本文へ移動
cccskills
無料GitHub で公開

ropa-250-exemption

Assesses the GDPR Article 30(5) exemption for organisations under 250 employees. Covers the three exception conditions that negate the exemption: non-occasional processing, risk to data subject rights, and special category data processing. Activate for Art. 30(5), 250 employee exemption, small business RoPA, SME exemption, occasional processing.

インストール方法を見る

含まれるファイル(5)

  • SKILL.md12.7 KB
  • assets/template.md5.1 KB
  • references/standards.md4.8 KB
  • references/workflows.md5.0 KB
  • scripts/process.py14.0 KB

SKILL.md(原文)

インストールする前に、エージェントに与えられる指示の中身を確認できます。

Art. 30(5) Exemption Assessment

Overview

GDPR Article 30(5) provides a limited exemption from the RoPA obligation for organisations with fewer than 250 employees. However, this exemption is narrowly constructed and rarely available in practice. The exemption does not apply when any of three exception conditions are met: (1) the processing is likely to result in a risk to data subjects' rights and freedoms, (2) the processing is not occasional, or (3) the processing includes special category data under Art. 9(1) or criminal conviction data under Art. 10. Since virtually all organisations with regular employee or customer processing engage in non-occasional processing, the EDPB has recommended that all organisations maintain RoPA regardless of size.

Art. 30(5) Text

The obligations referred to in paragraphs 1 and 2 shall not apply to an enterprise or an organisation employing fewer than 250 persons unless the processing it carries out is likely to result in a risk to the rights and freedoms of data subjects, the processing is not occasional, or the processing includes special categories of data as referred to in Article 9(1) or personal data relating to criminal convictions and offences referred to in Article 10.

Three Exception Conditions

The exemption is negated (RoPA is required) if any one of the following three conditions is met. They are disjunctive (OR), not conjunctive (AND).

Exception 1: Processing Likely to Result in a Risk

Test: Does the processing carried out by the organisation result in any risk (not just high risk) to the rights and freedoms of data subjects?

EDPB interpretation: The EDPB Position Paper on Art. 30 (2018) interpreted this condition broadly. Unlike Art. 35 (DPIA), which requires "high risk," Art. 30(5) refers to "a risk" — a lower threshold. The EDPB concluded that virtually all processing of personal data involves some risk to data subjects, making this condition almost always met.

Examples of processing that creates risk:

Processing ActivityRisk to Rights/FreedomsExemption Applies?
Employee payrollFinancial data exposure risk; employment impactNo — risk exists
Customer databaseIdentity theft risk; unwanted marketingNo — risk exists
Website with loginAccount compromise risk; privacy intrusionNo — risk exists
Email marketingUnsolicited communication; profilingNo — risk exists
CCTV at workplacePrivacy intrusion; chilling effectNo — risk exists
Manual paper-only visitor log retained < 24 hoursMinimal riskPotentially yes

Practical outcome: This condition alone makes the exemption unavailable for most organisations. Only processing that involves genuinely negligible risk (such as a one-time, paper-only process with immediate destruction) might satisfy this condition.

Exception 2: Processing Is Not Occasional

Test: Is the processing carried out on a regular or systematic basis, rather than being a one-time or infrequent occurrence?

"Occasional" means: One-time, infrequent, or ad hoc processing that is not part of the organisation's regular business operations. Examples might include a one-time market research survey conducted for a specific project with no repetition planned.

"Not occasional" (regular) processing includes:

Processing ActivityOccasional?Reasoning
Monthly payrollNo — recurringSystematic, monthly processing
Customer order processingNo — ongoingCore business activity, continuous
Employee email systemNo — continuousAlways-on processing
Website analyticsNo — continuousOngoing data collection
Annual customer satisfaction surveyNo — regularAnnually recurring; planned and systematic
One-time data migration projectPossibly yesSingle occurrence, defined end point
Ad hoc legal hold for single litigationPossibly yesOne-time, specific trigger

EDPB guidance: The EDPB stated that any organisation processing employee or customer personal data as part of its ongoing operations is engaged in non-occasional processing. This includes virtually all businesses that employ staff or serve customers.

Practical outcome: Any organisation with employees, customers, suppliers, or a website is engaged in non-occasional processing, negating the exemption.

Exception 3: Special Category or Criminal Data

Test: Does the processing include any special category data under Art. 9(1) or criminal conviction data under Art. 10?

Art. 9(1) special categories:

  • Racial or ethnic origin
  • Political opinions
  • Religious or philosophical beliefs
  • Trade union membership
  • Genetic data
  • Biometric data for identification purposes
  • Health data
  • Data concerning sex life or sexual orientation

Art. 10: Personal data relating to criminal convictions and offences or related security measures.

Common special category processing in small organisations:

Processing ActivitySpecial Category DataExemption Applies?
Employee sick leave recordsHealth data (Art. 9(1))No
Church tax processing (Germany)Religious beliefs (Art. 9(1))No
Background checksCriminal data (Art. 10)No
Biometric access control (fingerprint)Biometric data (Art. 9(1))No
Diversity monitoringRacial/ethnic origin (Art. 9(1))No
Employee emergency contacts with medical needsHealth data (Art. 9(1))No
Trade union membership deductionsTrade union membership (Art. 9(1))No
Employee photos on ID badgesNot biometric unless used for identificationDepends on use

Practical outcome: Almost all organisations process at least health data (sick leave) or religious data (in countries with church tax), making this condition met for most employers.

Exemption Assessment Flowchart

Does the organisation employ fewer than 250 persons?
│
├── No → Art. 30(5) exemption does NOT apply. Full RoPA required.
│
└── Yes → Assess three exception conditions:
    │
    ├── Condition 1: Is processing likely to result in a risk
    │   to data subjects' rights and freedoms?
    │   ├── Yes → Exemption does NOT apply for this processing.
    │   └── No → Continue to Condition 2
    │
    ├── Condition 2: Is the processing NOT occasional?
    │   (i.e., is it regular, recurring, or systematic?)
    │   ├── Yes (not occasional) → Exemption does NOT apply.
    │   └── No (truly occasional) → Continue to Condition 3
    │
    └── Condition 3: Does the processing include special
        category data (Art. 9(1)) or criminal data (Art. 10)?
        ├── Yes → Exemption does NOT apply.
        └── No → Exemption APPLIES for this specific processing
            activity. No RoPA entry required for this activity.

Critical point: The assessment is per processing activity, not per organisation. An organisation might have some processing activities that qualify for exemption and others that do not. In practice, the non-exempt activities (payroll, customer management, employee records) require a RoPA, and once you have a RoPA, you should document all processing for accountability purposes.

Assessment for Helix Biotech Solutions (Hypothetical Small Entity Scenario)

Assume Helix Biotech Diagnostics S.A.S. (recently acquired French subsidiary) has only 45 employees. Can it claim the Art. 30(5) exemption?

Employee Count Check

  • Helix Biotech Diagnostics S.A.S.: 45 employees
  • Threshold: fewer than 250 employees
  • Result: Passes the employee count threshold.

Exception Condition Assessment

Processing ActivityCondition 1: Risk?Condition 2: Not Occasional?Condition 3: Special Category?Exempt?
Employee payrollYes — financial data riskYes — monthlyYes — mutuelle health (health data)No
Customer diagnostic ordersYes — health data riskYes — daily operationsYes — diagnostic results (health data)No
Supplier managementYes — financial data riskYes — ongoingNoNo (fails Conditions 1 and 2)
Website (informational only, no analytics)Minimal riskYes — continuousNoNo (fails Condition 2)
Annual team photoMinimal riskArguably occasional (annual)No (photos not biometric unless used for ID)Possibly exempt
One-time office furniture procurementMinimal riskYes — occasional (one-time)NoPossibly exempt

Conclusion for Helix Biotech Diagnostics S.A.S.: Despite having only 45 employees, the entity must maintain a RoPA because:

  1. It processes health data (diagnostic results, employee health insurance) — Exception 3 is met.
  2. Its core business processing (customer orders, employee management) is non-occasional — Exception 2 is met.
  3. Processing of customer health data creates risk to data subjects — Exception 1 is met.

Only truly one-off, non-sensitive, low-risk activities (like a single furniture procurement) might technically qualify for the exemption, but documenting them in the RoPA anyway costs minimal effort and demonstrates accountability.

EDPB and Supervisory Authority Positions

EDPB Position Paper on Art. 30 (2018)

The EDPB formally recommended that all organisations maintain RoPA regardless of size:

"The EDPB considers that most processing operations carried out by employers or service providers can, in most cases, not be considered as 'occasional'. The EDPB therefore recommends that controllers and processors, including small and medium-sized enterprises, use and maintain records of their processing activities on a voluntary basis, even where they are not obliged to do so."

CNIL (France)

The CNIL provides a simplified RoPA template for small organisations and explicitly states that the Art. 30(5) exemption is "very limited in practice." The CNIL's Facilita RGPD tool is designed for organisations of all sizes, implicitly recommending that small organisations maintain records.

ICO (United Kingdom)

The ICO guidance states: "In practice, it is very likely that you will need to maintain records even if you employ fewer than 250 people because it is unlikely that all of your processing is only occasional."

BfDI (Germany)

The BfDI guidance explicitly states that the exemption "hardly ever applies in practice" (kommt in der Praxis kaum zur Anwendung) because virtually all organisations engage in non-occasional processing of employee and customer data.

AEPD (Spain)

The AEPD recommends that all organisations maintain RoPA as a matter of best practice and accountability, regardless of the Art. 30(5) exemption.

Practical Recommendation

For organisations under 250 employees: maintain a RoPA regardless of the Art. 30(5) exemption.

Reasons:

  1. The exemption almost never applies: Meeting any one of the three exception conditions negates it, and virtually all organisations meet at least one.

  2. Accountability obligation remains: Art. 5(2) and Art. 24 require the controller to demonstrate compliance regardless of size. A RoPA is the primary demonstration tool.

  3. Supervisory authority expectation: All major SAs recommend maintaining RoPA regardless of size.

  4. Minimal burden for small organisations: A small organisation may have 5-15 processing activities. Documenting these in a simple template requires minimal effort compared to the compliance risk of not maintaining records.

  5. Investigation readiness: If a supervisory authority investigates (e.g., following a data breach or complaint), the first request will be for processing records. Not having them significantly worsens the organisation's position.

  6. Foundation for other obligations: The RoPA feeds into DPIA assessments, privacy notices, data subject rights responses, and breach impact assessments. Without it, these processes are also impaired.

レビュー

まだレビューはありません。使ってみた感想をお寄せください。

同じリポジトリのスキル

概要と使いどころ

Implements 42 CFR Part 2 protections for substance use disorder patient records. Covers written consent requirements stricter than HIPAA, re-disclosure prohibition, court order procedures, qualified service organization agreements, and 2024 amendments aligning Part 2 with HIPAA. Keywords: 42 CFR Part 2, substance use disorder, SUD records, re-disclosure, consent, Part 2 amendments.

日本語の概要は準備中です。原文の説明を表示しています。

mukul975/Privacy-Data-Protection-Skills3032026年3月17日 更新

Guides assessment of third-country adequacy decisions under GDPR Article 45 for international data transfers. Covers the current EC adequacy decisions list, adequacy assessment criteria, partial adequacy handling, and monitoring of adequacy decision reviews. Keywords: adequacy decision, Article 45, third country, adequate protection, EC adequacy list.

日本語の概要は準備中です。原文の説明を表示しています。

mukul975/Privacy-Data-Protection-Skills3032026年3月17日 更新

Implements age-gating mechanisms for online services to restrict access based on user age. Covers hard gates versus soft gates, neutral age prompts, re-verification triggers, circumvention prevention, and regulatory requirements under GDPR, COPPA, UK Online Safety Act, and DSA. Keywords: age gate, age restriction, neutral prompt, children, online services, access control.

日本語の概要は準備中です。原文の説明を表示しています。

mukul975/Privacy-Data-Protection-Skills3032026年3月17日 更新

Evaluates and implements age estimation and verification technologies for online services. Covers facial age estimation, digital ID verification, self-declaration with risk assessment, AI-based age estimation, and the accuracy versus privacy tradeoff. Includes ICO guidance and euCONSENT framework. Keywords: age verification, age estimation, facial analysis, digital ID, children, online safety.

日本語の概要は準備中です。原文の説明を表示しています。

mukul975/Privacy-Data-Protection-Skills3032026年3月17日 更新

Preparing EU AI Act compliance documentation for high-risk AI systems. Covers Annex III classification, technical documentation under Art. 11, conformity assessment, risk management systems, and CE marking requirements. Keywords: EU AI Act, high-risk AI, Annex III, conformity assessment, CE marking.

日本語の概要は準備中です。原文の説明を表示しています。

mukul975/Privacy-Data-Protection-Skills3032026年3月17日 更新

Implements GDPR Art. 22 automated decision-making and AI Act Art. 14 human oversight requirements for AI systems. Covers identification of solely automated decisions, meaningful human intervention design, logic explanation mechanisms, and contestation procedures. Keywords: Art. 22, automated decision, human oversight, AI Act, profiling, contestation.

日本語の概要は準備中です。原文の説明を表示しています。

mukul975/Privacy-Data-Protection-Skills3032026年3月17日 更新

mukul975 のスキルをすべて見る

このスキルの問題を報告する