本文へ移動
cccskills
無料GitHub で公開

employee-monitoring-dpia

Conducts Data Protection Impact Assessments for employee monitoring systems per EDPB Guidelines 3/2019 on workplace data processing. Covers video surveillance, email monitoring, GPS tracking, keystroke logging, and productivity tools. Applies proportionality testing under Art. 35 GDPR. Keywords: DPIA, employee monitoring, surveillance, proportionality, EDPB, workplace privacy, keystroke logging, GPS tracking.

インストール方法を見る

含まれるファイル(5)

  • SKILL.md18.8 KB
  • assets/template.md6.7 KB
  • references/standards.md4.0 KB
  • references/workflows.md7.0 KB
  • scripts/process.py20.4 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。

Employee Monitoring DPIA

Overview

Employee monitoring represents one of the highest-risk processing activities under GDPR because it combines multiple EDPB WP248rev.01 risk factors: systematic monitoring (criterion 3), data concerning vulnerable data subjects — employees are explicitly classified as vulnerable due to the inherent power imbalance in the employment relationship (criterion 7), and often innovative technology (criterion 8). The European Data Protection Board's Guidelines 3/2019 on processing of personal data through video devices and the Article 29 Working Party's Opinion 2/2017 on data processing at work establish that any employee monitoring system requires a DPIA under Art. 35(1) GDPR before deployment.

This skill provides a structured DPIA methodology tailored specifically to employee monitoring scenarios, incorporating the proportionality framework from Barbulescu v Romania (Grand Chamber, ECHR, Application No. 61496/08, 5 September 2017) and national supervisory authority guidance from the CNIL, ICO, and German Federal Commissioner for Data Protection.

DPIA Trigger Analysis for Employee Monitoring

Why Employee Monitoring Always Requires a DPIA

Employee monitoring meets at least three of the nine EDPB WP248rev.01 criteria:

CriterionApplicability to Employee Monitoring
Criterion 3: Systematic monitoringAll forms of employee monitoring constitute systematic observation of individuals in the workplace
Criterion 7: Vulnerable data subjectsEmployees are explicitly listed as vulnerable data subjects by the EDPB due to the power imbalance inherent in the employment relationship
Criterion 5: Large-scale processingEnterprise monitoring systems typically process data about all employees continuously
Criterion 8: Innovative technologyAI-powered productivity tools, keystroke dynamics, screen capture, and behavioural analytics involve novel technologies
Criterion 1: Evaluation or scoringMonitoring data used for performance evaluation constitutes scoring of individuals

Meeting two or more criteria triggers a presumptive DPIA requirement. Employee monitoring typically meets three to five, making a DPIA mandatory in virtually all cases.

National Supervisory Authority Confirmation

  • CNIL (France): Deliberation No. 2018-327 explicitly lists "systematic employee monitoring (productivity, keystroke, screen capture)" as requiring a DPIA.
  • ICO (UK): Employment Practices Data Protection Code, Part 3: Monitoring at Work confirms DPIA requirement for all workplace monitoring systems.
  • BfDI (Germany): Section 26 BDSG (Federal Data Protection Act) imposes additional requirements for employee data processing, reinforcing the DPIA obligation.
  • AEPD (Spain): Guide on Data Protection in Labour Relations (2021) mandates DPIA for all systematic employee monitoring.

Monitoring Categories and Risk Profiles

Category 1: Video Surveillance (CCTV)

Description: Fixed or mobile camera systems recording visual images of employees in the workplace.

Applicable EDPB Guidance: Guidelines 3/2019 on processing of personal data through video devices, Section 8 (Processing in the employment context).

Legal Basis Analysis:

  • Art. 6(1)(f) legitimate interest is the most common lawful basis for workplace CCTV
  • Art. 6(1)(c) legal obligation where national law mandates security monitoring (e.g., financial institutions)
  • Art. 6(1)(a) consent is not valid due to the employment power imbalance (WP29 Opinion 2/2017, Section 5.1)

Proportionality Requirements:

  • Cameras must not be positioned in break rooms, toilets, changing areas, or trade union meeting rooms
  • Recording should be limited to security-relevant areas: entrances, server rooms, cash handling areas, warehouse loading docks
  • Continuous recording must be justified; motion-activated recording is less intrusive
  • Retention period must not exceed 72 hours unless a specific security incident requires longer retention for investigation — CNIL recommends a maximum of 30 days
  • Audio recording alongside video is prohibited in most jurisdictions absent specific statutory authorisation

Atlas Manufacturing Group Example: Atlas Manufacturing Group operates 14 CCTV cameras across its production facility. The DPIA identified that 3 cameras positioned in the staff canteen were disproportionate to the stated security purpose. The DPIA recommended removal of canteen cameras and retention reduction from 30 days to 72 hours for all remaining cameras, with exception procedures for documented security incidents.

Category 2: Email and Internet Monitoring

Description: Systems that log, scan, or analyse employee email communications and internet browsing activity.

Legal Basis Analysis:

  • Art. 6(1)(f) legitimate interest for preventing data leakage, detecting security threats, and enforcing acceptable use policies
  • Must be balanced against Art. 8 ECHR right to respect for private life and correspondence

Proportionality Requirements per Barbulescu v Romania (Grand Chamber): The ECHR Grand Chamber established six criteria that must be satisfied:

  1. Has the employee been notified in advance of the possibility that their communications may be monitored?
  2. What is the extent of the monitoring and the degree of intrusion into the employee's privacy?
  3. Does the employer have legitimate reasons to justify monitoring?
  4. Would monitoring with less intrusive methods be possible?
  5. What are the consequences of the monitoring for the employee?
  6. Has the employee been provided with adequate safeguards?

Risk Assessment Specific Factors:

  • Content monitoring (reading actual emails) is significantly more intrusive than metadata monitoring (sender/recipient/time)
  • Keyword scanning represents an intermediate level of intrusion
  • Personal email accounts must never be monitored even if accessed from corporate devices
  • Privileged communications (with legal counsel, trade union representatives, medical professionals) must be excluded

Category 3: GPS and Location Tracking

Description: Vehicle tracking systems, mobile device location monitoring, and geofencing for field employees.

Legal Basis Analysis:

  • Art. 6(1)(f) legitimate interest for fleet management, lone worker safety, and route optimisation
  • Art. 6(1)(b) performance of employment contract where location tracking is essential to the role (delivery drivers, field engineers)

Proportionality Requirements:

  • Tracking must be limited to working hours; tracking outside working hours is disproportionate unless the vehicle is used exclusively for business
  • Real-time tracking is more intrusive than periodic location logging and requires stronger justification
  • Geofencing alerts are less intrusive than continuous tracking and should be preferred where sufficient
  • Employees must be able to disable tracking when using a company vehicle for personal purposes (if permitted)
  • Speed monitoring and driving behaviour analysis must be justified by specific safety concerns

Atlas Manufacturing Group Example: Atlas Manufacturing Group implemented GPS tracking on 23 delivery vehicles. The DPIA identified that real-time tracking was active 24/7 despite drivers being permitted limited personal vehicle use. The DPIA required implementation of a "personal use" toggle on the vehicle dashboard that suspends tracking and notifies fleet management that the vehicle is in personal use mode.

Category 4: Keystroke Logging and Screen Capture

Description: Software that records individual keystrokes, takes periodic or triggered screenshots, and monitors application usage.

Risk Classification: Very High — this is the most intrusive form of employee monitoring.

Legal Basis Analysis:

  • Art. 6(1)(f) legitimate interest is difficult to establish because the severity of the intrusion typically outweighs any legitimate interest
  • National DPAs have consistently found keystroke logging to be disproportionate for general productivity monitoring
  • CNIL Decision No. MED-2019-006: Keystroke logging is justified only when investigating a specific, documented security incident and must be time-limited

Proportionality Requirements:

  • Keystroke logging as a default monitoring tool is almost certainly disproportionate and will fail the DPIA
  • Screen capture at fixed intervals (e.g., every 5 minutes) captures private information and personal passwords
  • Application-level monitoring (which applications are open and for how long) is less intrusive than screen capture
  • Any deployment must be time-limited, targeted to specific roles or individuals, and supported by a documented security justification

Supervisory Authority Enforcement:

  • CNIL (France): Fined an employer EUR 20,000 for deploying keystroke logging software on employee workstations without conducting a DPIA and without adequate transparency (Decision SAN-2021-003).
  • Garante (Italy): Prohibited a company from using continuous screen capture software, ruling it disproportionate to the stated productivity monitoring purpose (Provvedimento No. 9834141, 2022).

Category 5: Productivity and Behavioural Analytics

Description: Software platforms that aggregate data from multiple sources (email, calendar, application usage, badge swipes, collaboration tools) to generate productivity scores and behavioural profiles.

Risk Classification: Very High — combines evaluation/scoring with systematic monitoring and innovative technology.

Legal Basis Analysis:

  • Art. 6(1)(f) legitimate interest requires a rigorous balancing test
  • If productivity scores are used for employment decisions (promotion, dismissal, pay), Art. 22 automated decision-making restrictions apply
  • Profiling under Art. 4(4) is triggered when monitoring data is used to evaluate work performance aspects

Proportionality Requirements:

  • Aggregate/team-level analytics are less intrusive than individual-level scoring
  • Employees must be informed of exactly which data sources contribute to productivity metrics
  • Employees must have the right to access and challenge their productivity profiles under Art. 15 and Art. 22(3)
  • Human review must be guaranteed before any adverse employment decision based on monitoring data

DPIA Methodology for Employee Monitoring

Phase 1: Preliminary Assessment (Week 1)

  1. Identify the monitoring system: Document the specific technology, vendor, deployment scope, and data flows.
  2. Classify the monitoring category: Map to one or more of the five categories above.
  3. Confirm DPIA obligation: Count WP248rev.01 criteria met (employee monitoring will always meet at least two).
  4. Assemble the DPIA team: Processing owner (typically HR Director), DPO, IT Security Officer, Legal Counsel, and employee representative or works council member per Art. 35(9).
  5. Notify the works council: In jurisdictions with works council requirements (Germany, France, Netherlands, Austria), the works council has co-determination rights over monitoring systems under national labour law.

Phase 2: Systematic Description (Week 2)

Document the following for each monitoring system:

ElementDetails Required
Data categoriesPrecisely what data is collected (e.g., email metadata, email content, URLs visited, keystrokes, screenshots, GPS coordinates, video images)
Data subjectsAll employees, specific departments, specific roles, contractors, visitors
VolumeNumber of employees monitored, frequency of data collection, daily data volume
RetentionHow long monitoring data is stored, deletion procedures, archive policies
AccessWho can access raw monitoring data, who can access reports/dashboards, role-based access controls
RecipientsInternal recipients (HR, line managers, IT security), external recipients (monitoring software vendor, cloud hosting provider)
TransfersWhether monitoring data is transferred outside the EEA (common with US-based SaaS monitoring tools)
Legal basisSpecific Art. 6(1) basis and, where applicable, Art. 9(2) condition

Phase 3: Necessity and Proportionality Assessment (Week 3)

Apply the following proportionality test for each monitoring measure:

Step 1 — Legitimate aim: What specific, documented objective does the monitoring serve? (Security, fraud prevention, regulatory compliance, productivity management, health and safety)

Step 2 — Necessity: Is monitoring necessary to achieve the objective, or can the objective be achieved through less intrusive means?

Monitoring MeasureLess Intrusive Alternative
Continuous video surveillanceMotion-activated recording, access control logs
Email content scanningMetadata analysis, data loss prevention rules on attachments only
Keystroke loggingApplication usage logging, output-based performance measurement
Real-time GPS trackingRoute completion verification, periodic check-ins
Continuous screen captureActive window logging, time-tracking software with manual entries
Behavioural analytics scoringRegular supervisor reviews, objective output metrics

Step 3 — Proportionality: Even if necessary, is the monitoring proportionate to the aim? Apply the Barbulescu six-factor test.

Step 4 — Safeguards: What measures mitigate the impact on employees? (Transparency, access rights, retention limits, data minimisation, grievance procedures)

Phase 4: Risk Assessment (Week 3-4)

Assess risks specific to employee monitoring:

RiskLikelihoodSeverityInherent Risk
Chilling effect on legitimate workplace communication and trade union activityLikelySignificantHigh
Disproportionate surveillance creating hostile work environmentPossibleSignificantHigh
Monitoring data used for discriminatory employment decisionsPossibleMaximumHigh
Unauthorised access to monitoring data by line managersLikelyLimitedMedium
Function creep: monitoring data used for purposes beyond original justificationLikelySignificantHigh
Cross-border transfer of monitoring data to non-adequate jurisdictionsPossibleSignificantHigh
Employee inability to exercise DSAR rights over monitoring dataPossibleSignificantHigh
Capture of privileged communications (legal, medical, union)PossibleMaximumVery High

Phase 5: Mitigation Measures (Week 4-5)

For each identified risk, document specific technical and organisational measures:

Technical Measures:

  • Automated exclusion of privileged communications from monitoring scope (whitelisting legal counsel, medical providers, and trade union email addresses)
  • Role-based access controls restricting monitoring dashboard access to authorised HR personnel only
  • Automated retention enforcement with irreversible deletion at expiry
  • Data minimisation: collect metadata before content; aggregate before individual-level data
  • Encryption of monitoring data at rest (AES-256) and in transit (TLS 1.3)
  • Audit logging of all access to monitoring data with tamper-proof logs

Organisational Measures:

  • Acceptable Use Policy updated to reflect monitoring scope and provided to all employees before monitoring begins
  • Privacy notice under Art. 13/14 specifically addressing monitoring, delivered before system activation
  • Employee grievance procedure for challenging monitoring decisions
  • Annual proportionality review to confirm monitoring remains necessary
  • Training for managers on permissible use of monitoring data
  • Restriction on using monitoring data as sole basis for adverse employment decisions

Phase 6: DPO Review and Works Council Consultation (Week 5-6)

  1. Present completed DPIA to the DPO for independent advice per Art. 35(2).
  2. Document DPO advice and whether it was accepted; if not, record the justification.
  3. Submit DPIA findings to the works council where co-determination rights apply.
  4. Conduct employee consultation per Art. 35(9) — this may be satisfied through works council engagement.
  5. Obtain senior management sign-off acknowledging residual risks.
  6. If residual risk remains High or Very High after mitigation, initiate Art. 36 prior consultation with the supervisory authority.

Ongoing Review Requirements

Employee monitoring DPIAs must be reviewed:

  • Before any new monitoring technology is deployed
  • When the monitoring scope is expanded (new departments, new data categories)
  • When monitoring software is upgraded or vendor is changed
  • Following any employee complaint related to monitoring
  • Following any supervisory authority guidance or enforcement action relevant to employee monitoring
  • At minimum annually as part of the privacy programme calendar

Enforcement Precedents

AuthorityCaseFine/OutcomeRelevance
CNIL (France)SAN-2021-003EUR 20,000Keystroke logging deployed without DPIA or transparency
Garante (Italy)Provvedimento 9834141, 2022Processing prohibitedContinuous screen capture ruled disproportionate
AEPD (Spain)PS/00120/2021EUR 60,000GPS tracking of employee vehicles outside working hours
Datainspektionen (Sweden)DI-2020-11370SEK 300,000Facial recognition for employee time tracking without valid consent or DPIA
Hellenic DPA (Greece)Decision 26/2019EUR 150,000Continuous employee CCTV monitoring without proportionality assessment
ICO (UK)ENF/2021/00352Enforcement noticeEmployer required to cease covert monitoring of employee personal devices

Integration Points

  • Employment Consent Limits: This DPIA must confirm that consent has not been relied upon as the lawful basis for monitoring (see employment-consent-limits skill).
  • Workplace Email Privacy: Email monitoring components of this DPIA feed into the workplace-email-privacy skill for detailed Barbulescu compliance.
  • Employee DSAR Response: Monitoring data is within scope of employee DSARs (see employee-dsar-response skill).
  • Remote Work Monitoring: Remote monitoring raises additional proportionality concerns covered in the remote-work-monitoring skill.
  • Art. 36 Prior Consultation: Where residual risk remains high, escalation to the supervisory authority is required.

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