The situation
The client had run an annual engagement survey for six years. Participation had fallen to 41 per cent, results were reported at company level only, and when the regulator asked how psychosocial hazards were being identified and controlled, the survey could not answer the question. Leadership suspected strain was concentrated somewhere in the operation but had no way to locate it.
What we did
We ran a scoping session across rosters, incident history, claims exposure and current controls, then configured the return on investment model to the client's own leave and turnover numbers. The survey was rewritten in the language of the operation, referencing swing rosters, camp conditions and lone work rather than generic office items. Measurement ran across the seven psychosocial domains at crew level.
The survey finally told us something we could act on, and it was not what any of us would have guessed.Head of Health, Safety and Environment
What the measurement found
Strain did not sit where leadership expected. It concentrated almost entirely in night shift supervisors, who carried high demand across every domain while holding the lowest reported support scores in the business. The mechanism was structural rather than cultural: the supervisor layer had absorbed a scheduling change eighteen months earlier that had never been resourced.
The result
Four controls were added to the risk register, two of which were scheduling changes rather than training. Supervisor capability training was delivered to the group the data identified rather than to the whole workforce. At the twelve month re measure, unplanned leave in the supervisor cohort had fallen 18 per cent and support domain scores had moved into the mid range. The client now holds two dated cycles of evidence of a reviewed psychosocial risk process.