Join our Newsletter — 33% off our NHI Course
Home› Guides› DORA Major ICT Incident Reporting Guide: Classification Criteria,…
Guide Regulations & Frameworks

DORA Major ICT Incident Reporting Guide: Classification Criteria, Thresholds and Report Deadlines

PDF for newsletter subscribers · free to read online

← All guides
By NHI Mgmt Group, edited by Lalit Choda Updated 27 October 2026 16 min read
On this page

Under DORA a financial entity can have four hours to tell its supervisor about a major ICT-related incident, and the decision on whether an incident is "major" rests on a classification test that many incident teams have never run under pressure. Regulation (EU) 2022/2554 sets the duty in Articles 17 to 19, and three Level 2 acts set the detail: Delegated Regulation (EU) 2024/1772 on classification and materiality thresholds, Delegated Regulation (EU) 2025/301 on report content and time limits, and Implementing Regulation (EU) 2025/302 on the reporting template. This guide works through the classification test criterion by criterion, the three report stages and their clocks, how DORA reporting differs from NIS2 and GDPR, outsourced and aggregated reporting, and what identity and machine credential evidence a report needs. It is for incident managers, ICT risk, compliance and identity teams in EU financial entities and the ICT providers that report for them. For the wider regulation see the DORA Guide.

Status checked 27 October 2026: DORA has applied since 17 January 2025. Delegated Regulation 2024/1772 entered into force on 15 July 2024; Delegated Regulation 2025/301 and Implementing Regulation 2025/302 were published in the Official Journal on 20 February 2025. The Commission's Digital Omnibus proposal of 19 November 2025 would add a single reporting interface for incident notifications across NIS2, GDPR and DORA; it is a proposal, not law, and does not change the DORA deadlines below.

Key takeaways

  • Major means critical services plus a threshold: under Article 8 of RTS 2024/1772 an incident is major if it affects critical services and either involves successful, malicious and unauthorised access that may result in data losses, or meets two or more of the other materiality thresholds.
  • The clock starts at classification, with a backstop: the initial notification is due within four hours of classifying the incident as major and no later than 24 hours after becoming aware of it (RTS 2025/301 Article 5(1)(a)).
  • Three reports, one template: an intermediate report follows within 72 hours of the initial notification even if nothing has changed, and a final report within one month of the latest intermediate report, all on the Annex I template of ITS 2025/302.
  • Unauthorised access is a fast track: a credential compromise that gives an attacker malicious access to systems supporting critical services can make an incident major on its own, which puts identity telemetry at the centre of classification.
  • Outsourcing the report does not outsource the duty: Article 19(5) lets a third party submit reports, but the financial entity stays fully responsible and must tell its competent authority about the arrangement before the first report.

Scope: who and what is covered

The duty applies to every financial entity listed in Article 2(1) of DORA, from credit institutions and payment institutions to insurers, investment firms and crypto-asset service providers. Microenterprises and the entities in Article 16(1) on the simplified framework still classify and report major incidents, but the recurring incident aggregation rule in Article 8(2) of RTS 2024/1772 does not apply to them.

What is reportedLegal basisMandatory?Recipient
Major ICT-related incidentsDORA Art. 19(1); RTS 2024/1772; RTS 2025/301YesThe competent authority under Article 46, or a single authority designated by the Member State; significant credit institutions report through the national authority to the ECB
Major operational or security payment-related incidents (credit, payment, account information and e-money institutions)DORA Art. 1(1) and Art. 23YesAs above; these replace the former PSD2 payment incident reports
Significant cyber threatsDORA Art. 19(2); RTS 2024/1772 Art. 10; RTS 2025/301 Art. 6No, voluntaryCompetent authority, which may share it with other authorities
Information to clientsDORA Art. 19(3)Yes, where clients' financial interests are affectedAffected clients, without undue delay, with the mitigation measures taken

Member States may also require entities to send the initial notification and reports to their national CSIRTs. Competent authorities pass details on to the European Supervisory Authorities, the ECB where relevant, NIS2 single points of contact or CSIRTs, and resolution authorities (Article 19(6)).

Key obligations

Step 1: the classification criteria (RTS 2024/1772 Articles 1 to 7)

CriterionWhat you measureMateriality threshold (Art. 9)
Clients, financial counterparts and transactions (Art. 1)Clients unable to use the service or adversely affected, counterparts with a contractual arrangement, transactions with any part carried out in the EUOver 10% of clients using the service; over 100,000 clients; over 30% of counterparts; over 10% of the daily average number or value of transactions; or relevant clients or counterparts affected
Reputational impact (Art. 2)Media coverage, repeated complaints, inability to meet regulatory requirements, likely loss of clients or counterpartsAny one of the four conditions
Duration and service downtime (Art. 3)Duration from occurrence (or first log record) to resolution; downtime until the pre-incident service level is restoredDuration over 24 hours, or downtime over 2 hours for ICT services supporting critical or important functions
Geographical spread (Art. 4)Impact on clients, branches, group entities or market infrastructures in other Member StatesTwo or more Member States
Data losses (Art. 5)Impact on availability, authenticity, integrity or confidentiality of data(a) impact adversely affects business objectives or regulatory compliance; or (b) successful, malicious and unauthorised access that may result in data losses
Economic impact (Art. 7)Direct and indirect costs and losses before recoveries: stolen funds, replacement hardware and software, staff costs, redress, forgone revenue, legal and forensic feesExceeds, or is likely to exceed, EUR 100,000
Criticality of services affected (Art. 6)ICT services supporting critical or important functions; authorised or supervised financial services; successful, malicious and unauthorised accessGateway condition: must be met for any major incident
  • Estimates are expected: where actual client, transaction or cost figures are unknown, entities estimate them from comparable reference periods. Do not wait for exact numbers.
  • Excluded costs: routine maintenance, post-incident business enhancements such as upgrades and risk assessments, and insurance premiums do not count towards the EUR 100,000 threshold.
  • Recurring incidents (Art. 8(2)): incidents that are not major alone count as one major incident if they occurred at least twice within six months, share the same apparent root cause and together meet the criteria. Run this check monthly and report in aggregated form (ITS 2025/302 Article 3).

Step 2: the decision rule (RTS 2024/1772 Article 8)

  1. Is the criticality of services criterion met? If not, the incident is not major.
  2. Is the Article 9(5)(b) threshold met, meaning successful, malicious and unauthorised access that may result in data losses? If yes, the incident is major.
  3. Otherwise, are two or more of the thresholds in Article 9(1) to (6) met? If yes, the incident is major.

Step 3: the three reports and their content (RTS 2025/301 Articles 1 to 4)

ReportDeadline (Art. 5)Core content
Initial notificationAs early as possible, within 4 hours of classification as major and no later than 24 hours after awareness; if classified later, within 4 hours of that classificationEntity reference code; date and time of detection and classification; description; the criteria that made it major; Member States affected; how it was discovered; origin where available; whether a business continuity plan was activated
Intermediate reportAt the latest 72 hours after the initial notification, even if status is unchanged; updated without undue delay and once regular activities are recoveredTime of occurrence and recovery; how each criterion was met; incident type; threats and techniques used; affected functions, processes and infrastructure; impact on clients' financial interests; reports to other authorities; temporary actions; indicators of compromise
Final reportNo later than one month after the latest intermediate reportRoot causes; when the incident was resolved and root causes addressed; how it was resolved; information for resolution authorities; direct and indirect costs, losses and recoveries; recurring incident information
  • Template: Annex I of ITS 2025/302 has a general section (fields 1.1 to 1.15), the initial notification (2.1 to 2.10), the intermediate report (3.1 to 3.35) and the final report (4.1 to 4.16). Annex II is the data glossary and shows which fields are mandatory at each stage.
  • Combining reports: once regular activities have recovered or the root cause analysis is complete, the stages may be combined into two or one submission, provided each deadline is met (ITS Article 2).
  • Channel: use the secure electronic channel your competent authority provides; if it is unavailable, notify by other agreed secure means and resubmit when asked.
  • Weekends and bank holidays: if a deadline falls on one, most entities may submit by noon the next working day. This does not apply to the initial notification or intermediate report of credit institutions, central counterparties, trading venue operators and entities identified as essential or important under NIS2, and authorities may remove the extension for other significant or systemic entities (RTS 2025/301 Article 5(4) to (6)).
  • Missed deadline: tell the authority without undue delay, and no later than the deadline, with the reasons (Article 5(3)).
  • Reclassification: if an incident reported as major turns out never to have met the criteria, say so using the template's report type and other information fields (ITS Article 5).

Outsourced and aggregated reporting

  • Outsourcing (DORA Art. 19(5); ITS Art. 6): notify the competent authority once the arrangement is concluded and no later than the first report, giving the third party's name, contact details and identification code; notify again when it ends.
  • Aggregated reports (ITS Art. 7): a provider reporting for several entities may submit one aggregated report only where the incident originates from that provider, every covered entity has classified it as major, all are in one Member State under the same competent authority, and the authority has explicitly permitted it. Significant credit institutions, trading venue operators and central counterparties must always report individually.

How the DORA clocks differ from NIS2 and GDPR

RegimeFirst reportFollow-upFinalTrigger
DORA4 hours from classification, 24 hours from awareness at the latest72 hours from initial notification1 month from latest intermediate reportMajor ICT-related incident under RTS 2024/1772
NIS2 Art. 23Early warning within 24 hours of awarenessIncident notification within 72 hours of awareness1 month after the incident notificationSignificant incident
GDPR Art. 3372 hours from awareness of a personal data breachInformation in phases where neededNone fixedPersonal data breach unless unlikely to result in a risk

DORA is the sector-specific act for financial entities, so NIS2 incident reporting generally gives way to it, but a GDPR notification to the data protection authority is separate and still required where personal data is breached. See the DORA and NIS2 Overlap Mapping Guide and the Cyber Incident Reporting Requirements Map.

Timeline and current status

  1. 14 December 2022: DORA adopted; published 27 December 2022.
  2. 13 March 2024: Delegated Regulation 2024/1772 adopted; in force 15 July 2024.
  3. 23 October 2024: Delegated Regulation 2025/301 and Implementing Regulation 2025/302 adopted.
  4. 17 January 2025: DORA applies, including major incident reporting.
  5. 20 February 2025: RTS 2025/301 and ITS 2025/302 published in the Official Journal; each enters into force on the twentieth day after publication.
  6. 19 November 2025: Digital Omnibus proposal published, including a single reporting interface; under negotiation. See the EU Digital Omnibus Guide.

Competent authorities publish their own reporting portals and national guidance. Check your authority's instructions and EUR-Lex for amendments before relying on these dates.

Enforcement and penalties

  • No fixed EU fines: Article 50 requires Member States to set administrative penalties and remedial measures that are effective, proportionate and dissuasive; it sets no maximum amounts for financial entities, so exposure depends on national law.
  • Minimum powers (Art. 50(4)): orders to cease conduct, temporary or permanent halts of practices, any measure including financial ones to secure compliance, access to existing telecom traffic records where national law allows, and public notices naming the responsible person and the breach.
  • Personal exposure (Art. 50(5)): Member States must allow measures against members of the management body and other individuals responsible, under national law conditions.
  • Criminal penalties: Member States may also impose criminal penalties (Article 52).
  • Practical exposure: late or missing reports are easy for supervisors to evidence, because portals timestamp every submission. Repeated late initial notifications are a likely early enforcement theme.

What it means for AI agents and non-human identities

  • Machine credential compromise can be the trigger: a stolen API key, service account password or cloud access token used to reach systems supporting critical services is "successful, malicious and unauthorised access". Your classification depends on being able to tell, within hours, whether that access happened. MITRE ATT&CK T1078 Valid Accounts and T1078.004 Cloud Accounts describe the pattern. See the Service Account Security Guide.
  • Intermediate reports ask for techniques and indicators: the 72-hour report wants threats and techniques used and indicators of compromise. Logs that tie each action to a named workload identity, token or agent make this answerable.
  • Agents can change incident duration: an AI agent that keeps retrying, reprocessing or acting on bad data can extend downtime or multiply affected transactions. Include agent kill switches and credential revocation in the containment plan. See the AI Agent Incident Response Playbook.
  • Provider-originated incidents: when an ICT provider's integration credentials are abused, you still classify and report. Contract terms must make the provider tell you fast enough to meet the four-hour clock. See the Third-Party Incident Notification Clauses Guide.
  • Root cause in the final report: if the root cause is an unowned, long-lived machine credential, the remediation section should show ownership, rotation and decommissioning, not only a password reset.

Framework mapping

DORA reporting stepNIST CSF 2.0NIST SP 800-53 Rev 5ISO/IEC 27001:2022 Annex A
Detection and declaring an incidentDE.AE-08IR-5 Incident Monitoring, AU-6A.5.25 Assessment and decision on information security events
Classification against RTS 2024/1772RS.MA-02, RS.MA-03IR-4 Incident HandlingA.5.25
Initial, intermediate and final reportsRS.CO-02, GV.OC-03IR-6 Incident ReportingA.5.24 Incident management planning; A.5.31 Legal and regulatory requirements
Root cause for the final reportRS.AN-03IR-4A.5.27 Learning from information security incidents
Recovery status in updated intermediate reportsRC.CO-03IR-8 Incident Response PlanA.5.26 Response to information security incidents
Evidence for regulatorsRS.AN-03AU-11 Audit Record RetentionA.5.28 Collection of evidence
Provider notification termsGV.SC-05SR-8 Notification AgreementsA.5.20 Supplier agreements

Compliance steps

  1. Map critical services: link each ICT service and system to the critical or important functions it supports, using the same mapping as your register of information. Without it you cannot test the gateway criterion. See the DORA ICT Third-Party Risk and Register of Information Guide.
  2. Pre-compute denominators: keep current figures for clients per service, counterparts, and daily average transaction number and value, so the 10%, 30% and 100,000 thresholds can be tested in minutes.
  3. Build the classification worksheet: one row per criterion, with the threshold, data source, owner and an "estimate" flag. Add it to your Incident Severity Classification Matrix Template.
  4. Record awareness and classification times: the incident ticket must capture both, because the 24-hour backstop runs from awareness and the four-hour clock from classification.
  5. Pre-fill the template: store fields 1.1 to 1.15 (names, LEI codes, contacts, parent undertaking) for each legal entity so the initial notification only needs incident facts.
  6. Agree outsourced and group reporting: decide who submits for each entity, notify the authority before the first report, and check whether aggregated reporting is permitted.
  7. Test it: run a tabletop starting on a Friday evening with a compromised service account, time the classification and the initial notification, and confirm whether the weekend extension applies to you.
  8. Run the monthly recurrence check: review non-major incidents for a shared root cause over the last six months.
  9. Roll back safely: if an incident is reported and later found not to be major, reclassify through the template instead of withdrawing silently.

Detection and measurement

  • Awareness to classification time: median and worst case per quarter; target well inside 20 hours so the 24-hour backstop is never at risk.
  • Classification to initial notification: percentage submitted within four hours.
  • Intermediate and final reports on time: percentage meeting the 72-hour and one-month deadlines.
  • Reclassification rate: share of reported incidents later reclassified as non-major; a high rate suggests weak data at classification.
  • Identity attribution: percentage of incidents where the human or machine identity used for access was known before the intermediate report.

Common failure patterns

  • Starting the clock at the wrong point: waiting for full impact figures before classifying, when estimates are allowed and the 24-hour backstop runs from awareness.
  • Ignoring the unauthorised access route: treating a contained credential compromise as minor because no downtime occurred, although Article 9(5)(b) can make it major on its own.
  • No denominators: nobody can say what 10% of clients of a service is at 2am.
  • Assuming the provider reports: the entity remains responsible, and aggregated reporting needs explicit authority permission.
  • Missing the recurrence check: repeated small outages from one faulty integration that together cross the thresholds go unreported. The July 2024 CrowdStrike outage showed how one provider fault reaches many firms at once; see the Third-Party Concentration Risk Guide.
  • Forgetting GDPR: filing the DORA report and assuming it covers the separate personal data breach notification.

Practitioner checklist

  • Map every ICT service to critical or important functions and keep the mapping current.
  • Maintain client, counterpart and transaction denominators per service.
  • Embed the RTS 2024/1772 decision rule in the incident runbook.
  • Capture awareness time and classification time in every incident ticket.
  • Pre-fill the ITS 2025/302 general fields for each legal entity.
  • Confirm whether the weekend and bank holiday extension applies to you.
  • Notify the authority of any outsourced reporting before the first report.
  • Log authentication by service accounts, API keys and agents well enough to confirm or rule out unauthorised access within hours.
  • Run the six-month recurrence check every month.
  • Exercise the full report sequence at least once a year.

Standards and references

This guide summarises DORA major ICT-related incident reporting under Regulation (EU) 2022/2554 and its Level 2 acts as of 27 October 2026 and is not legal advice. Related NHI Mgmt Group resources: DORA Guide · DORA and NIS2 Overlap Mapping Guide · Incident Response Guide · Financial Services Identity Security Guide

PDF for subscribers

Take this guide with you

Every section, table and checklist in one PDF, with links to the sources and related guides. Subscribe to our weekly NHI & AI Identity Journal and we’ll email you a download link.

Free weekly newsletter

Subscribe to the NHI & AI Identity Journal

The latest on NHI and Agentic AI security – articles, research, breaches, news and events every week.

Bonus 33% off our NHI Course when you subscribe.

NHIMG Editorial Note
Researched and written by the NHI Mgmt Group editorial team using AI tools, under the editorial direction of Lalit Choda. Facts are drawn from the sources listed on this page and checked by our automated quality process. Last updated 27 October 2026. Spotted an error? Use our contact page.
NHI Mgmt Group, the independent authority on Non-Human Identity, IAM and Agentic AI security. nhimg.org

PDF Downloads for Newsletter Subscribers

Get the guide

Enter your details and we’ll email you a download link.

The full guide is free to read online. The PDF is a thank you for subscribing.

Fill in all fields marked * and tick the PDF box to continue.