FAQ item index

Search every question across sub-FAQs

Find the exact question, open the source answer card, and copy a direct link to the anchored sub-FAQ response.

Indexed coverage
29of29items
Across 7 modules • Updated May 9, 2026
Author
Sorena AI
Published
May 9, 2026
Updated
May 9, 2026
GDPR processor vs controller: role boundaries and evidence

What evidence should teams keep for the role decision?

Keep evidence at processing-activity level. A useful role file shows the purpose, essential means, party responsibilities, instructions, contract or arrangement, relevant records of processing, and the trigger for reassessing the label.

Article 30 records support this work because controllers must record processing activities under their responsibility, while processors must record categories of processing carried out on behalf of each controller. RoPA entries should therefore preserve the controller, processor, joint-controller, and subprocessor distinctions instead of collapsing every party into a vendor list.

  • Role assessment showing who decides the purpose and essential means for the specific processing activity.
  • Article 28 contract or legal act, documented instructions, subprocessor approvals, assistance logs, deletion or return record, and audit evidence for processor relationships.
  • Article 26 arrangement, responsibility allocation, contact point if designated, and published essence evidence for joint-controller relationships.
  • Controller RoPA entry with purposes, data-subject categories, personal-data categories, recipients, transfers, retention where possible, and Article 32 measure description where possible.
  • Processor RoPA entry with each controller on whose behalf the processor acts, processing categories for each controller, transfers where applicable, and security-measure description where possible.
Citations
When does the EU GDPR require a DPIA?

Short answer: when is a DPIA mandatory under EU GDPR Article 35?

A DPIA is mandatory when the planned processing is likely to result in a high risk to individuals. Article 35 says the controller must assess the impact before the processing starts, especially where new technologies are used and the nature, scope, context, and purposes of the operation make the risk high.

Article 35(3) gives three cases where a DPIA is required in particular: systematic and extensive automated evaluation, including profiling, that produces legal or similarly significant effects; large-scale processing of special-category data or criminal-offence data; and large-scale systematic monitoring of a publicly accessible area.

  • Treat the controller as accountable for the DPIA threshold decision, even where a processor or vendor provides inputs.
  • Run the threshold check before launch and again when the risk represented by the processing changes.
  • Use one DPIA for a set of similar processing operations only where they present similar high risks.
  • If the assessment shows residual high risk without measures to mitigate it, escalate to prior consultation under Article 36.
Citations
When does the EU GDPR require a DPIA?

How should the high-risk threshold be checked?

Start with the three Article 35(3) cases, then test the broader high-risk criteria described in DPIA guidance: evaluation or scoring, automated decisions with legal or similar significant effects, systematic monitoring, sensitive or criminal-offence data, large scale, matched or combined datasets, vulnerable people, innovative technology, cross-border transfers outside the EU, and processing that prevents people from exercising a right or using a service or contract.

The DPC guidance reports the WP29 rule of thumb: the more criteria a processing operation meets, the more likely it is to require a DPIA; operations meeting at least two criteria will require a DPIA, and a controller that decides otherwise should thoroughly document why the processing is not likely to be high risk.

  • Describe the processing operation, not just the product name or vendor system.
  • Record which Article 35(3) case or high-risk criteria are present, absent, or uncertain.
  • For large-scale processing, document the number or proportion of people affected, data volume and variety, duration or permanence, and geographic extent.
  • Check whether data subjects include children, employees, patients, asylum seekers, elderly people, or another group with a power imbalance or special vulnerability.
  • Escalate borderline cases where multiple criteria are present, the technology is novel, or people cannot reasonably avoid the processing.
Citations
When does the EU GDPR require a DPIA?

How should supervisory-authority list references be used?

Article 35(4) requires each supervisory authority to establish and publish a list of processing operations that are subject to the DPIA requirement, and Article 35(5) allows a supervisory authority to publish a list of operations for which no DPIA is required. Use those lists as a jurisdiction-specific check only where the relevant list is actually available and applicable to the processing operation.

Do not treat a general guidance page, sector label, vendor statement, or non-applicable national list as an exemption. If a no-DPIA list is used, the decision should explain why the processing falls strictly within the listed procedure and continues to meet the relevant requirements.

  • Identify the competent supervisory authority before relying on a DPIA-required or no-DPIA list.
  • Record the list entry, the processing facts that match it, and any conditions or limits stated in the list.
  • Where processing involves several Member States or behavioural monitoring across Member States, flag that Article 35 list issues may involve GDPR consistency-mechanism considerations.
  • If the folder does not ground a specific national list entry, leave the page at Article 35 list mechanics rather than naming that entry.
Citations
When does the EU GDPR require a DPIA?

What must the DPIA contain once the threshold is met?

If the threshold is met, Article 35(7) requires the DPIA to contain at least four elements: a systematic description of the envisaged processing and purposes, an assessment of necessity and proportionality, an assessment of risks to data subjects' rights and freedoms, and the measures envisaged to address those risks and demonstrate GDPR compliance.

CNIL's PIA methodology maps those elements into practical records: context and processing description; fundamental-principles analysis covering purpose, lawfulness, minimization, storage, information, rights, processors, and transfers; privacy-risk analysis; and formal validation involving DPO advice and, where appropriate, views of data subjects or their representatives.

  • Keep the nature, scope, context, purposes, personal data, recipients, retention periods, functional description, and supporting assets in the DPIA file.
  • Document necessity and proportionality against the purpose, lawful basis, data minimization, storage limits, transparency, data-subject rights, processor controls, and transfer safeguards.
  • Assess risk from the perspective of affected individuals, including severity, likelihood, risk sources, and impacts such as illegitimate access, unwanted change, or disappearance of data.
  • Record safeguards, security measures, corrective actions, residual risk, DPO advice where a DPO is designated, and views of data subjects or representatives where appropriate.
Citations
When does the GDPR 72-hour breach notification clock start?

When does the GDPR 72-hour breach notification clock start?

Do not start the Article 33 clock from every raw security alert. The EDPB says a controller becomes aware when it has a reasonable degree of certainty that a security incident has occurred and has led to personal data being compromised.

That short investigation period is not a pause button. The controller is expected to investigate promptly, establish whether personal data was breached, contain the incident, assess risk to individuals, and notify the supervisory authority if Article 33 is triggered.

  • Record the first alert, who received it, and why it was or was not immediately enough to establish a personal data breach.
  • Record the awareness timestamp separately: the point when the controller had reasonable certainty that personal data was compromised.
  • Assess whether the breach is unlikely to result in a risk to rights and freedoms; if not, prepare supervisory-authority notification without undue delay and, where feasible, within 72 hours after awareness.
  • Keep the Article 34 high-risk assessment separate from the Article 33 authority notification threshold; communication to data subjects is triggered by likely high risk.
Citations
Regulation (EU) 2016/679 (GDPR), Article 33

Article 33 sets the controller's supervisory-authority notification duty, the 72-hour timing after awareness, the risk exception, delayed-notification reasons, processor escalation, phased information, and breach documentation duty.

When does the GDPR 72-hour breach notification clock start?

What should happen when a processor finds the breach first?

A processor that becomes aware of a personal data breach affecting personal data processed for a controller must notify the controller without undue delay. The processor does not decide the Article 33 risk threshold for the controller before escalating.

Once the processor informs the controller, the controller uses that notice to run the Article 33 assessment and, if required, notify the supervisory authority. The controller keeps legal responsibility for notification even if a processor is authorised to submit a notice on its behalf.

  • Processor record: when the processor became aware, when it notified the controller, affected services, affected personal data, and facts still unknown.
  • Controller intake record: when the controller received processor notice, whether that notice gave reasonable certainty of a personal data breach, and who opened the Article 33 assessment.
  • Contract check: whether the controller-processor arrangement specifies early breach notice, phased updates, and any authority-notification support.
  • Multi-controller incident check: whether the processor must report details to each affected controller.
Citations
When does the GDPR 72-hour breach notification clock start?

What if the controller cannot complete everything within 72 hours?

The GDPR allows notification information to be provided in phases when it is not possible to provide all information at the same time. The first notice should say what is known, what is not yet known, and that more information will follow without undue further delay.

If the supervisory-authority notification is not made within 72 hours after awareness, Article 33 requires reasons for the delay. Those reasons should be specific to the breach, such as why facts could not be established sooner or why a bundled notification was used for closely related breaches.

  • Minimum notification content: nature of the breach, affected data-subject and record categories where possible, DPO or contact point, likely consequences, and measures taken or proposed.
  • Phased-update log: missing facts, owner for each investigation item, next update trigger, and later information sent to the supervisory authority.
  • Delay record: why notification exceeded 72 hours, when each material fact became available, and why the delay was not excessive.
  • Bundling check: only group similar breaches over a short period when that gives a meaningful notification; different data or breach types should be assessed separately.
Citations
When does the GDPR 72-hour breach notification clock start?

Which records prove the breach-clock decision?

Article 33(5) requires the controller to document personal data breaches, including the facts, effects, and remedial action. The record must allow a supervisory authority to verify compliance with Article 33.

Keep records for both notifiable and non-notifiable breaches. If the controller decides not to notify, the record should explain why the breach was unlikely to result in risk to individuals; if data subjects are not told, the record should support the Article 34 high-risk decision or applicable Article 34 condition.

  • Timeline: first alert, investigation start, awareness point, risk assessment, notification decision, authority submission, phased updates, and closure.
  • Facts: breach type, cause, affected systems, affected personal data, affected data subjects, known or approximate volumes, and whether data remained intelligible.
  • Effects and risk: likely consequences, likelihood and severity assessment, high-risk data-subject communication decision, and reviewer approvals.
  • Remedial action: containment, recovery, mitigation steps, communications sent, processor updates, authority correspondence, and lessons learned.
Citations
Page 2 of 2