Good Manufacturing Practice (GMP) Needs to Move Beyond Training Records
This blog explores the gap between delivering training and demonstrating capability matters - and how a formal competency framework could close it.
If you have sat through a GMP inspection, you know the moment. The inspector asks to see evidence that personnel are competent. The company produces a training matrix with names, SOPs, and tick boxes. They pull out only CVs, list of courses attended, and read-and-understood signatures without any evidence of examination or practical demonstration.
The inspector listens, then asks: “How do you know they can actually perform this task to the required standard?”
This is where the room goes quiet. The question is not “was the person trained?” It is “can the person perform this task, in this context, to the required GMP standard, and can you prove it?”
Training is an input. Competence is an outcome. And too often, the pharmaceutical industry documents the former and assumes the latter.
The PIC/S GMP Guide (PE 009-17) contains a clear expectation that training effectiveness must be assessed, not merely delivered.
Part I, Clause 2.11 states: “Continuing training should also be given, and its practical effectiveness should be periodically assessed.” This is a requirement to assess whether the training worked; not to check whether people attended.
Annex 1 (Manufacture of Sterile Medicinal Products) treats competence as something to be demonstrated, documented, and periodically reconfirmed:
Personnel must receive gowning and aseptic behaviours qualification and be confirmed by assessment and periodic reassessment at least annually (Clauses 7.3-7.4);
Initial validation requires at least three consecutive satisfactory simulation tests covering all working shifts. Each operator must then participate in at least one successful Aseptic Process Simulation (APS) annually. Where manual operations are performed, operators must complete three consecutive successful APSs, with revalidation every six months. (Clauses 9.38-9.39);
Where investigation reveals operator-related failure, activities must be limited until the operator is retrained and requalified (Clause 9.46).
Annex 17 (Parametric Release) (Clause 3.6) contains perhaps the most explicit competency requirement:
Personnel should be given specific training on Real Time Release Testing (RTRT) technologies, principles and procedures. Key personnel should demonstrate adequate experience, product and process knowledge and understanding. Successful implementation of RTRT requires input from a cross-functional/multi-disciplinary team with relevant experience on specific topics, such as engineering, analytics, chemometric modelling or statistics.
Note the separation of four distinct concepts – experience, product, process knowledge and understanding - they are not the same thing.
Most companies are good at documenting training inputs - training matrices, SOP read-and-understood records, and onboarding checklists. These are necessary and valuable, but they are not the same as competency.
| Input (documented) | Output (expectations) |
| Qualification - a credential or certificate | Competency - demonstrated ability to perform to a defined standard |
| Experience - time spent in a role | Assessed performance - evidence the person can do the job consistently |
| Training - instruction delivered | Training effectiveness - proof the instruction resulted in capability |
A person can hold a degree, have ten years of experience, and have attended every training session, and still not be competent in a task they have never performed or been assessed against.
The question is why this principle (demonstrated, assessed competence) is applied rigorously in sterile manufacturing but inconsistently everywhere else.
This gap parallels quality risk management. ICH Q9 and PIC/S Annex 20 require risk assessments, that companies conduct by identifying hazards, and scoring severity and probability. Too often the assessment is filed away and downstream work proceeds regardless. The risk assessment becomes a compliance artefact rather than a decision-making tool.
Competency assessment suffers the same pattern. A person who scores poorly receives the same training plan as someone who scored well.
The solution in both cases is that the assessment must drive downstream action - tailored training, supervision levels, authorised task lists, reassessment frequency, and CAPA effectiveness checks.
Beneath this gap lies a more fundamental problem. Many companies lack a structured mapping of skills to job roles. Job descriptions list responsibilities and qualifications but rarely define the specific skills required for each task, their criticality, expected competency level, appropriate assessment method, reassessment frequency, or evidence required.
Without this architecture, competency assessment becomes ad hoc. For example, a supervisor observes an operator, signs a form, and competence is declared.
Training matrices only tell what training someone has received, but role-skill matrices tell what competence someone needs, and whether they have demonstrated it.
The pharmaceutical industry has standards for almost everything except a shared, industry-aligned standard for personnel competency. Such a standard could provide:
No company will build a comprehensive competency framework overnight, but the journey can be staged:
The industry has the regulatory mandate. PIC/S GMP does not merely ask for training. In critical areas it expects demonstrated, documented, periodically reassessed competence. The industry has the risk management framework. What is missing is the connective tissue: a structured, industry-aligned approach to defining what competence looks like, assessing it consistently, and feeding results back into development and risk management.
Training records tell you what went in. The future of GMP compliance needs to show what comes out. The question is not whether your people were trained. It is whether they are competent; and whether you can prove it.
Register for our upcoming webinar to hear about a competency framework and the use of tools in a digital twin model that can help your team create a more engaging training experience and verify its effectiveness.