Cooperation with a Notified Body – when it is required, and when Module A is sufficient

The question of whether to work with a Notified Body too often comes up at the wrong stage of a project. Usually not when the product concept and safety logic are being defined, but only when the documentation is being assembled, the EU declaration of conformity is being prepared, or just before CE marking. That is when it becomes clear that the issue is not the procedure itself, but earlier assumptions: product classification, the limits of responsibility, the scope of the modernisation or integration, and the quality of the supporting evidence. That is why the right question is not whether to “engage” a Notified Body, but when its involvement is strictly required and when the manufacturer can remain with Module A and independently defend the entire conformity assessment route.

The most expensive mistake happens before the audit

The most costly mistakes do not arise from contact with a Notified Body itself, but from misjudging the situation at the start of the project. In practice, three separate issues are often conflated: safety design, conformity assessment, and the formal involvement of a third party. They are related, but they are not the same thing. A machine may be technically well designed and still not require the involvement of a Notified Body. Conversely, a product with a complex architecture and elevated technical risk does not automatically fall under such a procedure simply because the team considers it “difficult”. Whether third-party involvement is required depends on the legal classification of the product, its intended use, reasonably foreseeable misuse, and the correct conformity assessment procedure.

This directly affects how the project should be managed. The question of a Notified Body is secondary to the question of what is actually to be placed on the market: a machine, an assembly of machines, partly completed machinery, or another product covered by separate legislation. Only once that basis has been established can you determine which rules apply and which conformity assessment route is permitted. Yet in many organisations, Module A is adopted too early as the default option, before product classification has been completed and before responsibility for the manufacturer, importer, or integrator has been determined. That reverses the proper order of decisions.

In practice, the problem rarely becomes apparent in the first discussion about conformity. Most often, it returns only when the technical concept has already been locked down: the mechanical design is complete, components have been ordered, the manual has been prepared, and marking has been planned. At that point, it turns out that the chosen conformity assessment route was incorrect or incomplete. The result is not only the risk of a dispute with the market surveillance authority. Much more often, it means redesigning guards, changing the control system architecture, extending testing, revising the manual, declarations, and markings, or rebuilding the entire technical justification for the product’s safety. At that stage, the cost does not come from the procedure itself, but from having to revisit decisions that should have been made earlier.

To avoid this, four issues must be settled before the technical concept is frozen:

• the product classification and its intended use,
• the applicable legislation,
• the correct conformity assessment procedure,
• whether the involvement of a Notified Body is mandatory or whether Module A remains sufficient.

Only this sequence makes it possible to assess whether self-managed conformity assessment is organisationally and legally sound. In the Polish and EU context, this is not about caution for its own sake, but about ensuring that the conformity decision is grounded in the project, the documentation, and the actual conditions of use. Module A is not the starting point. It is the result of correct product classification and properly structured technical documentation.

Where Module A really ends

The boundary of Module A lies where the manufacturer is no longer able to independently and convincingly demonstrate product conformity under internal production control. It is not a “lighter” procedure and it does not reduce responsibility. On the contrary: because no third party is involved, the full burden of correct product classification, selecting the applicable requirements, risk assessment, adopting protective measures, and compiling the supporting evidence rests with the manufacturer. Module A is sufficient only when the team can demonstrate not only that the product is safe, but also why that particular conformity assessment route was permissible for that product, in its actual intended use and configuration.

From a design perspective, this means that conformity decisions cannot be separated from engineering decisions. For machinery, what matters is not only the product’s function, but also whether it falls into categories subject to specific procedural requirements and whether the manufacturer’s demonstration of conformity is actually based on technical solutions correctly selected for the identified hazards. Simply stating that the solutions are “standards-compliant” is not enough. Their suitability, consistency with the risk assessment, and actual implementation in the design, software, control system, and user documentation must be demonstrated.

In practice, this boundary is most visible in products with safety functions implemented in a non-standard way or clearly departing from solutions commonly accepted in a given industry. A manufacturer may want to stay with internal production control because they know the product, have their own engineering team, and have prepared the documentation. The problem arises when the risk assessment does not support the selection of protective measures, the description of the safety functions does not match the actual control logic, and the justification for the adopted deviations is incomplete. In that situation, the independent route becomes difficult to defend, even if the obligation to involve a Notified Body does not yet follow directly from an absolute procedural requirement.

The same applies to complex products subject to several conformity regimes at the same time. The technical correctness of the solution alone does not determine whether it is possible to remain within Module A. What also matters is how the individual procedures overlap and what conditions must be met in each of them. These are precisely the cases in which procedural independence is most easily mistaken for simplification that the documentation later cannot support.

• Module A is sufficient when product classification, risk assessment, selection of technical solutions, and documentation form a coherent conformity case.
• The involvement of a Notified Body should be verified in particular when the product belongs to special categories, departs from accepted technical solutions, or is subject to several regimes at the same time.
• If the team cannot independently defend the completeness of the technical documentation, staying with Module A becomes an organisational risk rather than a saving.

Only against this background is a procedural reference needed. The regulations applicable to the given type of product determine whether the conditions for remaining within Module A are met or whether third-party involvement is required. This is not decided by technical intuition or by the belief that the risk “is not high”. That is why the question of whether you can start with Module A and bring in a Notified Body later is often simply asked too late. If the design choices and evidence were not prepared from the outset with that possibility in mind, involving a third party later will usually expose weaknesses in the project rather than eliminate them.

The decision must be built into the design

Determining whether a given product requires the involvement of a Notified Body or whether the internal production control procedure is sufficient is a design decision. It should not be made when ordering components or just before final acceptance. The right time is the stage at which the team still has real influence over the safety architecture, the product’s limits of use, the scope of safety-related functions, and the way the technical documentation will be structured. If the decision is postponed until the end, the organisation usually falls into false economy: formally it stays with Module A, but designs the product as if filling the gaps later would be technically neutral. In practice, it is not.

The consequences are predictable. Changing the conformity assessment route near the end of the project usually means going back to the assumptions, reassessing risk, correcting control solutions, filling gaps in the evidence, and arguing over who in the organisation was responsible for the decision. That is exactly why a short but rigorous internal process is needed before the concept is frozen.

Such a process does not need to be elaborate, but it should be unambiguous. First, the product must be classified and its intended use and limits of use established. Next, a list of applicable regulations should be prepared and a preliminary risk assessment carried out, sufficient to identify critical safety functions and the points at which design choices may determine the subsequent conformity assessment route. Only on that basis can a procedural decision be made: either we stay with Module A or we prepare the project for third-party involvement to the extent required by the regulations or by the adopted evidence strategy.

For that decision to be workable, it needs an owner and clear input from engineering, automation, safety, conformity, and documentation. Without that, the issue returns at the worst possible moment: under delivery deadline pressure or after comments from the customer, importer, or market surveillance authority. At that point, the organisation is no longer solving a conformity problem; it is making up for the lack of a design decision.

• a description of the product, its intended use, and its limits of use,
• a preliminary risk assessment and identification of critical safety functions,
• a list of applicable regulations and a decision on the conformity assessment procedure,
• an evidence plan, including the responsibilities of suppliers and the integrator.

This is particularly clear in retrofits, line integration, and machine reconfiguration. For a new product, it is usually easier to identify the manufacturer and assign responsibility for the overall concept. With changes to existing systems, the situation is more difficult, because in addition to choosing the conformity assessment route, it is necessary to determine whether the scale of the intervention creates new manufacturer responsibility. If the change affects control logic, safety functions, the way machines interact, or the intended use, the conclusion that it is “just a retrofit” is no longer obvious. In that case, the question of a Notified Body is not a separate formality, but part of a broader decision on whether the scope of changes requires the entire safety concept to be reassessed and whether the information from suppliers and integrators is sufficient to support the final product.

If an organisation is considering working with a Notified Body, it should do so with a specific objective. That objective may be to confirm that a third-party route is mandatory, to assess a strictly defined scope, to organise the supporting evidence, or to reduce the risk of disputes over the technical solutions adopted. Such a discussion has value only if the team can already present its rationale: why it considered Module A sufficient, which technical solutions it adopted, which deviations it allowed, and who approved those decisions. Conformity cannot be credibly added after the fact. It has to be designed in parallel with the product.

Example first, regulations second

The most common mistake is to reduce the decision about a Notified Body to the cost of the procedure or to habit: “we have always done this under module A”. In practice, however, it is first and foremost a decision about product qualification and manufacturer responsibility, and about the quality of the design assumptions. If those two elements are inconsistent, a dispute over the procedure itself is only the consequence of an earlier problem. A team that assumes too early that module A is the obvious choice usually discovers the gap only when compiling the technical documentation, preparing the EU declaration of conformity, or during the review before CE marking. At that point, it turns out that the issue is not missing external confirmation, but a lack of continuity in the justification.

For the project, this has a very specific consequence: the technical team designs to assumptions that no one has formally verified, while the people responsible for placing the product on the market receive documents that are difficult to defend in the event of an inspection or customer questions. That is why cost must be assessed more broadly than through the procedure alone. The risk of later correction affects control systems, guards, operating logic, the instructions, marking, and the content of the EU declaration of conformity. For this reason, an earlier review of product qualification is often less costly than trying to rescue the project later, even if it ultimately confirms that the involvement of a Notified Body is not required.

A typical example concerns a line retrofit that was initially treated as a straightforward integration of existing machines. The team adopted module A as the natural choice because ready-made components were used and suppliers declared conformity for their equipment. Only when the supporting evidence was being organised did it become clear that the actual intervention had been much deeper: the logic governing interaction between stations had been changed, common safety functions had been added, stop and start-up sequences had been rebuilt, and responsibility for the final behaviour of the whole had in fact shifted to the integrator. So the problem was not the omission of a Notified Body as such, but the adoption of inconsistent assumptions.

The documentation lacked a coherent risk assessment for the entire system, there was no justification for the boundaries of the retrofit, and the parameters and architecture of the safety functions were described only fragmentarily, mainly by referring to component suppliers’ documents. When the question arose about the correct conformity assessment route and whether module A was sufficient, the team had to go back not to forms, but to the design itself: to organise the product qualification, complete the supporting evidence, and sometimes change the technical solutions. That is the key lesson from such cases. The involvement of a Notified Body may help structure the formal route and clarify the scope of the assessment, but it will not fix a weak design, replace a sound risk assessment, proper selection of protective measures, or coherent documentation.

Only after that groundwork has been done does it make sense to refer to the legal framework. It is the legislation applicable to the specific product, together with the conformity assessment procedures provided for it, that determines whether the involvement of a Notified Body is mandatory or whether module A can be used. The manufacturer, importer, integrator, or customer cannot choose freely between those routes if the product qualification and the applicable legal act lead to a different outcome. On the other hand, the mere presence of a Notified Body does not transfer responsibility for the content of the EU declaration of conformity or for whether the CE marking reflects the actual conformity assessment route. Responsibility for that remains with the entity placing the product on the market.

The practical takeaway is simple. The least expensive decision regarding a Notified Body is one made early: after product qualification, after checking the scope of the modification or integration, after assessing the quality of the supporting evidence, but before the project is closed out and the declaration of conformity and CE marking are prepared. If the documentation shows warning signs—a conflicting description of the product boundaries, no single owner of the risk assessment, excessive reliance on supplier declarations, or uncertainty about who is responsible for the overall safety functions—a review is needed before the final conformity assessment. Its purpose is not formalism. The point is to ensure that the conformity assessment route matches the actual product, rather than an assumption adopted too early and maintained only because the project is already nearly complete.