In GMP, we spend a great deal of time designing systems. We define responsibilities, write procedures, build training programmes, investigate deviations, manage CAPAs, review data and introduce controls intended to make performance reliable. All of that matters, but none of it works by itself. A procedure does not execute itself, a deviation system does not make people speak up, and a training record does not prove that somebody understands why a control matters. At some point, every system depends on people deciding how seriously to take it, when to challenge it, when to escalate and how to respond when reality does not quite match the process on paper.
That is why I keep returning to a simple way of looking at GMP performance: (Organisation + Systems) × Culture = Performance. It is deliberately not an additive formula, because organisation and systems give us structure, while culture determines how that structure behaves when people actually use it. Roles, interfaces, expectations, information flows and controls are essential, but without the right behaviours they can remain surprisingly fragile. The more interesting questions are therefore often very practical ones: do people raise a concern when something does not look right, can Quality and Operations disagree without the discussion becoming personal, does a manager want the real answer or the reassuring one, and when a procedure no longer fits reality, do people challenge it or quietly build a workaround around it?
Why culture acts as a multiplier
If the systems are weak, a healthy culture can compensate for a while. Experienced people notice gaps, communicate, help each other and keep the process moving. That can look impressive, but it is also fragile, because if the system only works because good people repeatedly rescue it, the organisation is borrowing reliability from individuals. The opposite is also true: an organisation can have beautifully written procedures, clear organisational charts, sophisticated digital systems and carefully defined governance, yet still be vulnerable if people do not trust the system, do not understand its purpose or do not feel able to challenge what they see.
That is what makes culture multiplicative. It amplifies whatever the organisation and its systems are capable of producing. A strong system in a culture of ownership, curiosity and professional challenge becomes more reliable; a strong-looking system in a culture of fear, silence or disengagement can become surprisingly brittle. The formal structure may be identical on paper, while the real performance of that structure is completely different.
Compliance and culture are not alternatives
I sometimes hear quality culture discussed as though it were the softer side of GMP: useful, perhaps, but secondary to the real work of procedures, specifications, validation and control. I see it differently. Compliance defines important boundaries and creates a common operating framework, while culture influences what happens inside that framework every day. We need both, and one should not be used as an excuse to weaken the other.
Good culture does not replace clear requirements. Psychological safety does not mean that every decision is negotiable, trust does not mean removing verification, and professional judgement does not mean ignoring procedure whenever somebody believes they know better. At the same time, more control is not automatically more compliant. If people stop understanding why a control exists, if they hesitate to raise problems, or if a process becomes so bureaucratic that everyone learns to work around it, adding another signature may increase activity without increasing control. The useful question is therefore not whether we need systems or culture, but whether both are reinforcing the same purpose.
You can often see culture in ordinary GMP moments
Culture is easy to discuss in broad terms and much easier to understand through small, ordinary moments. A technician notices that something is different from normal and decides whether to say something. An operator questions an instruction that technically complies with the procedure but does not make sense in the actual process. A deviation investigation reaches a convenient explanation and somebody asks whether we really understand what happened. An engineer, QA colleague and production specialist see the same design differently and remain curious long enough to understand what each perspective is protecting. A manager receives bad news and, through the way they respond, teaches the entire team whether transparency is actually safe.
None of these moments appears as a separate clause in a quality manual called “culture”, yet together they determine whether the formal system is alive or merely documented. This is also why culture becomes visible in recurring workarounds, weak escalations, repeated deviations, fragile interfaces and dependence on a few experienced people. Those are not automatically cultural problems, but they can be signals that the relationship between the formal system and the way people actually work deserves a closer look.
What should leaders look for?
If culture is part of GMP performance, I do not think the answer is to create another large culture programme simply because culture matters. That would be a rather ironic response. I would start much closer to the work and ask where people hesitate before escalating, which processes depend heavily on one experienced person, where unofficial trackers and reminders are needed to make the formal process function, which deviations keep returning in slightly different forms, and where Quality, Operations, Engineering or other disciplines are technically correct from their own perspective but still unable to reach a good system decision together.
I would also look closely at how the organisation responds when somebody says that something is difficult. Do we treat that as resistance, or as information? Do we assume the person needs more training, or do we consider whether the process itself is creating unnecessary friction? Those questions often tell us more about quality culture than a slogan on the wall.
Where compliance meets performance
This is also why I do not see compliance and performance as competing objectives. A well-designed GMP system should help people do the right work reliably. It should make risk visible, create clear ownership, support good decisions and allow problems to be raised early enough to do something useful about them. When the system does that, compliance supports performance, and when culture reinforces the same purpose, performance becomes more sustainable.
That is one of the central ideas behind From Compliance to Culture: procedures, structures and controls matter, but their real effectiveness depends on the conditions in which people use them. So perhaps the question is not simply whether your organisation has the right systems, but what your culture is multiplying. Whatever the answer is, it is likely to show up eventually in your GMP performance.
