Supply chain attacks on the rise: a 5 step approach to effectively protect your organisation

18/11/2026

The past weeks, European organisations have been facing an increased amount of cybersecurity incidents. The most recent one that caught the headlines was a large number of websites being used to distribute malware to try to obtain financial gains for the attackers. What catches the eye is that many of these incidents are in fact Supply Chain related incidents. A supply chain attack is a cyberattack that compromises a trusted supplier, service provider, software component, or delivery channel in order to reach downstream customers. Instead of attacking each organisation directly, an adversary targets an upstream element such as a vendor network, software build pipeline, code repository, update mechanism, third-party script, or hardware component and uses that trusted relationship to distribute malicious code, steal data, or gain access to customer systems.

In software supply chain attacks, common techniques include hijacking legitimate updates, undermining code signing, compromising open-source dependencies, and injecting malicious code during development or deployment. The SolarWinds Orion incident is a well-known example: attackers inserted a backdoor into a vendor update, which was then distributed to thousands of customers through a trusted update channel. And this is exactly the key: the affected organization usually has an existing trust relationship with these suppliers, which allows malware or cyber attacks inside the realm of these organisations way more effectively than when trying to penetrate their defenses from the outside world.

List of some recent events:

On the night of 11–12 October 2026, attackers remotely compromised refrigeration controllers at roughly 700 French supermarket and food-sector sites, altering defrost settings so refrigerators and freezers warmed to as much as 50°C; the disruption caused large-scale spoilage of fresh and frozen goods and at least 15 small fires. The affected retail groups included Casino, E.Leclerc, Coopérative U, Metro, Carrefour and Intermarché, with additional impacts reported at Monoprix and Netto; reported store-level impacts included about 80 Carrefour stores, around 30 Monoprix stores, 80 Coopérative U stores, and substantial numbers of Intermarché/Netto locations. The principal vulnerabilities were internet-exposed refrigeration controllers, weak or default administrative credentials, and outdated Danfoss equipment that could no longer receive security updates; attackers used the controllers' defrost function to raise temperatures. Expected direct damages are substantial: early retailer estimates put national losses in the tens of millions of euros, largely from destroyed perishable stock, with individual examples including €270,000 in losses at two Alsatian supermarkets. French authorities later attributed the operation to Iran's Shahid Kaveh cyber unit, although the immediate operational lesson is that remotely accessible OT/IoT refrigeration systems require network isolation, strong authentication, patching or compensating controls, and temperature-integrity monitoring.

In the early hours of 1-2 November 2026, a second, broader wave of cyberattacks targeted commercial refrigeration and freezer systems across more than 100 establishments in at least 11 European countries, including Belgium, Luxembourg, the Netherlands, Italy, Sweden, Estonia, Ireland, Portugal, Poland, Spain and Germany. The attackers remotely altered refrigerator and freezer settings, raised temperatures and forced defrost cycles, leading retailers to withdraw or discard perishable stock; the incident also affected French blood-bank refrigeration, although the French blood service reportedly lost no product. Reported affected organizations included Ahold Delhaize, Carrefour, Colruyt, ICA, Coop and Musgrave, with incidents affecting dozens of stores, although a complete chain-by-chain impact list and confirmed direct-loss total had not been publicly established. The likely vulnerabilities mirror those exploited in October: internet-exposed refrigeration supervisory controllers, weak or default credentials, obsolete or unpatched firmware, and inadequate segregation of operational-technology systems from public networks; newly disclosed Copeland/Dixell XWEB EVO flaws include unauthenticated configuration-file writes and root-level command execution. The direct damages were expected to be lower than the October French campaign because some sites detected and responded quickly, but they still included spoiled food, emergency disposal, operational disruption, and the cost of disconnecting, patching or replacing controllers; no official attribution or confirmed link to the October attacks had been made public.

On 16–17 November 2026, attackers exploited a vulnerability in Optimizely's Experimentation platform to inject malicious JavaScript into scripts used by websites belonging to 178 organizations, exposing visitors to a supply-chain-style web compromise rather than a direct breach of each organization's core systems. The affected sites included high-profile brands and services across retail, media, finance, travel and technology: IKEA, Zoom, the BBC, CNN, Fox News, Salesforce, GitLab, Equifax, HSBC UAE, Air France-KLM, Transavia, Alaska Airlines, Ledger and Peacock, with Belgian-linked sites including ABN AMRO MeesPierson, KLM and Just Russel. The malicious scripts sought to replace copied cryptocurrency wallet addresses with attacker-controlled addresses, intercept payment-card data entered in website forms, steal browser cookies and screenshots, and deliver exploit code to older iOS devices. The core vulnerability was a flaw in a trusted third-party web-personalization and experimentation service, allowing unauthorized JavaScript injection into customer projects; Optimizely said it identified and contained the incident. Expected direct damages are still unquantified: researchers had not confirmed that payment-card data was actually captured, but potential losses include fraudulent cryptocurrency transfers, card fraud, account-session theft, customer-notification and incident-response costs, and reputational harm to the affected brands. The incident underscores that third-party JavaScript, tag managers and marketing platforms must be treated as high-risk supply-chain components, with strict vendor security requirements, content-security-policy controls, script integrity monitoring, and rapid disablement procedures.

As already stated, the one thing all these three incidents have in common is the fact that they are supply chain attacks.

The recent Optimizely incident illustrates a web-based variant: a compromise in a third-party experimentation platform allowed malicious JavaScript to be injected into many customer websites, potentially exposing visitors to payment-card theft, cryptocurrency-address manipulation, and malware.

Effective mitigation requires vendor due diligence, least-privilege access, secure build pipelines, software bills of materials (SBOMs), dependency scanning, code-signing verification, network segmentation, and continuous monitoring of third-party components and update channels.

The NIS2 and CRA link

The good news is that regulators and security experts in the EU foresaw these attack types. That is the reason that in Network Information Security Directive 2 (NIS2) and more recently Cyber Resilience Act (CRA) these attack types were flagged and controls required for European organisations to protect themselves against these type of attacks. For nis 2 the most relevant article is 21.2.d:

"The measures referred to in paragraph 1 shall be based on an all-hazards approach that aims to protect network and information systems and the physical environment of those systems from incidents, and shall include at least the following: supply chain security, including security-related aspects concerning the relationships between each entity and its direct suppliers or service providers"

And for CRA it is for manufacturers Art 13 and Annex 1part 1 and 2: 

"Sets the essential product-security requirements, including risk-based secure-by-design/default development, attack-surface reduction and protection against exploitation. Supply-chain components must be considered as part of the product's security posture." and "Requires manufacturers to identify and document vulnerabilities and components, including an SBOM in a commonly used, machine-readable format covering at least top-level dependencies. This is the central CRA control for software supply-chain visibility."

Now, it is all good and well that these requirements are written down in legislation, but how should you as an organization now approach the supply chain risks in your organization? Hereunder you find a 5 step approach:

  • The first step of course is mapping out all third party dependencies, starting with business critical services. This on its own is already a challenge for many organisations since it requires an agreed scope of what are considered "critical services". In all honesty, some supermarkets would probably not have listed the refrigeration monitoring services as business critical, although they are an essential part of the likely mission of "keeping the stores open" adagium.
  • Step 2 would be to do a thorough analysis of all contractual and service documentation with this supplier to see what is already legally and operationally agreed. It would be good to have standard clauses prepared from within your organization as to how you would ideally have the agreement in place, as a starting point for likely negotiations.
  • Third step (which in fact can be done in parallel with step 2) is to see what is the current set up from an architectural point of view, and what would / should be the desired implementation. For the refrigeration service provider for instance, this would mean no direct open access to the refrigeration control panel, but either a read only access (if remote intervention is not needed) or a jump host architecture with multi factor Authentication.
  • The fourth step would be to start active discussions with the suppliers to discuss the contractual and physical changes that are requested. If the desired end situation is for whatever reason not feasible, do not forget to check with your management that the residual risk is accepted and that the risk register is updated accordingly.
  • Lastly, do not forget to also test the new situation to ensure it is implemented correctly and that all scenarios are covered as envisioned.

Share