Institute of Work Health and Safety / Risk

The gap between the procedure and the job

Safety systems describe how a task should be done. People do the task the way it can actually be done. Closing that distance, rather than denying it, is most of the discipline.

Every organisation of any size has documents describing how its work is performed. There are procedures, method statements, permits, inductions and checklists, and collectively they describe a version of the job in which the equipment is available, the weather cooperates, the previous crew finished on time, nothing is missing, and the person doing the work has as long as the task requires. That version is coherent and internally consistent, and it is not what happens on any particular day.

The actual job is done with what is present. A part is late so the sequence changes. A guard is awkward so somebody works around it in a way that is faster and mostly fine. A form is filled in at the end of the shift for the whole shift because filling it in as you go is impossible while wearing gloves. None of this is misconduct and most of it is competent adaptation by people who are trying to get work done under conditions the procedure did not anticipate.

The dangerous move is to treat that gap as a discipline problem. If the official response to every difference between the document and the work is that somebody failed to follow the procedure, then the only rational behaviour left to a worker is to stop telling anybody what the work is actually like. The gap does not close. It becomes invisible, which is considerably worse, because now the organisation's picture of its own operations is wrong and confidently held.

Treated the other way, the gap is the most valuable information available. Every workaround is a piece of evidence that a procedure does not fit the conditions, and the person performing the workaround usually knows precisely why. A supervisor who can ask what makes this hard and get an honest answer has access to a stream of diagnostic information that no audit produces, because an audit measures conformance to the document rather than the suitability of the document.

This is why the emphasis in serious practice has moved toward understanding normal work rather than cataloguing deviations. The question is not whether people follow the procedure. It is whether the procedure describes something a competent person can do, every time, under the conditions that actually occur, while also meeting the other demands placed on them. When the answer is no, the procedure is the thing that needs to change, and it will keep needing to change as the work changes.

There is a limit to this, and it should be stated plainly, because the argument gets stretched. Some rules exist because the consequence of getting it wrong is catastrophic and irreversible, and those are not negotiable regardless of how inconvenient they are. Isolation before working on energised equipment, fall protection at height, confined space entry controls: these are not procedures to be adapted to conditions. The discipline is knowing which rules are in that category and defending those absolutely, precisely so that the rest of the system has enough flexibility to be honest.

Most organisations get this backwards. They defend the whole document set with equal severity, which means the trivial and the critical carry the same weight, and workers learn from experience that the system does not distinguish between them. Then when a critical rule is broken, nobody in the chain reacted differently to it than they would have to a missing signature, because nothing in how the system was operated ever told them it was different.

The last observation about the gap concerns who is able to close it. A worker knows the work does not match the procedure and cannot change the procedure. A person who can change the procedure is rarely present when the mismatch occurs. Almost every durable improvement in this field comes from arranging a route between those two positions that does not require anybody to make an accusation, and almost every failed safety system has that route missing, blocked, or present on paper and understood by everybody to be unusable. The route matters more than the content of any particular procedure, because it is what allows all the others to be corrected.