Article 10 Significant incidents with regard to managed service providers and managed security service providers

This Regulation is to further specify the cases in which an incident should be considered to be significant for the purpose of Article 23(3) of Directive (EU) 2022/2555. The criteria should be such that relevant entities are able to assess whether an incident is significant, in order to notify the incident in accordance with Directive (EU) 2022/2555. Furthermore, the criteria set out in this Regulation should be considered exhaustive, without prejudice to Article 5 of Directive (EU) 2022/2555. This regulation specifies the cases in which an incident should be considered to be significant by setting out horizontal as well as entity-type specific cases.

Pursuant to Article 23(4) of Directive (EU) 2022/2555, relevant entities should be required to notify significant incidents within the deadlines set by that provision. Those notification deadlines are running from the moment the entity becomes aware of such significant incidents. The relevant entity is therefore required to report incidents that, based on its initial assessment, could cause severe operational disruption of the services or financial loss for that entity or affect other natural or legal persons by causing considerable material or non-material damage. Therefore, when a relevant entity has detected a suspicious event, or after a potential incident has been brought to its attention by a third party, such as an individual, a customer, an entity, an authority, a media organisation, or another source, the relevant entity should assess in a timely manner the suspicious event to determine whether it constitutes an incident and, if so, determine its nature and severity. The relevant entity is therefore to be regarded as having become ‘aware’ of the significant incident when, after such initial assessment, that entity has a reasonable degree of certainty that a significant incident has occurred.

With a view to establishing whether an incident is significant, where relevant, relevant entities should count the number of users impacted by the incident, taking into consideration business and end customers with whom the relevant entities have a contractual relationship as well as natural and legal persons that are associated with business customers. Where a relevant entity is unable to calculate the number of impacted users, the relevant entity’s estimate of the possible maximum number of affected users should be considered for the purpose of calculating the total number of users affected by the incident. The significance of an incident involving a trust service should not only be determined by the number of users but also by the number of relying parties as these can be equally affected by a significant incident involving a trust service in regard to operational disruption and material or non-material damage. Therefore, trust service providers should, where applicable, also take into account the number of relying parties when establishing whether an incident is significant. For that purpose, relying parties should be understood as natural or legal persons that rely upon a trust service.

Maintenance operations resulting in the limited availability or unavailability of the services should not be considered as significant incidents if the limited availability or unavailability of the service occurs according to a scheduled maintenance operation. Moreover, where a service is unavailable due to scheduled interruptions such as interruptions or non-availability based on pre-determined contractual agreement should not be considered as significant incident.

The duration of an incident which impacts availability of a service should be measured from the disruption of the proper provision of such service until the time of recovery. Where a relevant entity is unable to determine the moment when the disruption began, the duration of the incident should be measured from the moment the incident was detected, or from the moment when the incident was recorded in network or system logs or other data sources, whichever is earlier.

Complete unavailability of a service should be measured from the moment the service is fully unavailable to users, to the moment when regular activities or operations have been restored to the level of service that was provided prior to the incident. Where a relevant entity is unable to determine when the complete unavailability of a service began, the unavailability should be measured from the moment it was detected by that entity.

For the purpose of determining the direct financial losses resulting from an incident, relevant entities should take into account all the financial losses which they have incurred as a result of the incident, such as costs for replacement or relocation of software, hardware or infrastructure, staff costs, including costs associated with replacement or relocation of staff, recruitment of extra staff, remuneration of overtime and recovery of lost or impaired skills, fees due to non-compliance with contractual obligations, costs for redress and compensation to customers, losses due to forgone revenues, costs associated with internal and external communication, advisory costs, including costs associated with legal counselling, forensic services and remediation services, and other costs associated to the incident. However, administrative fines, as well as costs that are necessary for the day-to-day operation of the business, should not be considered as financial losses resulting from an incident, including costs for general maintenance of infrastructure, equipment, hardware and software, keeping skills of staff up to date, internal or external costs to enhance the business after the incident, including upgrades, improvements and risk assessment initiatives, and insurance premiums. The relevant entities should calculate the amounts of financial losses based on available data and, where the actual amounts of financial losses cannot be determined, the entities should estimate those amounts.

Relevant entities should also be obliged to report incidents that have caused or are capable of causing the death of natural persons or considerable damage to natural persons’ health as such incidents are particularly serious cases of causing considerable material or non-material damage. For instance, an incident affecting a relevant entity could cause unavailability of healthcare or emergency services, or the loss of confidentiality or integrity of data with an effect on the health of natural persons. For the purpose of determining whether an incident has caused or is capable of causing considerable damage to a natural person’s health, relevant entities should take into account whether the incident caused or is capable of causing severe injuries and ill-health. For that purpose, the relevant entities should not be required to collect additional information to which they do not have access.

Limited availability should be considered to occur in particular when a service provided by a relevant entity is considerably slower than average response time, or where not all functionalities of a service are available. Where possible, objective criteria based on the average response times of services provided by the relevant entities should be used to assess delays in response time. A functionality of a service may be, for instance, a chat functionality or an image search functionality.

Successful, suspectedly malicious and unauthorised access to a relevant entity’s network and information systems should be regarded as a significant incident, where such access is capable of causing severe operational disruption. For instance, where a cyber threat actor pre-positions itself in a relevant entity’s network and information systems with a view to causing disruption of services in the future, the incident should be considered to be significant.