A suspected personal data breach creates two separate GDPR compliance questions: must the incident be reported to a supervisory authority, and must affected individuals be informed? This decision tree explains how to assess risk, manage the 72-hour reporting window, coordinate with processors, document the decision, and keep the workflow ready for the next incident.
Overview
Under GDPR, a personal data breach is broadly an incident involving the accidental or unlawful destruction, loss, alteration, unauthorised disclosure of, or access to personal data. A breach can therefore involve more than a confirmed data theft. Misdelivery, exposed cloud storage, ransomware, lost devices, compromised credentials, and accidental deletion may all require review.
The notification decision depends primarily on the likely risk to the rights and freedoms of individuals. The organisation should not wait for every technical detail before starting its GDPR breach risk assessment. Begin with the facts available, record assumptions, and update the assessment as the investigation develops.
There are three possible outcomes:
- No notification to the supervisory authority: the organisation documents why the breach is unlikely to result in a risk to individuals.
- Notification to the supervisory authority: the breach is likely to result in a risk, but not necessarily a high risk, to individuals.
- Notification to both the supervisory authority and affected individuals: the breach is likely to result in a high risk to individuals, unless an applicable exception or effective remedial action changes that assessment.
This is operational guidance, not a substitute for legal advice or the instructions of the relevant supervisory authority. National procedures, sector rules, and cross-border facts should be checked during the incident.
What to track
The GDPR breach notification decision tree
- Is personal data involved? Confirm whether the affected information relates to identified or identifiable individuals. If no personal data is involved, use the security incident process rather than the GDPR breach notification process, while recording the basis for that conclusion.
- Has a breach occurred? Identify whether personal data was lost, destroyed, altered, disclosed, or accessed without authorisation. Treat credible indicators as an incident requiring triage; certainty can develop during containment and investigation.
- When did the organisation become aware? Record the date and time when the organisation had a reasonable degree of certainty that a security incident had occurred involving personal data. This timestamp starts the notification clock for the controller assessment.
- Could the breach create a risk to individuals? Assess the nature, sensitivity, volume, and context of the data, as well as the likelihood and severity of potential harm. If risk is unlikely, document the reasoning and continue monitoring.
- Is the risk high? Consider whether individuals could face serious effects such as identity misuse, financial harm, discrimination, loss of confidentiality, physical danger, or significant reputational or emotional harm. If high risk is likely, plan communication to affected individuals without undue delay.
- Has a processor discovered the incident? The processor should escalate the incident to the controller without undue delay under the relevant processing arrangement. The controller remains responsible for deciding whether notification is required, so processor communications should include discovery time, affected systems, data categories, containment actions, and known or suspected individuals affected.
- Is the incident cross-border? Determine the organisation's main establishment, the locations of affected individuals, and the relevant processing activities. Confirm the appropriate supervisory authority or authorities rather than relying on an informal assumption.
Risk assessment table
| Factor | Questions to record | Why it matters |
|---|---|---|
| Data involved | What categories of personal data were exposed, changed, lost, or deleted? | Credentials, financial information, health information, precise location, identification data, and confidential records may increase potential harm. |
| Scale | How many individuals and records may be affected? | Volume influences the possible reach and severity of the incident, even when exact numbers are not yet known. |
| People affected | Are children, employees, customers, vulnerable people, or other groups involved? | The context and vulnerability of individuals can change the risk assessment. |
| Exposure | Who could access the data, for how long, and was it downloaded or redistributed? | Actual access, onward disclosure, and exposure duration help distinguish theoretical from credible risk. |
| Controls and response | Was the data encrypted, pseudonymised, revoked, recovered, or otherwise made unintelligible? | Effective safeguards and prompt remediation may reduce risk, but should be evidenced rather than assumed. |
| Potential harm | What could happen to individuals if the data is misused or remains unavailable? | Likelihood and severity together determine whether the risk is low, material, or high. |
Notification content checklist
Where notification to a supervisory authority is required, prepare the information available within the applicable deadline. The notice should explain the nature of the breach, the categories and approximate number of individuals and records concerned, likely consequences, and measures taken or proposed to address the breach and reduce adverse effects. Include a contact point for follow-up. If some information is still unavailable, provide a clear initial notification and supplement it as the investigation produces reliable facts.
Communications to individuals should use clear, plain language. Explain what happened, what information may be involved, the likely consequences, the steps already taken, and practical actions individuals can take. Provide a reliable contact route and avoid speculation presented as fact.
Cadence and checkpoints
The first 72 hours
The 72-hour GDPR rule applies to notifying the supervisory authority after the controller becomes aware of a qualifying breach. It is not a target for beginning the investigation; triage should start immediately. If notification occurs after 72 hours, record the reasons for the delay and include them with the notification.
- First hours: appoint an incident owner, preserve evidence, contain the exposure, establish the discovery timestamp, and open a decision log.
- Same day: identify the controller and processor roles, classify affected data, estimate the scope, check contractual escalation requirements, and involve privacy, security, legal, communications, and business owners as needed.
- Before the deadline: make and approve the notification decision, identify the competent supervisory authority, prepare an initial notice if required, and decide whether communication to individuals is necessary.
- After notification: continue root-cause analysis, refine affected-person counts, coordinate follow-up responses, and retain evidence supporting every decision.
Use a quarterly checkpoint to test the workflow with a tabletop exercise. Confirm that the on-call route works, processor contacts are current, the breach register is accessible, and the team can retrieve cloud logs and identity records. A monthly operational check is appropriate for high-change environments: review open privacy incidents, unresolved risk decisions, vendor escalations, and log-retention gaps.
For supporting controls, align the process with an incident response plan for compliance and a documented evidence retention policy. These documents should make ownership and evidence requirements explicit rather than leaving them to an incident-time discussion.
How to interpret changes
A risk assessment is not necessarily finished when the first notification decision is made. New evidence may change the scope, likelihood, or severity of harm. For example, a suspected exposure may become more serious when logs show that data was downloaded, or less serious when access records demonstrate that an encrypted dataset could not be read.
Reassess when any of these variables changes:
- the number or categories of affected individuals;
- the sensitivity of the data or the identity of the people involved;
- whether unauthorised access, copying, or onward disclosure occurred;
- whether credentials, keys, sessions, or other access paths remain active;
- whether containment, deletion, recovery, or encryption materially reduces likely harm;
- whether a processor, subprocessor, or separate controller has additional facts;
- whether the incident affects more than one country or legal entity.
Do not treat “no notification” as “no incident.” GDPR compliance generally requires controllers to document personal data breaches, including the facts, effects, and remedial action, even when notification is not required. Keep the decision proportional but sufficiently clear that an independent reviewer can understand what was known, when it was known, who approved the conclusion, and what evidence supported it.
For cloud incidents, record the shared-responsibility boundary, relevant provider notifications, access logs, configuration changes, and retention limitations. A cloud security policy template can help define the controls that make this evidence available.
When to revisit
Review this decision tree before every annual incident-response exercise and whenever official GDPR guidance, supervisory authority procedures, national requirements, or internal processing activities change. Also revisit it after a real incident, a material cloud migration, a new high-risk processing activity, a major processor change, or a contract renewal that alters breach-escalation terms. A practical compliance calendar can help assign these recurring reviews; see the cybersecurity compliance calendar template.
At the next checkpoint, ask five practical questions: Can the team identify the discovery time? Can it locate the current processor escalation contacts? Can it produce a defensible risk assessment? Can an authorised owner approve notification within the required timeframe? Can the organisation explain and evidence a decision not to notify? If any answer is no, assign an owner and deadline, update the incident response materials, and test the revised process. This turns GDPR data breach notification from a one-time document into a repeatable privacy compliance capability.