Someone saw it. Now something happens.
An unsafe condition reported into a group chat is a rumour. Raised in EHSTEC it becomes a safety signal with a priority, a severity, an accountable owner and a response you can audit.
A condition or an act — before it becomes an event
Report a hazard when you see something that could hurt somebody but has not yet: a missing guardrail, a blocked exit, a frayed sling, a colleague working at height without a harness. Nothing has happened. The point of raising it is that nothing ever does.
This is the distinction that matters in EHSTEC. A hazard is a state of the world, so it is triaged and resolved. A near miss is an event that already occurred, so it is investigated like an incident. The same field report can begin either way — the built-in report template asks which it is — but from that point the two follow different workflows, because "who fixes this and by when" and "why did this nearly happen" are different questions.
Hazard reporting is also distinct from an inspection. An inspection is planned: you decide in advance what gets checked, on what schedule, against a scored form. A hazard report is unplanned and comes from whoever happened to be standing there. A program needs both, and the two feed the same corrective action queue.
Reported by the person who saw it
- Classify what was seen — hazard, unsafe act, unsafe condition — using your own categories.
- Attach photo evidence to the signal, so the person fixing it can see what the reporter saw.
- Set a location and, where relevant, the asset involved, so the report lands against the right record.
- Who reported it stays on the record even after it changes hands during triage.
Priority and severity are not the same question
- Priority is how urgently it needs attention. Severity is how bad it would be if it went wrong. EHSTEC tracks them separately, because a low-urgency, high-severity hazard is exactly the one programs lose.
- Severity runs from None through Low, Medium and High to Extreme — and "not yet assessed" is a distinct state, so an unassessed signal cannot hide among the harmless ones.
- Triage is complete only when there is an accountable owner, an assessed severity and a determined response. The signal shows you which of the three is missing.
- Watchers can follow a signal without owning it, so a supervisor stays informed without becoming the bottleneck.
Two honest endings
- Most hazards end in a corrective action — an owner, a due date, evidence on close — and the signal resolves when the fix is verified.
- Some turn out to be more than a hazard. A signal can be escalated into a full incident case, and keeps a link back to the report it came from.
- The register shows open, in review and resolved side by side, so nothing sits in an ambiguous middle state.
- Every change of hands is recorded, which is what makes "we knew about it" a defensible statement rather than an admission.
Where hazard reporting connects
Guides on hazard reporting
Give every hazard an owner
Start a free trial and turn reported hazards into triaged, owned and verified fixes.