2024 · Netherlands · J - Information and communication · Ransomware attack via shared customer-communications supplier (supply-chain)
Analysed 2026-11-06 by Hermes
Open this case in the tool →Ransomware attack via shared customer-communications supplier (supply-chain)
Strategy to Execution
Evidence coverage: 3/6 stages, weight 20/60
Decisive stage: entry (clearest evidence) · Alternative: Expertise to Execution
(1) The strategic expectation at the victims was documented and explicit: RBG and ABN AMRO outsourced customer communications (tax assessments, bank letters) to AddComm under contractual security arrangements that RBG states were also checked, and RBG's spokesperson said AddComm's information-security system met the European standards series; the expectation for a processor holding BSNs and account numbers of an entire municipal and bank population is doctrinal in the strongest sense - contracts plus assurance. (2) In practice the assurance did not hold in the supplier's operational reality: attackers held AddComm's systems from 5 to 17 May 2024 without being noticed, were discovered only when the systems encrypted themselves, had already copied out data of a selected group of customers, and AddComm could not determine what was taken of whom for weeks; the cascade reached 5,407 organisations and over 1.5 million people according to the Dutch Data Protection Authority's 2024 data-leak report, which opened its own investigation. (3) That is a Strategy to Execution gap rather than merely weak security: formal strategic artefacts on the vendor-assurance diagonal (supplier-security contracts, checked arrangements, standards compliance) existed and were relied on, yet were not realised in the day-to-day architecture and operations of the shared processor, where detection ran 12 days late and forensic scope could not be established. The classification rests on the entry/exposure stage (clearest evidence: origin documented in the victims' own statements); AddComm's own detection and egress gaps are the runner-up Expertise to Execution reading but no source states their origin and break, so they stay contributing evidence, not the decisive stage. AddComm's entry vector was never disclosed and is not load-bearing. (4) RM13 (Strategy to Execution lens): supplier-security requirements, contracts and monitoring on the RBG and ABN AMRO side did not translate into controlled third-party operation - the arrangements existed and were checked, yet the processor sat penetrated for twelve days and lost its customers' data (causal role: Enabling). RM04 (Strategy to Execution lens): the operating model concentrated the customer communications of about sixty named organisations - in the end 5,407 chain organisations - on one AddComm environment, so one compromise halted RBG's tax mailings, forced it to break all digital connections and led ABN AMRO to pause the supplier, multiplying the blast radius across the whole chain (causal role: Amplifying). RM34 (Strategy to Execution lens): monitoring and alert-handling requirements were not implemented or operated as intended at AddComm, which noticed the intrusion only when the encryption started, twelve days after access began (causal role: Obscuring). RM33 (Strategy to Execution lens): logging and evidence-preservation requirements were not implemented to a scope that could answer what was taken - which data of which customers had been stolen stayed undetermined for weeks, so downstream victims could only warn everyone and search themselves (causal role: Obscuring).
RM13 — Third-party / supply-chain trust boundaryRM04 — Control dependency concentrationRM34 — Monitoring / detection / alert-handling gapRM33 — Logging / forensic-readiness gapRow covers three victims (AddComm was the organisation actually breached; RBG and ABN AMRO were its clients whose data sat at AddComm and whose systems were confirmed not hacked). Sector set on AddComm, the breached organisation (J); RBG alone would be O Public administration and defence, ABN AMRO K Financial and insurance activities. The row's incident matches its stated victims; no data-quality mismatch found. The two addcomm.nl incident pages in source_url return HTTP 404 today and are not cited; AddComm's current crisis blog is cited instead (opened; a short first-person account confirming the 17 May discovery, negotiations, Bitcoin and the AP contact). The case description's claim that regional media (31mag) reported RBG data already leaking on 25 May could not be verified - the 31mag page would not open; the AP 2024 report instead states the data were read and possibly stolen, and no publication was confirmed, while ABN AMRO saw no indications of misuse. eurepoc_extra.intelligence_impact says no leaking of data while the row description reports a 25 May leak claim - internal dataset inconsistency, left unreconciled. AddComm reportedly paid ransom (AD, NOS; its own blog mentions negotiations and Bitcoin) and disclosed agreements with the criminals not to publish or keep the data; verification that the criminals complied is not documented in the sources reviewed. The AP opened an investigation, visited AddComm and closed the dossier without sanction, so the regulator-finding-of-no-failings rule does not apply (no finding on AddComm's controls was published). No legal adviser/NCSC advisory found naming the threat group; unattributed actor does not lower confidence per the method.