What counts as a near-miss on a construction site?

Any unplanned event that had the potential to cause injury, illness, or damage but, this time, did not. A dropped tool that lands clear of a worker, a crane load that swings closer than it should, a ladder that slips but is caught before anyone falls, a near-collision between a forklift and a pedestrian walkway, each is a near-miss. The common thread is potential, not outcome: nobody was hurt and nothing was damaged, but the conditions that could have caused harm were present. Reporting the event, not just the lucky outcome, is what makes it useful.

Why do near-misses matter more than their lack of injury suggests?

Because they are a leading indicator, a measurement of risk that arrives before the outcome, not after it. Safety research going back decades (the Heinrich triangle and its successors) finds that a large number of near-misses and unsafe conditions typically sit beneath every serious injury, meaning the hazard that eventually causes a recordable has usually shown itself many times first, as a near-miss nobody acted on. A program that reports and reviews near-misses is catching the hazard while it is still just a warning. A program that only tracks recordables is waiting for the warning to become an injury before it responds.

How does near-miss reporting relate to other leading indicators like toolbox talks and inspections?

They are the same family of measurement, prevention activity tracked before an outcome, just captured at different points in the day. A toolbox talk and a job hazard analysis happen before the work starts; a near-miss report happens during or right after an event that almost went wrong; a safety observation or PPE inspection happens as a spot-check while work is underway. Together they form the leading side of a safety program, sitting beside the lagging counts, recordables, TRIR, DART, that only show up after something has already happened. A mature program tracks both sides, not just the one that is easier to count after the fact.

Construction Glossary · Safety

Near-MissReporting

Nobody got hurt this time. That is exactly why the report matters, it is the earliest warning a safety program ever gets.

Definition

A near-miss is an unplanned event that could have caused injury, illness, or property damage but did not. It is a leading indicator, a signal that arrives before, not after, an outcome. Because many near-misses typically precede each recordable incident, a program that catches and reports them sees its risk earlier than one that only counts injuries after they happen.

The warnings sit below the outcomes

Near-misses outnumber recordables by a wide margin, which is the point: the base of the pyramid is where the earliest, cheapest warning lives.

What a near-miss report needs

A near-miss report is only as useful as what it captures, the event, the condition behind it, and what closes the gap before it recurs.

Field	Captures	Why it matters
What happened	The event, in plain language, as it unfolded	Context the count alone loses
Category	Struck-by, fall, caught-between, electrical, other	Groups the signal so patterns surface
Corrective action	What was fixed, or who owns the fix	Turns a report into a closed loop

Near-miss counts sit beside lagging rates like TRIR, and feed the same program as a toolbox talk.

How POD reads near-misses

POD reads near-miss counts and categories from your safety records and daily reports, gated to what you report. A near-miss is a leading indicator, and POD keeps it beside the lagging recordable counts so the warning signs sit next to the outcomes.

Near-miss reporting guide

Frequently asked questions

The leading side of the pyramid

Near-misses are one leading indicator among several. See how the whole leading set sits beside the lagging rates.

Leading vs Lagging Back to the Glossary