information security policy framework. It does not expand on Annex A controls: control guidance sits in ISO/IEC 27002. The 2017 edition dropped the project-style implementation plan of the 2010 edition, so it reads as an interpretation aid rather than a step-by-step project manual.

Use 27003 when you need to justify an interpretation to an auditor, for example how far "interested parties" in clause 4.2 should extend, or what counts as adequate competence evidence in clause 7.2. Its clause 9.1 guidance is short; 27004 is the detailed companion.

What clause 9.1 asks of you

ISO/IEC 27001:2022 clause 9.1 requires the organisation to evaluate information security performance and ISMS effectiveness and to determine: what needs to be monitored and measured, including processes and controls; the methods for monitoring, measurement, analysis and evaluation, which should give comparable and reproducible results to be valid; when monitoring and measuring is performed; who performs it; when results are analysed and evaluated; and who analyses and evaluates them. Documented information must be available as evidence of the results.

ISO/IEC 27004: the structure

ClauseContentHow to use it
5 RationaleNeed for measurement, fulfilling ISO/IEC 27001, validity of results, benefitsJustify the programme to management and set validity rules
6 CharacteristicsWhat to monitor, what to measure, when, and whoAnswer the clause 9.1 a) to f) questions directly
7 Types of measuresPerformance measures and effectiveness measuresBalance the measure set
8 ProcessesIdentify information needs through to review, improvement and retention of documented informationOperating procedure for the programme
Annex AAn information security measurement modelShared vocabulary for measures, attributes and indicators
Annex BMeasurement construct examples (35 according to the ISO/IEC 27001 Security site)Starting ideas, to be adapted, not adopted wholesale
Annex CAn example of free-text form measurement constructionTemplate for writing a construct in prose

Performance versus effectiveness: 27004 defines performance measures as expressing planned results in terms of the planned activity, such as headcounts, milestones met or the degree to which controls are implemented. Effectiveness measures express the effect that carrying out planned activities has on information security objectives. "Percentage of privileged accounts reviewed this quarter" is performance; "privileged accounts found with excessive rights at review, trending down" is closer to effectiveness. For an alternative classification (implementation, effectiveness, efficiency, impact) from NIST SP 800-55 Vol. 1, see the Security Metrics and KPIs Programme Guide.

What to monitor and measure: 27004 clause 6.2 lists candidates such as ISMS process implementation, incident management, vulnerability management, configuration management, awareness and training, access control and event logging, audit, risk assessment and treatment, third-party risk, business continuity, physical security and system monitoring. Clause 6.3 adds management system processes (planning, leadership, risk management, policy management, resources, communication, management review, documentation, auditing) and notes that controls can be measured on attributes such as how much they reduce likelihood or consequence, how many events they handle before failing and how quickly they detect.

Roles: clause 6.5 names roles that may be combined: measurement client, measurement planner, measurement reviewer, information owner, information collector, information analyst and information communicator. In a small organisation two or three people may hold all seven; the point is that every measure has a named person for each job.

Identity and NHI requirements

Neither standard mentions non-human identities, but access control and logging are named measurement candidates in 27004, and identity controls produce the most countable evidence in most ISMSs. Good identity measures connect to ISO/IEC 27001:2022 Annex A controls A.5.15 (access control), A.5.16 (identity management), A.5.17 (authentication information), A.5.18 (access rights) and A.8.2 (privileged access rights).

Information needMeasureTypeAnnex A link
Do we know every identity that can reach in-scope systems?Percentage of human and non-human identities in scope with a named accountable ownerPerformanceA.5.16
Are leaver and decommission processes working?Median hours from HR leave date or workload retirement to account disablementEffectivenessA.5.18
Are secrets kept short-lived?Percentage of service account credentials, API keys and tokens rotated within policy agePerformanceA.5.17
Is privilege shrinking?Excess entitlements removed per access review cycle, and the share found again at the next cycleEffectivenessA.5.18, A.8.2
Is strong authentication in place?Percentage of privileged human accounts using phishing-resistant MFA; percentage of workloads using federated or certificate-based credentials instead of static secretsPerformanceA.8.5
Do identity incidents recur?Incidents caused by a credential already flagged as stale or over-privilegedEffectivenessA.5.26, A.5.27

For a deeper identity measure set, see the Identity Security Metrics Guide and the NHI Programme Metrics guide. Measures need a reliable inventory as data source; the How to build a service account inventory from Active Directory guide covers one common starting point.

Implementation steps

The order below follows 27004 clause 8.1 steps a) to f), plus the review and retention activities in clauses 8.8 and 8.9.

  1. Identify information needs (8.2). Interview top management and risk owners: what decisions will measurement inform? Draw needs from information security objectives (clause 6.2 of ISO/IEC 27001), the risk treatment plan and the previous management review. Write each as a question.
  2. Find existing practices that can answer them (8.3.2). Inventory data you already produce: IAM reports, vulnerability scanner output, ticket systems, training records, SIEM. Reuse beats new collection.
  3. Develop and document measures (8.3.3, 8.3.4). For each, record the information need, the formula, the target or decision threshold, the data source, collection and reporting frequency, and the named roles from clause 6.5. Mark each as performance or effectiveness. Prioritise: start with 8 to 15 measures.
  4. Keep management engaged (8.3.5). Agree targets with the measurement client before collection starts; a target set after the data arrives will be challenged at audit.
  5. Establish procedures (8.4). Write how data is collected, stored, protected and validated, including who may change a formula. This is what makes results "comparable and reproducible".
  6. Monitor and measure (8.5). Run the first collection as a pilot cycle and record data quality problems rather than hiding them.
  7. Analyse results (8.6). Compare against targets and previous periods; explain variance; separate data errors from real change.
  8. Evaluate performance and effectiveness (8.7). Decide what the results mean for objectives and risks, and feed conclusions to management review (ISO/IEC 27001 clause 9.3) and corrective action (clause 10.2).
  9. Review the programme (8.8). At least annually retire measures that no longer inform decisions, fix those with poor data and add needs arising from new risks.
  10. Retain and communicate (8.9). Keep results, analysis and evaluation records as documented information, and report in a form each audience can act on.

Small versus large organisations: a 50-person firm can run the programme in a spreadsheet with one ISMS manager as planner, analyst and communicator and the leadership team as client, measuring quarterly. A large enterprise usually needs a measure catalogue with change control, automated collection, and separation between the people who own a control and those who analyse its results, so control owners are not marking their own homework.

Rolling back a bad measure: if a measure proves unreliable, do not quietly drop it. Record the decision, the reason and the replacement in the measure catalogue and report it to management review; an auditor will otherwise find a gap in the trend.

Evidence and assessment

Certification auditors test clause 9.1 by sampling: they pick measures, trace the definition, the data, the analysis and the management response. Typical evidence:

Clause 9.1 elementEvidence an auditor expectsCommon weakness
What is monitored and measuredMeasure catalogue linked to objectives and risksMeasures with no link to any objective
Methods (comparable, reproducible)Written formula, data source and collection procedureFormula changed between periods without record
When and who measuresSchedule and named information collector"Security team" as owner
When and who analyses and evaluatesAnalysis records, evaluation conclusions with dates and namesCharts with no interpretation
Results as documented informationRetained results for each periodLive dashboards with no history
Link to 9.3 and 10.2Management review minutes acting on results; corrective actions raised from missed targetsTargets missed for several cycles with no action

Internal audit should test the measurement programme itself under clause 9.2; the ISMS Internal Audit and Management Review guide covers that cycle and how results enter management review.

Metrics that show the measurement programme itself works:

  • Percentage of information security objectives with at least one effectiveness measure.
  • Percentage of measures reported on schedule with complete data.
  • Share of measures with a documented formula unchanged, or changed under control, across the last four periods.
  • Number of management review decisions or corrective actions traceable to a measurement result per year.
  • Measures retired or redesigned at the annual programme review (a healthy programme changes).

Mapping to other frameworks

ISO/IEC 27001 / 27004 elementNIST CSF 2.0NIST SP 800-53 Rev. 5
Clause 9.1 performance evaluationGV.OV-03 (risk management performance evaluated)PM-6 Measures of Performance
Monitoring of controlsID.IM-01, ID.IM-03CA-7 Continuous Monitoring, PM-31 Continuous Monitoring Strategy
Control effectiveness assessmentID.IM-02CA-2 Control Assessments
Management review using results (9.3)GV.OV-01, GV.OV-02PM-6
Identity measuresPR.AA-01, PR.AA-05AC-2, AC-6, IA-5

NIST SP 800-55 Vol. 1 (December 2024) covers identifying and selecting measures and pairs well with 27004 for organisations that report to both ISO and NIST audiences. For board-level reporting built on top of an ISMS measure set, see the Key Risk Indicators for Cyber Risk Guide; for the cloud estate, see Cloud Security Metrics.

Common pitfalls

  • Copying Annex B wholesale. The examples were written for illustration; the ISO/IEC 27001 Security site describes them as a mixed bunch. Adapt only those that answer a real information need.
  • All performance, no effectiveness. A catalogue full of "percentage completed" measures shows activity, not security. Pair each critical control with an outcome measure.
  • Measuring what is easy. Firewall rule counts and phishing click rates are easy to collect but often answer no management question.
  • Stale clause references. Measures and constructs still citing 2013 Annex A numbers suggest the programme was not reviewed after transition to the 2022 edition.
  • No retained history. Dashboards that overwrite values leave nothing to show auditors for previous periods.
  • Identity blind spots. Counting only human joiners and leavers while service accounts, API keys and workload identities sit outside the measure set.

Practitioner checklist

  • Confirm which editions you use (ISO/IEC 27003:2017, ISO/IEC 27004:2016) and record the mapping to ISO/IEC 27001:2022.
  • Write the information needs as questions, signed off by the measurement client.
  • Build a measure catalogue: need, formula, target, data source, frequency, type, and the clause 6.5 roles.
  • Keep 8 to 15 measures at the start, with at least one effectiveness measure per objective.
  • Include human and non-human identity measures tied to A.5.15 to A.5.18 and A.8.2.
  • Document collection procedures so results are comparable and reproducible.
  • Retain results, analysis and evaluation for every period.
  • Feed results into management review and raise corrective actions for missed targets.
  • Review the catalogue annually and log every retired or changed measure.
  • Track the 27003 and 27004 revisions and update references when they publish.

Standards and references

Related NHI Mgmt Group resources: ISO 27001 Certification Guide · ISMS Internal Audit and Management Review · Security Metrics and KPIs Programme Guide · ISO/IEC 27005 Information Security Risk Guide · KRI vs KPI in Cyber Security

PDF for subscribers

Take this guide with you

Every section, table and checklist in one PDF, with links to the sources and related guides. Subscribe to our weekly NHI & AI Identity Journal and we’ll email you a download link.

Free weekly newsletter

Subscribe to the NHI & AI Identity Journal

The latest on NHI and Agentic AI security – articles, research, breaches, news and events every week.

Bonus 33% off our NHI Course when you subscribe.

NHIMG Editorial Note
Researched and written by the NHI Mgmt Group editorial team using AI tools, under the editorial direction of Lalit Choda. Facts are drawn from the sources listed on this page and checked by our automated quality process. Last updated 14 November 2026. Spotted an error? Use our contact page.
NHI Mgmt Group, the independent authority on Non-Human Identity, IAM and Agentic AI security. nhimg.org

PDF Downloads for Newsletter Subscribers

Get the guide

Enter your details and we’ll email you a download link.

The full guide is free to read online. The PDF is a thank you for subscribing.

Fill in all fields marked * and tick the PDF box to continue.