The first question after a breach is always "how bad is it?", and teams that answer it from instinct either over-notify on a minor exposure or under-notify on a breach that later turns out to include passwords and tokens. This template gives incident, privacy and security teams a structured impact assessment to run once a breach is confirmed: which records and people are affected, which identities and credentials were exposed, what harm is likely, and how severe the breach is. It uses the GDPR Article 33 and 34 tests and the ENISA severity methodology for personal data, and adds a credential and identity section that most privacy-only assessments leave out. It is written for incident managers, data protection officers and security leads. Adapt the bracketed text before adoption.
How to use this template
- Replace text in square brackets, such as [Organisation], [DPO] and [Incident Manager], with your own names. "Must" is mandatory; "should" is recommended.
- Complete one assessment per incident. Open it as soon as a breach is confirmed and update it as facts change; each version keeps its date and author.
- Decision points are marked D1 to D4.
- The completed assessment feeds the notification decision and the register. The notification process itself is in the Data Breach Notification Checklist, the record in the Data Breach Register Template, and letters to individuals in the Customer Breach Notification Letter Template. This template does not repeat them.
- A DPIA assesses planned processing before it happens; this template assesses a breach after it happens. See the DPIA Guide for the former.
1. Purpose
1.1 This assessment establishes, with evidence, the scope and severity of a confirmed or suspected breach at [Organisation], so that notification, containment and support decisions rest on recorded facts.
1.2 Under GDPR Article 33(5), the controller must document every personal data breach, its facts, effects and remedial action, in a way that lets the supervisory authority verify compliance. This assessment is the core of that record.
2. Scope
2.1 Applies to any incident affecting the confidentiality, integrity or availability of personal data, customer data, employee data, or credentials and secrets held by or for [Organisation], including breaches at processors and suppliers.
2.2 GDPR Article 4(12) defines a personal data breach as a breach of security leading to accidental or unlawful destruction, loss, alteration, unauthorised disclosure of, or access to, personal data. Availability and integrity breaches count, not only theft.
2.3 Breaches involving only credentials or secrets, with no personal data, still require sections 3, 5 and 7 because exposed credentials lead to further data access.
3. Incident summary
| Field | Entry |
|---|---|
| Incident reference | [IR-2026-xxx] |
| Assessment version, date, author | [v1, date, name] |
| Date and time of awareness | [When [Organisation] had reasonable certainty a breach occurred; starts regulatory clocks] |
| Breach window | [First and last known unauthorised access] |
| Breach type | Confidentiality / integrity / availability (tick all) |
| Cause | [Malicious external, insider, error, supplier, unknown] |
| Systems and data stores involved | [List] |
| Status | [Contained / ongoing]; evidence of exfiltration: [confirmed / suspected / none found] |
| Controller or processor role | [Organisation] acts as [controller / processor] for this data |
3.1 If [Organisation] is a processor, GDPR Article 33(2) requires notifying the controller without undue delay; send this assessment with that notice.
4. Records and people affected
4.1 Record numbers as confirmed, estimated or upper bound, and say how each was derived (log analysis, file listing, database query). Article 33(3)(a) asks for approximate numbers where exact ones are not yet known.
| Data category | Data subjects (type) | Approx. people | Approx. records | Confirmed or estimated |
|---|---|---|---|---|
| Contact details | [Customers] | [ ] | [ ] | [ ] |
| Government ID numbers | [ ] | [ ] | [ ] | [ ] |
| Financial and payment data | [ ] | [ ] | [ ] | [ ] |
| Special category data (health, biometrics and others) | [ ] | [ ] | [ ] | [ ] |
| Children's data | [ ] | [ ] | [ ] | [ ] |
| Account credentials (see section 5) | [ ] | [ ] | [ ] | [ ] |
| Location or behavioural data | [ ] | [ ] | [ ] | [ ] |
4.2 List the jurisdictions where affected people live; this decides which regulators and clocks apply. See the Breach Notification Laws by Country Guide.
5. Identities and credentials exposed
5.1 For each credential type, record what was exposed, whose identity it belongs to, what it can reach and whether it has been rotated. A credential that still works is an open incident, not a historical one.
| Credential type | Form exposed | Belongs to | Access it grants | Rotated or revoked? |
|---|---|---|---|---|
| User passwords | Plaintext / reversible / hashed (algorithm and salting: [ ]) | [Customers, staff] | [Accounts; reuse risk elsewhere] | [Forced reset date] |
| Session tokens and cookies | [ ] | [ ] | [Live sessions, MFA bypass] | [Sessions revoked date] |
| MFA seeds or recovery codes | [ ] | [ ] | [ ] | [Re-enrolment date] |
| API keys and access tokens | [ ] | [Service, integration] | [Systems and data reachable] | [ ] |
| Cloud access keys and service account keys | [ ] | [Workload] | [Accounts, projects] | [ ] |
| Certificates and private keys | [ ] | [ ] | [Signing, TLS, SSH] | [Revoked and reissued] |
| OAuth refresh tokens held for customers | [ ] | [Customers' third-party accounts] | [Customer data in other services] | [ ] |
5.2 D1: downstream exposure. If any exposed credential grants access to further personal data or to customers' systems, the [Incident Manager] must open a linked assessment for that downstream access, or extend this one, before the severity in section 7 is settled. Machine credentials often reach far more data than the breached system held. See the Non-Human Identity (NHI) Guide.
5.3 Customer-held tokens (5.1, last row) may put [Organisation]'s customers in breach too; notify them as part of section 8.
6. Likely harm to individuals
6.1 For each data category, assess the realistic consequences for people, considering malicious intent and what the data enables. Article 33(3)(c) requires describing likely consequences.
| Harm | Typical data that drives it | Likelihood (low / medium / high) | Rationale |
|---|---|---|---|
| Identity theft and fraud | ID numbers, date of birth, address, financial data | [ ] | [ ] |
| Account takeover | Passwords, tokens, MFA material, security answers | [ ] | [ ] |
| Targeted phishing | Contact details with account or transaction context | [ ] | [ ] |
| Financial loss | Payment data, bank details | [ ] | [ ] |
| Discrimination, distress, reputational harm | Special category data, private communications | [ ] | [ ] |
| Physical risk | Home address or location of vulnerable people | [ ] | [ ] |
| Loss of access to services | Availability breaches of essential records | [ ] | [ ] |
6.2 Record mitigating factors: strong encryption with an uncompromised key, confirmed deletion by a trusted recipient, data already public, or credentials rotated before any use.
7. Severity scoring
7.1 [Organisation] uses the ENISA methodology for assessing the severity of personal data breaches (2013), developed with the Greek and German data protection authorities, as a consistent starting point: severity (SE) equals data processing context (DPC) multiplied by ease of identification (EI), plus circumstances of the breach (CB).
| Factor | Scale | Score for this breach |
|---|---|---|
| DPC: data processing context | Basic scores: simple 1, behavioural 2, financial 3, sensitive 4; contextual factors adjust within 1 to 4 | [ ] |
| EI: ease of identification | Negligible 0.25, limited 0.5, significant 0.75, maximum 1 | [ ] |
| CB: circumstances of the breach | Each loss of confidentiality, integrity or availability element adds 0, 0.25 or 0.5; malicious intent adds 0.5 | [ ] |
| SE = DPC x EI + CB | Below 2 low; 2 to below 3 medium; 3 to below 4 high; 4 or above very high | [ ] |
7.2 The score supports judgement; it does not replace it. Credentials that still work, children's data or a vulnerable population can justify a higher rating than the arithmetic gives. Record any override and its reason.
7.3 D2: risk level. The [DPO] sets the GDPR risk level: unlikely to result in a risk, risk, or high risk to the rights and freedoms of individuals.
8. Notification decision
| Risk level | Supervisory authority (Art. 33) | Individuals (Art. 34) |
|---|---|---|
| Unlikely to result in a risk | Not notified; reasoning documented here | Not informed |
| Risk | Without undue delay and, where feasible, within 72 hours of awareness; reasons for any delay stated | Not required; consider voluntary notice |
| High risk | As above | Without undue delay, unless an Article 34(3) condition applies |
8.1 Article 34(3) conditions: the data was made unintelligible to unauthorised persons, for example by encryption; later measures mean the high risk is no longer likely to materialise; or individual communication would involve disproportionate effort, in which case a public communication of equal effect is used. Record the condition relied on.
8.2 Information may be provided to the authority in phases without undue further delay (Article 33(4)). Do not hold the first notice for a complete assessment.
8.3 D3: other notices. Check sector regulators, contractual notice clauses with customers, insurers and law enforcement. In the UK, fraud and cybercrime are reported to Report Fraud (which replaced Action Fraud on 4 December 2025) at reportfraud.police.uk; in Scotland, to Police Scotland on 101.
9. Review and sign-off
9.1 D4: sign-off. The [DPO] and [Incident Manager] sign each version. Re-assess when new evidence changes record counts, credential exposure, exfiltration status or harm, and revisit the notification decision each time.
9.2 Retain the assessment with the register entry for [6] years or per [Organisation]'s retention schedule.
Implementation guidance
- Prepare data maps: the assessment is only as fast as your knowledge of what each system holds. Keep record-of-processing entries linked to systems and to the machine identities that can read them.
- Pre-agree counting methods: decide how to count people versus records and how to de-duplicate before an incident, not during it.
- Pilot with a tabletop: run one past or simulated incident through the template and time each section; see the Identity Attack Tabletop Exercise Scenarios.
- Calibrate scoring: score [five] past incidents with the ENISA method and compare against the decisions made, then adjust contextual factors.
- Track metrics: time from awareness to first assessment version, proportion of assessments revised upward after notification, and credentials still valid [72] hours after awareness.
| Framework | Reference |
|---|---|
| UK GDPR and EU GDPR | Art. 4(12) definition; Art. 33 notification to authority; Art. 34 communication to individuals |
| NIST SP 800-61 Rev. 3 | Incident response as a CSF 2.0 community profile |
| NIST CSF 2.0 | RS.AN-03 analysis of what has taken place and root cause |
| NIST SP 800-53 Rev. 5 | IR-4 Incident Handling; IR-6 Incident Reporting; RA-3 Risk Assessment |
Common failure patterns: counting records but not people; treating hashed passwords as harmless without recording the algorithm; ignoring machine credentials that reach other data; waiting for a complete assessment before the first notice; and never revisiting the assessment when exfiltration is later confirmed.
Standards and references
- EUR-Lex: Regulation (EU) 2016/679, General Data Protection Regulation
- ENISA: Recommendations for a methodology of the assessment of severity of personal data breaches
- NIST: SP 800-61 Rev. 3, Incident Response Recommendations and Considerations for Cybersecurity Risk Management
- NIST: SP 800-53 Rev. 5, Security and Privacy Controls for Information Systems and Organizations
- Report Fraud: Should I report to Report Fraud? (Scotland: Police Scotland on 101)
Related NHI Mgmt Group resources: Data Breach Notification Checklist · Data Breach Register Template · Incident Response Plan Template · DPIA Guide