CRA Reporting from 11 September 2026: 10 Misconceptions

Cyber Resilience Act reporting misconceptions under the EU Cyber Resilience Act Article 14

Article 14 of the EU Cyber Resilience Act (Regulation (EU) 2024/2847) takes effect on 11 September 2026: manufacturers must report actively exploited vulnerabilities and severe incidents to ENISA and their CSIRT. Most write-ups repeat the 24/72-hour summary. These are the Cyber Resilience Act reporting misconceptions we keep meeting in practice — assumptions that feel reasonable but leave a team unprepared, or in breach, the first time the clock starts.

TL;DR

  • Ten common assumptions about CRA reporting that leave teams in breach when Article 14 starts on 11 September 2026.
  • Chief among them: “December 2027 is the real deadline” and “we’re SaaS, so we’re exempt” — both wrong.

Cyber Resilience Act reporting misconceptions, myth 1: “The real deadline is December 2027 — we have time”

The most expensive of them all. 11 December 2027 is when the full product requirements apply (conformity assessment, technical documentation, CE marking). The Article 14 reporting duty starts 15 months earlier, on 11 September 2026 — and the transitional rule carves reporting out: it applies to all in-scope products placed on the market before December 2027, modified or not. A product you shipped in 2023 can be outside the CE-marking rules and still reportable today if a vulnerability in it is being actively exploited. Your installed base is in scope for reporting even where it is out of scope for everything else.

2. “We’re SaaS, so we’re exempt”

Partly true, easy to get wrong. Standalone SaaS generally sits outside the CRA — but as soon as a cloud component is a “remote data processing solution” (the product cannot fulfil its function without it), it falls back into scope. “We’re SaaS” is a claim to verify against your architecture, not a conclusion.

3. “Every vulnerability we find has to be reported”

No. The duty covers only two things: vulnerabilities with reliable evidence of actual exploitation in the wild, and severe incidents affecting your product’s security. A vulnerability you find and patch before anyone exploits it stays in your ordinary vulnerability-handling process and never touches the 24-hour clock — see our CRA vulnerability reporting guide.

Cyber Resilience Act reporting misconceptions, myth 4: “If a library we use gets exploited somewhere, we must report it”

The trigger is exploitation of a vulnerability in your product — not the fact that a component you bundle was exploited in someone else’s product elsewhere. Shared dependencies mean shared vulnerabilities, not shared reporting triggers. This is one of the assumptions a good component inventory quietly removes: knowing where a component is actually deployed, and whether the exploitation touches your own attack surface, matters more than which CVE is trending.

5. “We’re legally required to continuously monitor for exploits”

Article 14 creates no standalone monitoring mandate. The duty is triggered by awareness — reasonable certainty that your product is affected — and most reportable cases start with an inbound: a CVD disclosure, a customer flag, a report in your PSIRT inbox. Strong detection still matters enormously, but continuous monitoring is a requirement to design for full applicability in December 2027, not for September 2026.

6. “We need full SBOMs for every legacy product before September”

Over-investing against the wrong deadline. Formal SBOM and vulnerability-handling obligations (Annex I) apply from December 2027. By September you need enough component visibility to know, when an advisory arrives, whether the affected component is in one of your shipped products — not a fully versioned SBOM for every SKU you ever sold.

Cyber Resilience Act reporting misconceptions, myth 7: “One person handling this is enough”

The Single Reporting Platform enables two named accounts per manufacturer — a primary and a secondary representative — both authenticating via EU Login, and both creatable today, before the platform goes live (see our CRA Single Reporting Platform update). A single named reporter who is asleep, on leave, or has left the company is exactly the failure mode a 24-hour deadline exposes. Setting up both EU Login accounts now costs almost nothing and removes a real risk.

8. “Filing a report makes the vulnerability public”

A common reason teams hesitate. The receiving CSIRT can delay or withhold onward dissemination on cybersecurity grounds, and full public disclosure is typically held back until a fix exists — publicising an unpatched, actively exploited flaw would hand attackers a roadmap. None of this pauses your own 24/72-hour clock; it only affects who sees the content and when. Confidentiality is among the most persistent of these fears, and the fear of exposure is exactly what makes teams file late.

9. “We’ll wait until we’re completely certain”

Waiting for full forensic confirmation is not a valid reason to miss the 24-hour window. The threshold is “reasonable certainty” after an initial assessment, not proof beyond doubt. An investigation still aiming at total certainty does not excuse a late report once that threshold is crossed — see our CRA reporting obligations guide on “becoming aware”.

Cyber Resilience Act reporting misconceptions, myth 10: “We’ll automate submissions through an API”

There isn’t one, so far. Submission happens through a person filling in a web form. You can automate detection, triage and drafting internally, but the actual filing needs a named human, reachable and able to act — including outside office hours and on weekends.

What these Cyber Resilience Act reporting misconceptions add up to

These Cyber Resilience Act reporting misconceptions share one root: treating Article 14 as a document to finalise once, rather than a live operational capability. The teams that will be fine on 11 September are not the ones with the thickest binder — they are the ones who have named their two reporters, created the EU Login accounts, built a rough 24-hour template, and know who picks up the phone when a disclosure lands in the inbox.

● Article 14 is live on 11 September 2026 — a live capability beats a thick binder.

Most of these misconceptions dissolve once reporting is a rehearsed capability rather than a document. COMPLYD ships a ready-to-use workflow for both reporting tracks, built around the 24h/72h/final stages. See the CRA Incident Reporting Software, try our free CRA online check tool, or book a demo.

Frequently asked questions

Are legacy products in scope for CRA reporting?

Yes. The Article 14 reporting duty applies to in-scope products placed on the market before December 2027, even where they are outside the CE-marking and conformity requirements.

Do I have to report every vulnerability?

No — only actively exploited vulnerabilities (evidence of exploitation in the wild) and severe incidents affecting your product. Ordinary discovered-and-patched vulnerabilities are not reportable under Article 14.

Can I automate the submission?

Not currently. You can automate detection, triage and drafting, but the filing itself is a web form completed by a named, reachable human — so name a primary and secondary reporter now.

Related to this article

Check for free

Does the Cyber Resilience Act apply to you? →

Product Check and Readiness Check, no login.

Glossary

Cyber Resilience Act glossary →

30 key terms explained, each with its legal reference.

Where does your product stand on the Cyber Resilience Act?

A first answer in minutes — or see the software that turns the legal text into a workflow.

Free CRA online checkExplore the CRA software


Last updated on 25 September 2026 · Sources: Regulation (EU) 2024/2847 (CRA) — EUR-Lex, ENISA — Single Reporting Platform

Stephan Hutterer

About the author

Stephan Hutterer

Founder & CEO, COMPLYD

Product-security specialist with 20+ years protecting critical OT systems and deep IEC 62443 expertise. He founded COMPLYD to turn the EU Cyber Resilience Act into a workable process for product teams.

LinkedIn profile