TL;DR:
- A wave of regulation is underway: NIS-2, DORA, and the Cyber Resilience Act (CRA) are drastically increasing legal requirements for IT security, risk management, and supply chain integrity.
- “Paper-based compliance” falls short: Relying solely on checklists and static PDF documents offers no protection against actual cyberattacks, nor can it keep pace with agile development processes.
- Engineering as the solution: Modern information security services integrate security directly into software development and cloud infrastructure—for instance, through dynamic ISMS processes, automated DevSecOps pipelines, and proven incident response frameworks.
- The added value: Companies not only meet their regulatory obligations but also sustainably strengthen their cyber resilience, securing a genuine competitive advantage.
This article is a (partly) automated translation of the German original.
When terms like NIS-2 (Network and Information Security), DORA (Digital Operational Resilience Act), or the CRA (Cyber Resilience Act) come up in meetings, many IT leaders envision one thing above all else: a massive increase in bureaucratic overhead. This is entirely understandable. New EU regulations are often associated with just one thing: more forms, lengthy approval processes, and less time for day-to-day operations.
The problem rarely lies with the regulation itself, but rather with how we approach it in practice. We tend to organize security into folders and checklists. Yet, a completed Excel spreadsheet will not stop an attack when it actually happens, nor will it make systems any more resilient.
So, how do we shift from mere “paper compliance” to genuine technical resilience?
Why regulatory pressure is on the rise
The EU is increasingly replacing previous guidance with mandatory requirements. This currently affects almost every level of IT. Here is a brief summary of the key classification criteria and regulations:
NIS-2: This applies to companies with 50 or more employees or an annual turnover of at least €10 million within critical sectors, classifying them as either “important” or “essential” entities. This encompasses 18 sectors, as defined in Annexes 1 and 2 of the BSI Act (the Act on the Federal Office for Information Security and the security of information technology in various entities). Furthermore, regardless of company size or annual turnover, an organization may fall under the scope of NIS legislation if it operates within one of these sectors (e.g., as a DNS provider, a trust service provider, or an operator of critical infrastructure). The specific details regarding classification into NIS-2 sectors are highly complex, and the assessment process can be extremely time-consuming. Therefore, the classification process should be carried out by experienced consultants specializing in critical infrastructure (KRITIS). A key aspect involves the strict reporting deadlines for significant incidents: an initial “early warning” must be submitted to the BSI (Federal Office for Information Security) within 24 hours, followed by a subsequent report within 72 hours of an incident’s discovery. A final report must be submitted within one month of the initial notification.
DORA: This regulation applies not only to banks and insurance companies but also—indirectly, via contractual agreements—to their IT service providers (e.g., cloud providers, data centers, and software vendors). A core component is “Threat-Led Penetration Testing” (TLPT)—that is, conducting realistic, simulated attacks on IT infrastructure during live operations.
CRA: This requires manufacturers of products with digital elements to provide a Software Bill of Materials (SBOM) as part of their technical documentation. It also mandates the provision of free security updates for the product’s expected lifespan (with a minimum of five years). An exception applies if the anticipated service life is shorter. Furthermore, there are additional obligations, such as vulnerability management and risk management.
Even those not directly subject to these laws may still feel their impact through the supply chain; regulated clients often simply pass these requirements on to their suppliers, software partners, and service providers.
The Pitfall: Treating Security Merely as a Documentation Project
In many organizations, the immediate reaction to new regulatory requirements is to acquire templates, fill out PDFs, and draft process descriptions—all for the sake of an audit.
This approach has two major drawbacks:
- It offers no actual protection: Documents do nothing to address configuration errors, weak passwords, or vulnerabilities in the code.
- A brake on agile teams: When approval forms are required for every minor change, development speed suffers.
Effective security is not achieved through retrospective documentation, but rather through robust IT architecture, automated processes, and security integrated directly into the development lifecycle.
A practical approach: Integrating security directly into the technology
To ensure that security teams and developers aren’t working at cross-purposes, security must be woven into daily operations as seamlessly as possible. This is best accomplished through automation and clearly defined responsibilities.
1. Keeping the ISMS Alive
In practice, an ISMS rarely fails because of its set of rules; rather, it falters when it comes to day-to-day application. Instead of spending months drafting policies upfront, the following steps have proven effective:
- The “Minimum Viable ISMS” as a starting point: Do not aim to perfectly cover all 93 controls of ISO 27001 on day one. Focus first on the 20% of risks that cause 80% of the potential damage—typically identity management, missing patches, and insecure backups.
- Keep policies concise and readable: No one reads a policy that is longer than two pages. Formulate clear rules of conduct rather than legalistic, defensive text. For example: “MFA is mandatory for all external access” is better than three paragraphs explaining the theoretical significance of authentication.
- Embed processes where work already happens: If your team uses Jira or GitLab, the risk register shouldn’t be a separate Excel spreadsheet on SharePoint. Instead, create risks as specific ticket types within the same system used to prioritize other tasks. This allows management to see—during resource planning—exactly which risk is being mitigated by a given technical ticket.
- Automated evidence instead of manual screenshots: Auditors require proof. From the outset, set up automated scripts that query the state of your systems—either on a monthly basis or in preparation for an audit (e.g., “Are all hard drives encrypted?” or “Do all admins have MFA enabled?”). This saves days of searching for documentation when an audit approaches.
2. Automating Technical Checks (DevSecOps)
To ensure that automation remains more than just a buzzword, agile teams rely on established open-source and enterprise tools:
- Static Application Security Testing (SAST): Tools such as SonarQube, Semgrep, or Checkmarx examine source code during the pull-request process for common vulnerabilities (e.g., the OWASP Top 10) and block the merge if critical issues are detected.
- Software Composition Analysis (SCA): Tools like Snyk, Trivy, OWASP Dependency-Check, or https://dependencytrack.org/ analyze the libraries in use and automatically generate an SBOM—as required by the Cyber Resilience Act (CRA)—in either the SPDX or CycloneDX format.
- Infrastructure as Code (IaC) Scanning: Tools such as tfsec (now part of Trivy) or Checkov scan Terraform or Kubernetes configuration files prior to deployment, preventing misconfigurations—for instance, ensuring that cloud storage (such as S3 buckets) is not inadvertently left publicly accessible.
3. Realistic Incident Response Testing
NIS-2 and DORA mandate very short reporting deadlines in the event of an incident. Searching for an emergency response plan only when a crisis actually occurs wastes valuable time. A more effective approach involves conducting short, regular exercises—known as “Game Days”—to test monitoring tools and reporting chains under realistic conditions.
What does a pragmatic “Game Day” look like in practice? Rather than disrupting live operations, the security team plans a two-hour scenario within a staging environment (e.g., a simulated data breach or the failure of a central database):
- Injection: A technical ticket or a simulated alert is triggered—without prior warning—for the on-call team.
- Response: The team must follow the predefined reporting chain, analyze log files, and isolate the breach.
- Retrospective: Afterward, the focus is not on assigning blame, but on evaluation—were the log files sufficiently informative? Did everyone know who was responsible for preparing the notification to the authorities?
Three such exercises per year provide more genuine security than any hundred-page emergency manual. Of course, processes, workflows, and documentation are necessary—but ideally, these should be the natural outcome of established, day-to-day practices.
A simple real-world example: from abstract risk to solution
What does this pragmatic approach look like in practice? Let’s take the issue of third-party risk (a key focus of DORA and NIS-2):
- The old way: The security team sends a 50-page questionnaire to every new software vendor. The vendor fills it out in a rigid, perfunctory manner, and someone simply files the document away. In reality, nothing has been effectively assessed.
- The pragmatic approach: Define simple criteria (e.g., does the provider process personal data? Do they have access to the production network? Are there other types of critical access?). If any of these apply, a developer or “security champion” checks just two specific things: Is there a current SOC 2 or ISO certificate, and is there a clean API interface for managing permissions? The answer is then recorded directly in the system alongside the relevant tool.
This reduces the effort required from several days to just two hours—while producing a result that is audit-ready and genuinely valuable.
Liability is a leadership issue: Why senior management must be on board
An often-overlooked aspect of NIS-2 and DORA is the personal responsibility of senior management. Executives can no longer simply “delegate” responsibility for cyber risks to the IT department.
In practical terms, this means the following for company leadership:
- Regular security briefings: A standing agenda item in monthly management meetings (e.g., a 15-minute status update on cyber risks and critical vulnerabilities).
- Mandatory documentation: Verifiable training for senior management on risk management and cyber threats, ensuring compliance with legal due diligence obligations.
Conclusion
There is no denying that these new regulations require effort. However, if we view these requirements not merely as red tape, but as an opportunity to modernize legacy processes and automate testing, everyone wins:
IT becomes more stable, workflows become clearer, and the company meets regulatory requirements as a natural part of its ongoing operations. Those getting started now should begin with a thorough impact and gap analysis to establish priorities. It is precisely at this intersection of regulation and modern platforms that we help teams integrate compliance seamlessly into their existing cloud and DevOps environments. to integrate.