What does GMP really require? At first glance, the answer seems obvious: documented processes, qualified equipment, trained people, controlled materials, traceable records, effective investigations, appropriate oversight and all the other elements needed to manufacture a product consistently and safely. Of course those things matter. But after years of working in regulated environments, I have become increasingly interested in the space between having the system and being able to rely on it.
A procedure can exist and still be difficult to use. A role can be defined and still leave an interface unclear. An escalation pathway can be documented and still fail because people hesitate to use it, and a review step can be completed while adding very little if the reviewer does not understand what they are expected to challenge. So for me, the more useful question is not simply whether the GMP system is present, but whether it is trustworthy enough to support good decisions under real operating conditions.
Trust is not the opposite of control
Trust in a GMP environment is sometimes misunderstood as something informal: confidence in good people, a positive atmosphere or the absence of challenge. That is not what I mean. Useful trust is built on visibility, competence, clear ownership and a system that behaves predictably enough for people to rely on it. I trust a process when I know what should happen, who owns the next step, what information is available and what happens when something does not go according to plan. I trust a colleague when I know they will raise a concern rather than quietly work around it, and I trust a quality system when an uncomfortable observation can become visible without the person raising it becoming the problem.
The same applies to decisions. A decision becomes easier to trust when the assumptions behind it are visible, when it is clear what is known and what is uncertain, and when we understand what would make us reconsider it. None of that removes control; it makes control more meaningful because it gives people something they can actually use and challenge.
GMP has to work under pressure too
This becomes especially visible in radiopharmaceutical operations, where time matters, materials move quickly, products decay and logistics are tightly connected to production and patient schedules. Pharmaceutical requirements and radiological considerations can interact in ways that are not obvious from one discipline alone. In that environment, a system that only works when everybody has unlimited time and perfect information is not a reliable system.
People need to know where to look, who can decide, which information matters and when an issue must be escalated. That is why transparency is useful, not as surveillance but as operational visibility. If a team can see the process, the current state, the open questions and the ownership clearly enough, it can act earlier and with more confidence. When information is hidden across inboxes, individual spreadsheets or informal conversations, the organisation is forced to rely on people remembering what the system should already make visible.
Reliable systems reduce the need for heroics
There is a pattern I have seen many times in operational environments: the process looks stable because experienced people keep rescuing it. They know whom to call, they remember the workaround, they know which approval really matters and they spot missing information before anybody else notices. That expertise is valuable, but if the same expertise is required every day simply to keep routine work functioning, the system is using people as an invisible control layer.
That is not the same as resilience. A reliable GMP system should make competent performance easier, not require competent people to compensate continually for weak interfaces, unclear ownership or poorly designed workflows. Experience should add judgement and depth; it should not be the hidden mechanism that keeps the process from falling apart.
Professional judgement needs structure
I also do not believe the answer is to remove structure and tell experienced people to use common sense. Professional judgement is strongest when the system gives people enough context to use it well. What are we trying to protect? What is known and what remains uncertain? Which requirements are fixed, which assumptions are still open, who owns the decision and what would trigger reconsideration? Those questions make judgement more transparent and more defensible.
Without that structure, “professional judgement” can become little more than another way of saying that the decision lives in somebody’s head. With it, judgement becomes something the organisation can understand, discuss and learn from rather than something that disappears when the experienced person leaves the room.
Compliance should support daily operability
This is where I think the phrase Where Compliance Meets Performance becomes practical rather than promotional. The objective is not to make GMP lighter simply because a control is inconvenient, nor is it to add controls because more activity feels safer. The objective is to understand what the control is protecting and design the simplest reliable way to protect it in daily operation.
Sometimes that means strengthening a control, sometimes clarifying ownership or improving information flow, and sometimes redesigning a process that technically complies but repeatedly creates workarounds, delays or confusion. It can also mean asking whether a control that once made sense is still doing useful work. The point is not less control or more control by default, but control that works in reality.
Trust is visible in the way the system behaves
Trust is not a value statement on a wall. It is visible in very ordinary things: whether somebody can challenge a decision without being treated as difficult, whether the team can see what is waiting and who owns the next step, whether an investigation is trying to understand or merely to close the event, and whether people know which issues they can resolve themselves and which require escalation. It is also visible when two disciplines see a problem differently: can those perspectives be brought together before one simply overrules the other?
One of my favourite practical tests is also one of the simplest: does the process still work when the person who usually saves it is on holiday? That tells us a great deal about whether the GMP system can actually be relied upon.
From compliance to performance
Compliance is essential. It gives us boundaries, traceability, consistency and disciplined control. But sustainable GMP performance asks one more thing from the organisation: make those controls work in reality. That requires capable people, clear ownership, useful information, professional judgement and enough trust for problems to become visible before they become expensive.
That is one of the themes I explore in From Compliance to Culture. The real strength of a GMP system is not demonstrated only when everything goes according to plan; it becomes visible when something changes, something fails, information is incomplete or two legitimate objectives compete, and the organisation can still make a clear, defensible and workable decision. So perhaps the question is not only whether we are compliant, but whether the people using the system can actually rely on it. That is where compliance begins to become performance.
