/
In our experience, most MDR technical documentation is not rejected because the device is unsafe. It is rejected because the file cannot prove that it is safe. A notified body reviewer never meets your product on the bench; they meet your Annex II documentation, and if that file has gaps, the CE certificate stops there, no matter how good the device actually is. That gap between «our device works» and «our file demonstrates it works» is where we spend most of our time with manufacturers, and it is worth understanding before you submit rather than after the first deficiency letter arrives.
What the notified body is actually reading
The technical documentation is not a folder you assemble to taste. Annexes II and III of Regulation (EU) 2017/745 fix its contents: device description and variants, the information the design and manufacturing rely on, the general safety and performance requirements, the benefit-risk analysis and risk management, the verification and validation results including the clinical evaluation, and the post-market surveillance file that Annex III demands. Under the Article 52 conformity assessment route, the reviewer works down that structure as a checklist. Every item that is missing, vague or unsupported is not a stylistic quibble; it becomes a formal nonconformity, and each nonconformity turns into a question that adds a review cycle. The manufacturers who clear assessment fastest are rarely the ones with the best devices. They are the ones whose file answers the checklist before it is asked.
The GSPR checklist that says «not applicable» too often
The general safety and performance requirements in Annex I are the backbone of the whole submission, and they are where we see files fall apart first. Each requirement has to be addressed one by one: is it applicable, by what method is it met, and what is the evidence, the harmonised standard or the common specification that proves it. The recurring failure is a GSPR checklist peppered with «N/A» and no justification, or a column that cites a standard without showing the test report behind it. A reviewer reads an unjustified «not applicable» as «not considered». The question you should ask of your own file is blunt: for every single requirement, could a stranger find the evidence from the checklist alone, without emailing your R&D team?
A clinical evaluation that asserts instead of demonstrating
Clinical evaluation under Article 61 and Annex XIV is the second great rejection zone, and the MDR raised the bar sharply from the old directives. Equivalence, the route many manufacturers still lean on, now requires technical, biological and clinical equivalence and, in practice, contractual access to the equivalent device’s own data, which a competitor will almost never grant. We have watched files built entirely on an equivalence claim collapse on that single point. The literature route, done properly, means a documented search protocol, appraisal criteria and an honest treatment of the evidence that goes against you, not a bibliography stapled to the back. If your clinical evaluation report asserts a conclusion the underlying data does not carry, the notified body will find the gap; that is precisely what the guidance in the MDCG documents trains them to look for.
Post-market surveillance treated as an afterthought
Annex III is where a surprising number of otherwise solid files come undone, because manufacturers write it last and write it thin. The post-market surveillance plan, the post-market clinical follow-up under Annex XIV Part B, and the reporting cadence set by Article 86 all have to be in the documentation as a live system, not a promise. Reviewers now open the PMS section early precisely because an empty PMCF plan is a fast signal that the rest of the file may be aspirational too. A PMS plan that merely says the company «will monitor complaints» tells the reviewer nothing about how, against what indicators, or with what trigger for action.
Why non-EU manufacturers get caught out
If your dossier was built for the FDA 510(k) or a Chinese NMPA submission, it is not MDR technical documentation, and reusing it wholesale is one of the most expensive assumptions we correct. The structures do not map: the MDR is built on the GSPR, not on special controls or predicate devices, the risk and clinical expectations differ, and a manufacturer outside the Union also needs an EU authorised representative named under Article 11 before any of it reaches a notified body. Reformatting a US file into Annex II order is not enough; the evidence itself often has to be regenerated to answer European requirements. This is the point where a manufacturer entering the EU market benefits most from a review before submission rather than after. You can see how we handle that entry through the EU authorised representative role, and it pairs with what the importer must verify once the device is on the market.
How long does a notified body take to review technical documentation?
Longer than anyone plans for, and every deficiency letter adds a full cycle on top. The review time you cannot control; the completeness of the file you submit, you can. In practice the single most effective way to shorten the calendar is to close the gaps above before the first submission, because a clean file moves and a file full of «not applicable» stalls. That is the review we do at ASC Services: we read your technical documentation against Annexes I, II, III and XIV the way a notified body will, flag what it cannot yet prove, and tell you what to fix before you file. Tell us about your device and where you are in the process through our contact page, or see the rest of our work in advisory services and our news section.