WHS Institute / Systems That Actually Work

Near misses are free information and mostly wasted

The events where nobody was hurt contain the same lessons at a fraction of the price.

A near miss is an incident that ran out of energy before it reached anyone. Structurally it is identical to the harmful version: same conditions, same gap, same sequence, different luck. It is therefore the cheapest possible source of information about how a workplace fails, and it is available in far greater quantity than actual harm. Organisations that take near misses seriously get many opportunities to fix a pattern before it costs someone. Organisations that do not are left waiting for the version with a person in the way, which arrives eventually and teaches the same lesson at a much higher price.

The obstacles to capturing them are practical and cultural in roughly equal measure. Practically, reporting has to be quick, or it competes with the job and loses. A process that takes twenty minutes and requires a computer at the end of a shift will collect almost nothing. Culturally, people report near misses only when they are confident the report will not be read as an admission. If the first question after a report is who was involved, the reporting rate will fall, and the organisation will interpret the fall as improvement, which is the most damaging possible misreading of the data.

Reporting volume is therefore a poor performance measure in the direction most people assume. A rising count often indicates growing trust and better visibility rather than deteriorating conditions, and a falling count frequently means people have stopped bothering. What matters is not the number but what happens next: whether reports are read, whether patterns are noticed across them, whether anything visibly changes, and whether the person who reported ever hears back. Without that loop, a reporting system is just a queue that quietly convinces the workforce that speaking up is pointless.