Join our Newsletter — 33% off our NHI Course
Home› Guides› ISO 27001 Annex A.8 Technological controls: What Auditors…
Guide Regulations & Frameworks

ISO 27001 Annex A.8 Technological controls: What Auditors Expect for All 34 Controls

PDF for newsletter subscribers · free to read online

← All guides
By NHI Mgmt Group, edited by Lalit Choda Updated 14 November 2026 16 min read
On this page

Annex A.8 is where an ISMS stops being paperwork and starts touching real systems, and it is where most certification audit nonconformities on technical controls are raised. The 34 technological controls in ISO/IEC 27001:2022 Annex A cover endpoints, privileged access, vulnerabilities, configuration, logging, networks, cryptography and the development life cycle, each in a single line that leaves the implementation open. This guide works through every A.8 control by sub-group and states what each one is for in plain words, what a control owner should build, and the evidence an auditor will normally sample. It is written for ISMS managers, control owners and internal auditors preparing for a stage 2 or surveillance audit. Control intent is paraphrased here; the normative wording is in ISO/IEC 27001:2022 and the guidance in ISO/IEC 27002:2022, both published by ISO and IEC, and should be read in the standards themselves. For the full control catalogue and attributes, see the ISO 27002 Controls Guide.

Key takeaways

  • Evidence beats policy. For A.8 an auditor expects to see the control running: a configuration export, a scan report, a log query, a change ticket, not only a policy statement.
  • Seven of the 34 are new in 2022. 8.9 configuration management, 8.10 information deletion, 8.11 data masking, 8.12 data leakage prevention, 8.16 monitoring activities, 8.23 web filtering and 8.28 secure coding are where immature evidence shows up most often.
  • Several controls lean on organisational ones. 8.2 to 8.5 implement the access rules in 5.15 to 5.18; 8.30 depends on supplier controls 5.19 to 5.22; 8.32 is the technical side of change governance.
  • Cloud inheritance is allowed, but must be argued. A provider can operate 8.14 or 8.20 for you, yet the Statement of Applicability must say so and you must review the provider's assurance.
  • Machine identities sit inside A.8. Service accounts, API keys and certificates fall under 8.2, 8.5, 8.15 and 8.24; ignoring them is a frequent audit gap.

How A.8 differs from the rest of Annex A

ISO/IEC 27002:2022, edition 3, groups its 93 controls into four themes, and the technological theme in clause 8 holds 34 of them, numbered 8.1 to 8.34. ISO lists no amendment to ISO/IEC 27002:2022. ISO/IEC 27001:2022 itself was amended in February 2024 by Amendment 1 (climate action changes), which does not change the Annex A control list. The organisational theme (A.5) says what the organisation decides; A.8 says what its systems must do. That makes A.8 the theme where an auditor most often asks to watch a screen share or receive an export rather than read a document.

Sub-groupControlsTypical ownerMain evidence type
Endpoint and access8.1 to 8.5IT operations, IAM teamMDM compliance reports, PAM records, access review outputs, MFA coverage
Resilience and hygiene8.6 to 8.8, 8.13, 8.14Infrastructure, vulnerability managementCapacity dashboards, EDR coverage, scan and patch reports, restore test records
Data protection8.10 to 8.12, 8.24Data owners, security engineeringDeletion certificates, masking rules, DLP policy and incident logs, key inventory
Configuration and change8.9, 8.18, 8.19, 8.32Platform engineering, change managerBaselines, drift reports, change tickets, software allow lists
Logging and monitoring8.15 to 8.17SOC, SIEM engineeringLog source inventory, use cases, alert handling records, NTP configuration
Network8.20 to 8.23Network teamNetwork diagrams, firewall rule reviews, segment design, web filter policy
Development life cycle8.25 to 8.31, 8.33, 8.34Engineering, AppSecSDLC standard, security requirements, code review and test results, environment separation

Endpoint and access controls: 8.1 to 8.5

ControlIntent in plain wordsEvidence auditors usually sample
8.1 User endpoint devicesProtect information held on or reached from laptops, phones and other user devicesDevice standard; MDM or UEM compliance report showing encryption, screen lock and OS version; BYOD rules
8.2 Privileged access rightsRestrict and manage elevated rights so they are granted, used and removed deliberatelyPrivileged account inventory (human and machine), approval records, PAM session logs, quarterly review sign off
8.3 Information access restrictionEnforce the access rules agreed under 5.15 inside each systemRole definitions, sample of user to role assignments, results of the last access review
8.4 Access to source codeControl who can read and change source code, tools and librariesRepository permission export, branch protection settings, removal of leavers from the source platform
8.5 Secure authenticationUse authentication strength that matches the risk of the system and the accessMFA coverage report, authentication standard, conditional access or equivalent policy, lockout settings

What to build first: one privileged account register that includes service accounts and automation identities, then MFA on every administrative and remote access path. Auditors often pick five privileged accounts at random and ask who approved each and when it was last reviewed. See the Privileged Access Management Guide and the MFA Guide.

Resilience and hygiene: 8.6 to 8.8, 8.13 and 8.14

ControlIntent in plain wordsEvidence auditors usually sample
8.6 Capacity managementMonitor and plan resource use so services stay availableCapacity thresholds and alerts, trend reports, a capacity review minute
8.7 Protection against malwarePrevent, detect and recover from malware, backed by user awarenessEDR or anti-malware coverage against the asset inventory, alert handling sample, exclusion list review
8.8 Management of technical vulnerabilitiesFind vulnerabilities, judge exposure and fix them on timeScan scope versus inventory, remediation SLAs, overdue items with risk acceptance, penetration test follow up
8.13 Information backupKeep and test backup copies in line with an agreed backup policyBackup policy, job success reports, restore test records with dates and outcomes
8.14 Redundancy of information processing facilitiesBuild in enough redundancy to meet availability requirementsAvailability requirements per service, architecture showing redundancy, failover test results

For 8.8 the question that sinks audits is "how do you know the scanner sees everything?" Reconcile scan targets with the asset inventory from 5.9 at least monthly and keep the reconciliation. For 8.13 a restore test is evidence; a green backup job is not. See the Vulnerability Management Guide, the Capacity Management as a Security Control guide and the Backup and Recovery Testing Checklist.

Data protection: 8.10 to 8.12 and 8.24

ControlIntent in plain wordsEvidence auditors usually sample
8.10 Information deletionDelete information when it is no longer needed, on systems, devices and with suppliersRetention schedule, deletion job logs, disposal certificates, supplier deletion confirmations on contract exit
8.11 Data maskingHide or transform sensitive data where full values are not needed, in line with access policy and legal dutiesMasking or tokenisation rules per data set, non-production data handling standard, sample of masked records
8.12 Data leakage preventionApply measures to systems, networks and devices that handle sensitive information to prevent leakageDLP policy scope, rule set mapped to classification labels, incident and false positive records
8.24 Use of cryptographyDefine and apply rules for cryptography, including key managementCryptography standard (algorithms, key lengths), key and certificate inventory, rotation records, encryption at rest and in transit evidence

These controls depend on classification under 5.12. Without labels, 8.12 rules have nothing to key on and 8.11 cannot say which fields are sensitive. A sensible order is classification, then retention and deletion, then masking for non-production, then DLP on the highest risk channels. See the Data Loss Prevention (DLP) Guide, the Data Retention and Secure Deletion guide and the Cryptographic Key Management guide.

Configuration, change and software control: 8.9, 8.18, 8.19 and 8.32

  • 8.9 Configuration management: establish, document, apply, monitor and review secure configurations for hardware, software, services and networks. Evidence: hardening baselines (for example CIS Benchmarks or vendor baselines) with your deviations recorded, and a drift report showing what is out of baseline today. See the Configuration Drift and Baseline Enforcement Guide.
  • 8.18 Use of privileged utility programs: restrict and control tools that can override system and application controls. Evidence: list of such utilities, who may run them, and logging of their use.
  • 8.19 Installation of software on operational systems: manage software installation on production systems securely. Evidence: an allow list or approved software catalogue, removal of local admin rights, and installation logs.
  • 8.32 Change management: subject changes to processing facilities and systems to change control. Evidence: a sample of changes with risk assessment, approval, testing and rollback plan, plus emergency changes reviewed after the fact. See the Change Management Security guide.

Infrastructure as code helps all four: a merged pull request with review is both the change record (8.32) and the configuration baseline (8.9). Say so in the implementation statement so the auditor samples the repository rather than a ticketing tool.

Logging and monitoring: 8.15 to 8.17

ControlIntent in plain wordsEvidence auditors usually sample
8.15 LoggingProduce, store, protect and analyse logs of activities, exceptions and faultsLogging standard naming required events, log source inventory, retention settings, access restrictions on logs
8.16 Monitoring activitiesWatch networks, systems and applications for anomalous behaviour and act on itDetection use case list, alert triage records, escalation into incident management (5.25, 5.26)
8.17 Clock synchronizationSynchronise clocks to approved time sources so events can be correlatedNTP configuration for servers, network devices and cloud services; drift monitoring

A common question for 8.16 is "show me an alert from last month and what happened next." Keep a handful of closed alerts with timestamps for each stage. See the Log Management and Retention guide and the SIEM Guide.

Network controls: 8.20 to 8.23

  • 8.20 Networks security: secure, manage and control networks and network devices. Evidence: current network diagram, device hardening, management plane access restrictions.
  • 8.21 Security of network services: identify and monitor the security mechanisms, service levels and requirements of network services, including those bought in. Evidence: service agreements with security terms, SD-WAN or ISP security configuration.
  • 8.22 Segregation of networks: separate groups of services, users and systems. Evidence: segment design with the rationale, firewall rule review, test showing a blocked path. See the Network Segmentation Guide.
  • 8.23 Web filtering: manage access to external websites to reduce exposure to malicious content. Evidence: filter policy by category, coverage for remote users, block logs. See the DNS Filtering vs Secure Web Gateway guide.

Development life cycle: 8.25 to 8.31, 8.33 and 8.34

ControlIntent in plain wordsEvidence auditors usually sample
8.25 Secure development life cycleSet rules for building software and systems securelySDLC standard with security gates, training records for developers
8.26 Application security requirementsIdentify and approve security requirements when building or buying applicationsRequirements template (for example based on OWASP ASVS), sample of approved requirements for a recent project
8.27 Secure system architecture and engineering principlesDefine and apply secure engineering principles to all developmentArchitecture principles, design review or threat model records
8.28 Secure codingApply secure coding principles in software developmentCoding standard, SAST and dependency scan results, peer review enforcement
8.29 Security testing in development and acceptanceDefine and run security tests within the life cycleTest plans, DAST or penetration test reports, release sign off showing findings handled
8.30 Outsourced developmentDirect, monitor and review development done by othersContract security clauses, supplier code review or test evidence, acceptance testing
8.31 Separation of development, test and production environmentsKeep environments separate and securedSeparate accounts or subscriptions, access lists per environment, deployment pipeline controls
8.33 Test informationSelect, protect and manage information used for testingRules on production data in test, masking evidence (links to 8.11), approvals for exceptions
8.34 Protection of information systems during audit testingPlan and agree audit tests on operational systems to limit disruptionAgreed scope and timing for technical audits and penetration tests, read only access for auditors

Organisations that only buy software can often scope 8.25 to 8.29 narrowly, but 8.26 still applies to procurement and 8.30 to any customisation. See the Secure Software Development Lifecycle (SSDLC) Guide and the Secure Coding Guide.

Where identity fits

Non-human identities appear across A.8 even though no control names them. Service accounts and workload identities are privileged under 8.2; API keys, tokens and certificates are authentication information protected under 8.5 and governed by 5.17; their activity belongs in the 8.15 log sources; and the keys behind them fall under 8.24. Pipelines that deploy code hold credentials that bridge 8.31 environments. Put machine identities in the same privileged register and review cycle as people, with owners and expiry. See the Service Account Security Guide, the Secrets Management Guide and the NHI and AI Agent Audit Evidence guide.

Rolling out and evidencing A.8

  1. Confirm applicability: mark each of the 34 controls in the Statement of Applicability as applicable, inherited from a named provider, or excluded with a reason. See the ISO 27001 Statement of Applicability Template.
  2. Assign one owner per control: a named role, not a team, with the evidence source written next to it.
  3. Write implementation statements: two to four sentences per control naming the tools, scope and frequency. Avoid restating the control wording.
  4. Collect a baseline of evidence: run each report once, store it with the date, and note gaps as risks or corrective actions.
  5. Close the new controls first: 8.9, 8.10, 8.11, 8.12, 8.16, 8.23 and 8.28 usually have the least mature evidence.
  6. Test with an internal audit: sample like a certification auditor would, for example five changes, five privileged accounts and one restore. See the ISMS Internal Audit and Management Review guide.
  7. Automate collection: schedule exports so evidence for surveillance audits is a query, not a project. If an automated control breaks, fall back to the manual record and log the gap as a nonconformity rather than hiding it.

Decision guidance: small, large and cloud-first

SituationPractical approachWatch out for
Small organisation, mostly SaaSLean on provider controls for 8.14, 8.20, 8.21; focus on 8.1, 8.2, 8.5, 8.12, 8.13Assuming SaaS backups cover you; check the shared responsibility terms
Large, hybrid estateSplit A.8 owners by platform; one evidence register; automate 8.8, 8.9, 8.15Inconsistent baselines between business units
Software companyTreat 8.25 to 8.31 as the core of the audit; use pipeline evidenceProduction data copied to test without 8.33 approval
Buyer of software onlyScope development controls to procurement and configurationExcluding 8.26 and 8.30 entirely despite customisations

Detection and measurement

  • EDR and MDM coverage: percentage of inventory assets reporting in the last 7 days (8.1, 8.7).
  • Critical vulnerability age: median days to remediate critical findings against the SLA (8.8).
  • Baseline compliance: percentage of systems matching the approved configuration (8.9).
  • Restore success: share of scheduled restore tests that met the recovery objective (8.13).
  • Log source coverage: percentage of in-scope systems sending required events to the central log store (8.15).
  • Privileged accounts reviewed: share of human and machine privileged accounts reviewed in the last cycle (8.2).

Common failure patterns

  • Policy only: a statement for 8.16 with no detection use cases or alert records behind it.
  • Scanner blind spots: 8.8 evidence that covers servers but not cloud workloads, containers or network devices.
  • Machine accounts missing: 8.2 reviews that list people only, while long-lived service accounts with admin rights go unreviewed.
  • Untested backups: 8.13 evidence made only of job success emails.
  • Inheritance without proof: controls marked as provider-operated with no reviewed provider certificate or report under 5.22.
  • Production data in test: 8.33 breached by routine database copies to development.

Framework mapping

NIST publishes an ISO/IEC 27001:2022 crosswalk for SP 800-53 Rev. 5 through its OLIR catalogue and warns that mappings are not always one to one. The table shows closely related controls for orientation, not equivalence.

A.8 groupNIST SP 800-53 Rev. 5NIST CSF 2.0
8.2, 8.3, 8.5AC-6, AC-3, IA-2, IA-5PR.AA-01, PR.AA-05
8.7, 8.8SI-3, RA-5, SI-2ID.RA-01, PR.PS-02
8.9, 8.19, 8.32CM-2, CM-6, CM-11, CM-3PR.PS-01
8.13, 8.14CP-9, CP-10PR.DS-11, PR.IR-03
8.15 to 8.17AU-2, AU-6, AU-8, SI-4PR.PS-04, DE.CM-01, DE.CM-09
8.20, 8.22SC-7PR.IR-01
8.24SC-12, SC-13, SC-28, SC-8PR.DS-01, PR.DS-02
8.25 to 8.29SA-3, SA-8, SA-11, SA-15PR.PS-06

Practitioner checklist

  • Record applicability, owner and evidence source for all 34 A.8 controls in the Statement of Applicability.
  • Include service accounts, API keys and certificates in the 8.2 privileged register and review cycle.
  • Reconcile vulnerability scan scope with the asset inventory monthly and keep the result.
  • Keep dated restore test records for each critical service.
  • Publish hardening baselines with recorded deviations and a current drift report.
  • Maintain a log source inventory and at least a few closed alerts per quarter as 8.16 evidence.
  • Ban unmasked production data in test unless an approved exception exists.
  • Run an internal audit that samples A.8 the way the certification body will.

Standards and references

Related NHI Mgmt Group resources: ISO 27001 Certification Guide · ISO 27001 · ISO 27001 Annex A.5 Organisational controls · NIST SP 800-53 to ISO 27001 Mapping Guide

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.