The wall of shame is NIS2's sharpest enforcement tool, and it favours companies that report incidents first
Thu, 19th Nov 2026 (Today)
NIS2 sets a 24-hour deadline for organisations to send an early warning after becoming aware of a significant incident - one that causes or is capable of causing severe operational disruption or financial loss, or considerable material or non-material damage to other people or organisations.
Meeting that deadline depends on the tooling already running when the incident happens. Detection and monitoring have to pick up activity early enough to investigate it, while a centralised, tamper-resistant record of access and privileged activity shows which accounts touched which systems, when, and under whose authority.
That record has to come from one place and be exportable to the team's monitoring tools; if credentials and secrets live in spreadsheets, shared drives and individual tools, nobody can assemble it inside 24 hours.
The reporting process also has to be ready, with an incident response workflow, a notification template and the correct national reporting portal where one is available.
Before any of that can happen, though, somebody has to recognise that the incident meets the reporting threshold and have enough evidence to act on it. That recognition is where most of the difficulty starts.
Key insights
- NIS2 Article 23 turns incident disclosure from a liability question into a reporting duty: the early warning is due within 24 hours of becoming aware of a significant incident, and making it doesn't add to the organisation's legal exposure.
- The early warning asks for a first account, with full forensics due in the final report a month later, so the 24-hour window is a question of knowing what you know, since the detailed analysis can follow.
- A report sent on time earns a response: the receiving CSIRT or authority has to answer, usually within 24 hours, and the organisation gains access to help and threat intelligence it wouldn't get otherwise.
- The national authority has to account for its own sector to the rest of Europe, which gives it a reason to help early reporters and penalise silence.
- Contributors to a 2026 NIS2 access management report by Passwork argue that honesty about incidents, more than any technical control, decides whether an organisation can meet the reporting obligation.
- The UK's Cyber Security and Resilience Bill, now in the House of Lords, brings the same 24-hour initial notification to UK data centres, MSPs and operators of essential services, so the preparation described here applies on both sides of the Channel.
- The organisation that reports first decides what gets said about the incident; the one that waits leaves that to an authority with the power to publish infringements.
Reporting means admitting there is an incident
Organisations can be cautious about disclosing an incident while they're still trying to understand it and what its legal consequences might be. Security investigates, legal gets involved, and information stays within a small group.
NIS2 starts the reporting clock while much of that work is still going on, and the people who receive those reports can see the hesitation it causes.
Dr. Yiannis Pavlosoglou, Deputy Governor at the Greek National Cybersecurity Authority, receives NIS2 incident reports from the other side of that process. He talked about his experience in a 2026 NIS2 access management report from European password and secrets manager Passwork. Asked what holds organisations back, he points first to honesty. "I would say the biggest barrier is understanding that you have to be honest with having cyber security incidents."
He also talks about a "wall of shame". In his account, the phrase applies to national authorities as much as to the organisations they supervise, since each authority has to answer for its sector in front of the rest of Europe. The old model, where a critical entity lost payroll for a week and nobody treated it as a big deal, no longer holds.
The directive makes the same demand of organisations. Article 23 expects the early warning while the investigation is still underway, so the organisation may not yet know the cause or full impact. It also states that notifying an incident doesn't, by itself, increase the notifying entity's liability.
The report doesn't disappear into a filing system, but reaches a national CSIRT or competent authority that answers with feedback and, on request, operational advice. A report that never arrives, on the other hand, can itself amount to an infringement.
The UK is on the same clock
UK organisations reading this as a European problem should check the calendar. The Cyber Security and Resilience (Network and Information Systems) Bill passed the House of Commons on 16 June 2026 and is now before the Lords.
It extends the UK's NIS Regulations 2018 to data centres, managed service providers and large load controllers, and it adopts the same two-stage structure as Article 23: an initial notification to the regulator and the NCSC within 24 hours of becoming aware of a significant incident, with a fuller report to follow.
Many UK firms already answer to NIS2 anyway, either through an EU subsidiary or because an EU customer in scope has written the directive's supply-chain duties into the contract.
What the 24-hour notification needs behind it
At 2am, the 24-hour deadline isn't much help if nobody can establish whether the incident is significant. Someone has to apply the Article 23 test to what's known so far, and for some providers the criteria are more specific.
Commission Implementing Regulation (EU) 2024/2690 sets more specific significance criteria for the digital infrastructure and digital service providers it covers. Those thresholds don't apply across every NIS2 sector, so most organisations work from the directive's own definition.
Once an incident meets the threshold, the team needs enough information for the early warning and a known route to the relevant CSIRT or authority, both of which depend on records that already exist.
In practice, the national cybersecurity authorities and independent practitioners interviewed for that same NIS2 access management report find that the record a 24-hour early warning depends on is the same one an audit later asks for: a central log of who accessed which credential or system, when, and with what authority, held separately from the systems under investigation.
One of the report's contributors, Ioannis Ntokos, IT Governance Risk & Compliance Expert who has worked for FedEx, PwC, and Randstad, describes incident handling as the hardest NIS2 requirement to put into operation because of the number of moving parts. "You need to have reporting channels. You need to know who to reach out to."
The tooling has to give the team something useful to work with. Monitoring can establish when suspicious activity began, and access and privileged-activity records can help identify the accounts involved. Those records, along with the decisions made during the investigation, need to remain available as the response develops.
The same central store that produces the record also has to support the response. If a compromised credential appears in the early-warning evidence, the team needs to revoke or rotate it, and everything that shares it, within the same 24 hours, with a log entry showing who did so and when. A tool that can only report on access, and not change it, leaves the containment step to whoever has admin rights on each system, which is the arrangement the incident has just shown to be unsafe.
Article 23(1) can also require communication with customers. Where appropriate, an entity must notify recipients of its services without undue delay when a significant incident is likely to adversely affect the provision of those services.
The 72-hour incident notification follows the early warning, and it includes an initial assessment of severity and impact and, where available, indicators of compromise. The authority can request an intermediate report while the response is still under way, and the final report is generally due within one month of the incident notification.
Reporting can bring help back
The information sent under NIS2 can feed back into the organisation's own response. Article 23(5) requires the CSIRT or competent authority to respond to an early warning without undue delay and, where possible, within 24 hours. The entity receives initial feedback, and if it asks for further help, the CSIRT or authority can provide guidance or operational advice on mitigation. A CSIRT can also provide additional technical support.
Pavlosoglou has described the authority turning up to help organisations after an incident. Reporting puts the incident in front of people who may already have information about similar activity elsewhere, and that exchange runs in both directions.
He also describes NIS2 as the first time threat intelligence became something organisations are allowed to do and can learn from. A peer who sees you under attack can call to ask whether they're affected, and the answer comes from the security team, with no team of lawyers in between.
None of this means the report goes everywhere at once. The entity reports the incident to its national CSIRT or competent authority, and the information travels further only if the incident affects more than one country. Under Article 23(6), where appropriate, the CSIRT, competent authority, or single point of contact must inform other affected member states and ENISA without undue delay. In doing so, they must protect the reporting organisation's security and commercial interests and keep the information confidential.
National teams also share information through the CSIRTs network under Article 15, while anonymised and aggregated notification data goes to ENISA under Article 23(9). Entities can also share information voluntarily under Article 29 or notify incidents, cyber threats, and near misses under Article 30.
Each report adds to a picture that no single organisation can build on its own.
An incident is a bad time to find out the process doesn't work
That picture only helps if the process behind it works, and many organisations don't test it until they have to. Gabriel Dinu, Deputy Director at Romania's National Cyber Security Directorate, has seen how an incident changes priorities. Discussing what pushes organisations to improve their cybersecurity, he said: "if they have an incident, they will do it the next day."
NIS2 incident reporting doesn't leave much room for that approach. Take a plausible significant incident and see how long it takes to decide whether it needs reporting. Then try to retrieve the evidence that would support the early warning.
Pavlosoglou spoke of authorities arriving after an incident and asking for logs, only to discover that nobody had ever enabled logging. Missing logs slow the investigation and make it difficult to establish what happened when the regulator starts asking questions.
Customers may have questions as well, if their services or data were affected. The records the team collects during the response become part of what the organisation can show about how it handled the incident.
Public disclosure backs the reporting regime
NIS2 also gives national authorities powers to make information about infringements public. Article 32 and Article 33 set out supervisory and enforcement powers for essential and important entities, respectively, and those powers include requiring entities to make aspects of infringements public
Failure to meet the reporting obligations can itself amount to an infringement, which brings incident reporting within NIS2's enforcement framework.
Article 34 provides for administrative fines. Member states must ensure that breaches of Articles 21 or 23 by essential entities can attract maximum fines of at least €10 million or 2% of worldwide annual turnover, whichever is higher. For important entities, the corresponding figures are at least €7 million or 1.4%.
Two limits apply, though. Suffering a significant incident doesn't automatically amount to an infringement, and NIS2 creates no EU-wide public list of entities that have suffered incidents.
Whatever the authority decides to do with it, the 24-hour window closes quickly, but the record an organisation builds to meet it keeps working. It feeds the 72-hour notification and the final report, it answers the customers who ask what happened to their data, and it gives the auditor who turns up months later something to read.
So build it now: put every credential and every privileged action into one log that nobody involved in an incident can edit, confirm the team can pull it in under an hour, and run that drill before the directive, or the UK bill that follows it, runs it for you.