Key takeaways
- Disruptions of the past few years, from power interruptions to supply shortages and system outages, pushed almost every business to build temporary workarounds to keep running.
- Temporary controls are easy to introduce, and nobody is ever accountable for removing them.
- The common survivors are manual record fallbacks, stretched calibration intervals, deferred maintenance, relaxed checks, skipped approvals and equipment nobody registered.
- A control that is still operating but no longer justified is an undocumented change to your management system.
- A stable period is the opportunity. This work is far easier now than in the middle of a disruption.
Over the past few years, almost every operating business has had to build workarounds. Power interruptions, supply shortages, system outages and staff absences arrived on somebody else's schedule, and the work still had to get done. Those workarounds were sensible, proportionate and often ingenious.
Most of them are still running. And most were never written down, never formally approved, and have never been reviewed against conditions that no longer apply.
Why temporary controls become permanent
This is rarely carelessness. The instinct is to treat it as a discipline failure, but it is a structural problem.
Introducing a temporary control is easy and urgent. Somebody has an immediate problem, they solve it, and the work continues. Everybody understands why. The decision is made quickly, usually verbally, usually by whoever is closest to the work.
Removing one is neither easy nor urgent. It requires somebody to notice that a condition has changed, to establish that the original reason has fallen away, to confirm nothing else has come to depend on the workaround, and to take responsibility for reverting it. There is no deadline forcing any of that, and no consequence for leaving it. So it stays.
Then something more subtle happens. New people join and learn the workaround as the normal method. The original justification is forgotten, because it was never recorded. Within two or three years, an emergency adaptation has become simply the way the process runs, and nobody in the building can tell you why.
What is most likely still in place
These are the ones we see most often across manufacturing, engineering and processing operations in South Africa.
| The workaround | Why it was introduced | What it means now |
|---|---|---|
| Manual record fallbacks | Systems were down, so results were written on paper and captured later | Gaps and late entries became normal. Traceability is weaker than the records suggest |
| Stretched calibration intervals | Instruments could not be scheduled reliably, so intervals slipped | The interval was never restored. The basis for the current interval is no longer defensible |
| Deferred maintenance | Planned work was displaced by unplanned interruptions | A backlog that was rational at the time is now simply a backlog |
| Relaxed in-process checks | Checks were reduced to recover lost production time | Reduced sampling became the standard sampling. Nobody re-justified it |
| Skipped approval steps | Approvers were unreachable, so work proceeded and was signed later | Signing after the fact is now habit, which makes the approval meaningless |
| Generators and fuel storage | Installed quickly, often under real pressure | Frequently missing from the environmental aspects register, the legal register and the hazard identification for refuelling |
| Temporary suppliers | Approved suppliers could not deliver, so alternatives were used | Still in use, often never formally evaluated or added to the approved list |
What the standard actually requires
ISO 9001 anticipates this situation. Clause 6.3 requires that changes to the quality management system are carried out in a planned manner, with consideration of the purpose of the change, its potential consequences, the integrity of the system, resources, and responsibilities.
Emergency change is a legitimate part of operating a business, and no standard expects a formal change process in the middle of an emergency. That is not the issue.
The issue is that an emergency change which becomes permanent is a permanent change that was never assessed. If an auditor asks why your calibration interval is what it is, and the answer is that it drifted during a disruption and nobody moved it back, the interval is not wrong so much as unjustified. And the difference between wrong and unjustified matters less than people hope, because you cannot demonstrate either way.
This connects to something we have written about before, which is why corrective actions keep coming back. A control introduced without analysis is the same failure in a different direction: something changed in the system without anyone establishing whether it should have.
The environmental and safety tail
The quality consequences are the most numerous, but equipment installed in a hurry is usually the most serious, and it is the one most often missed.
A backup diesel generator brings a set of obligations with it. Air emissions. Noise. Bunded fuel storage and spill containment. Refuelling as a hazardous activity requiring its own risk assessment. Possibly a permit or a registration, depending on size and location.
Many of these installations went in during a genuine operational emergency, commissioned by whoever could get one delivered. They frequently never reached the environmental aspects register, the legal and other requirements register, or the hazard identification. They are still on site, still fuelled, and in an integrated system they are still your responsibility. ISO 45001 expects changes like these to be assessed for their health and safety effect, which makes a retrospective assessment the sensible first step.
Circumstances change, too. Equipment that was indispensable a few years ago may now be an expensive asset carrying compliance obligations for a risk that has reduced. That is worth a deliberate decision rather than a default.
How to find them, practically
Do not start with a document review. The workarounds are not in the documents, which is the entire problem.
Ask the people who run each process one question: what do you do differently now compared with a few years ago? You will get a more accurate list in twenty minutes on the floor than in a week of reading procedures. People remember what they changed, even when nobody wrote it down, and they are usually glad to be asked.
Then, for each item, three questions. Is the original reason still true? If not, should this revert, or has it turned out to be genuinely better? And either way, where is that decision recorded?
Some workarounds will turn out to be improvements worth keeping. That happens often, and it is a good outcome. But keeping one should be a decision with a name attached to it, with the procedure, training and records updated to match.
Why now is the moment
This work is far easier in a stable period than in the middle of a disruption, and it will be easier now than if conditions tighten again.
A business that closes out its old workarounds now is in a stronger position either way: cleaner and more defensible if conditions hold, and starting from a known baseline if they do not. One that leaves them in place is carrying undocumented controls whose justification has expired, and it will discover them at an audit rather than at a time of its choosing.
If you would like a considered view of where your management system has drifted from what is actually documented, contact us. It is a common piece of work, and it is usually less daunting than businesses expect once the list exists.