Keith Butterfield, Yacht and Marine Surveyor. Costa del Sol, Spain.
If decisions increasingly favour the cleanest record, are we rewarding safety — or its appearance?
If selection rewards fewer findings, the incentive shapes what becomes visible.
It is what gets reported, what gets recorded, how findings are treated, how performance is judged — and how behaviour gradually adapts. This has unintended consequences on certification for ever-cleaner inspections, surveys and outcomes. It affects technical engagement and the confidence placed in apparently good outcomes if the underlying incentives remain unchanged.
For example: non-OEM bearings inside an extended service interval nobody re-verified. Who verified the parts? Name the person. Not the department. The person.
A generator has 13,466 hours into a 24,000-hour interval. Four oil analyses. No significant wear particles.
Last sample: 19 days and 127 running hours before a bearing failure. Every indicator was green. Every record was accurate.
The bearing replacement interval was never a number. It was a manufacturers promise with a condition attached on the basis that genuine parts were used. Non-genuine bearings went in during overhaul. Copper backing instead of aluminium. Weaker bonding. Nobody re-verified the condition. No form had a field for it. So no file existed. That file would have listed every component and its post-overhaul certification.
Nobody chose that. It fell out of two sets rules that never spoke to each other. It would have been the only systematic parts-traceability record on board.
Oversight was not skipped. It was displaced. Records intervals. It does not record what those intervals assume. Find the condition it rests on. Then name who checks it. If nobody owns the assumption, the interval is not a control. It is a hope with a number on it.