한국어
A.5.12

Classification of information

Organizational controlsCheckpoints 6Evidence 6Cases 6

Control objective

This control requires information to be classified according to its confidentiality/integrity/availability requirements and its legal/business importance, using consistent criteria, so that a level of protection proportionate to each class can be applied. A clear classification result lets the organization decide follow-on safeguards such as access control, encryption, storage/transmission, and retention/disposal in proportion to risk. The aim is to prevent both wasted resources from over-protection and exposure from under-protection, while ensuring the whole organization handles information against the same baseline.

Key checkpoints

  1. Is the classification scheme (class definitions, classification criteria, reclassification procedure) documented and approved/published at the organizational level?
  2. Do the criteria reflect not only confidentiality but also integrity/availability and legal/regulatory requirements (such as personal data)?
  3. Is each asset in the information asset inventory assigned an owner and a classification level, and kept up to date?
  4. Are handling rules per classification level (access/storage/transmission/printing/disposal) defined and applied in actual operations?
  5. Are reclassification and declassification procedures operated as information ages or its context changes?
  6. When information is exchanged with external parties, is a basis for interpreting and mapping classification levels agreed upon?

Implementation guidance

  • Design the class scheme to be simple and fit the organization's size and business (for example, 3 to 4 levels such as public/internal/restricted/confidential). Too many levels reduce practical compliance.
  • Describe in each class definition the impact from a confidentiality/integrity/availability perspective (expected damage from disclosure/alteration/disruption), so staff have a basis for deciding the level.
  • Link classification to the information asset inventory (A.5.9) so that asset owners assign/approve the class, and specify a procedure for periodic re-review.
  • Identify legally protected data (personal data, unique identification information such as resident registration and passport numbers, trade secrets) with separate tags or labels so they connect automatically to the handling rules.
  • Define how to mark (label) classification per medium (documents/files/email/databases), and where possible turn the class into metadata using DLP and document security tools for enforcement.
  • Assign responsibility for the initial classification of newly created and collected information, and set a default (how to treat unclassified items) to prevent gaps.
  • Define the approval authority and recording method for reclassification/declassification/disposal, so aged information is neither over-protected nor left unreviewed.

Evidence

  • Classification criteria and class definition document, and handling rules per class (policy/guideline)
  • Information asset inventory (including owner, classification level, and last review date per asset)
  • Labelling examples (document header/footer, file properties, email classification tags)
  • Reclassification and declassification approval records
  • Training material and completion records related to information classification
  • DLP/document security policy configuration screens (basis for mapping controls to each class)

Nonconformity examples

  • Classification levels are defined, but actual assets carry no level, so most information is left unclassified.
  • The class definitions cover only confidentiality and do not reflect integrity/availability or legal requirements (such as personal data).
  • The asset inventory and the classification scheme are not linked, so owners and levels are not kept current.
  • Restricted documents carry no label, so no level-based control is applied when they are taken outside or sent by email.
  • No reclassification/declassification procedure exists, so aged confidential information stays over-protected or, conversely, is left unreviewed.
  • No classification-mapping basis is agreed with external partners, so a counterpart handles the information at a lower level of protection than intended.

Source

Source/limitation: Control numbers, titles, and theme classification are based on the publicly available list of ISO/IEC 27001:2022 Annex A. The explanatory text (control objective, key checkpoints, implementation guidance, evidence, nonconformity examples) and the attribute classification are original material written by this collection for practical reference; they are not the normative text of the ISO/IEC 27001:2022 or 27002:2022 standards. For certification, verify against a licensed copy of the standard.
On this page
Item info
SetA.5
Key checkpoints6
Evidence6
Cases6