Hazard vs Risk: What's the Difference? (With Manufacturing Examples)
Taylor Scott, Victorian safety practitioner, Ardmor Operations — Last updated 11 October 2026·7 min read
A hazard is anything with the potential to cause harm — a source, situation, or piece of equipment. A risk is the likelihood that a hazard will actually cause harm, combined with how severe that harm would be. A hazard exists whether or not anyone is exposed to it; a risk only exists where exposure creates a real chance of harm.
That distinction sounds academic until you're standing on a production floor trying to decide what to fix first. Get it wrong and you end up treating every hazard as equally urgent, or worse, deciding a hazard "isn't a problem" because nothing has gone wrong yet — which is exactly the reasoning that shows up in incident investigations after something finally does.
What is a hazard?
Safe Work Australia's glossary defines a hazard plainly:
"A situation or thing that has the potential to harm a person." — Safe Work Australia, Glossary
WorkSafe Victoria uses almost identical language in its office health and safety guidance:
"A hazard is something that can cause harm." — WorkSafe Victoria, Office health and safety: risk management
Note what's not in either definition: nothing about likelihood, nothing about how bad the outcome would be, nothing about whether anyone is currently exposed. A hazard is a fact about the world — a spinning blade, a raised platform, a flammable solvent, a wet floor, a noisy compressor, a repetitive lifting task. It exists independently of context. An unguarded blade is a hazard sitting in a locked storeroom exactly as much as it's a hazard bolted into a running production line.
What is a risk?
Risk is the second half of the equation — what happens when exposure to that hazard is added in. Safe Work Australia's glossary defines it as:
"The possibility that harm (death, injury or illness) might occur when exposed to a hazard." — Safe Work Australia, Glossary
WorkSafe Victoria's version: "A risk is the chance of a hazard causing harm."
Risk is always a function of two variables:
- Likelihood — how probable is it that someone is exposed, and that exposure results in harm?
- Consequence (severity) — if harm occurs, how serious is it: first aid, lost time, permanent injury, fatality?
Change either variable and the risk changes, even though the hazard itself hasn't moved. That's the whole point of risk assessment: it isn't asking "is this dangerous in theory?" — it's asking "given how this hazard actually sits in this workplace, right now, what's the real chance someone gets hurt, and how badly?"
Hazard vs risk: what's the difference?
| Hazard | Risk | |
|---|---|---|
| What it is | A source or situation with the potential to cause harm | The likelihood and severity of that harm actually occurring |
| Depends on context? | No — exists regardless of exposure | Yes — driven entirely by exposure, controls, and environment |
| Can it be eliminated? | Sometimes (remove the hazard entirely) | Always managed down, even when the hazard stays |
| Example | An unguarded blade | High risk on an open line; low risk locked in a switch room |
| The question it answers | "What could go wrong?" | "How likely is it to go wrong, and how bad would it be?" |
| What you do about it | Identify it | Assess it, then control it |
The same hazard, different risk: manufacturing examples
This is where the distinction actually pays off — because the same hazard can sit at wildly different risk levels depending on where it is and who's near it.
Example 1: the unguarded blade. A slitting or trimming blade with no interlocked guard is a hazard no matter where it lives. Bolted into an operator-facing production line running two shifts, with product being manually fed, that hazard carries a high risk — frequent exposure, hands near the blade, plausible severe-injury outcome. The identical blade, unpowered, stored in a locked electrical/maintenance room that only two trained technicians enter under isolation procedure, carries a comparatively low risk — the hazard hasn't changed, but exposure and likelihood have collapsed.
Example 2: forklift movement. A forklift reversing is a hazard in any warehouse. In an aisle where pedestrians and forklifts share the same path with no separation, that's high risk — frequent proximity, poor sightlines, severe potential consequence (a documented pattern in Victorian manufacturing incidents). The same forklift operating inside a fenced, pedestrian-excluded zone with mirrors and beacons is a lower risk, even though the machine itself — the hazard — hasn't changed at all.
Example 3: a flammable solvent drum. An open drum of solvent is a hazard wherever it sits. Stored correctly in a bunded, ventilated flammables cabinet away from ignition sources, it's a manageable risk. The same drum sitting open next to a grinding station throwing sparks is now a severe fire risk — same hazard, same chemical, completely different likelihood of ignition and completely different consequence if it does.
In every case, the hazard identification step alone tells you almost nothing about priority. It's the risk assessment — likelihood plus consequence, in that specific location, with those specific controls — that tells you where to act first.
Why this distinction matters practically
Conflating hazard and risk is one of the most common ways a risk assessment breaks down. Two failure patterns show up repeatedly:
- Treating hazard identification as the whole job. A hazard register that lists "unguarded blade," "forklift traffic," "flammable liquids" and stops there tells you nothing about which one needs attention this week versus this year. Without the likelihood/severity step, everything looks equally urgent — or equally ignorable.
- Downgrading a hazard because "nothing's happened yet." Low incident history reflects low realised risk, not the absence of a hazard. The hazard is still there; you've been lucky on exposure, timing, or controls holding. This is exactly the gap that incident investigations routinely uncover after the fact — a hazard that had sat "acceptable" for years the moment exposure and consequence lined up badly.
Once a risk is properly assessed — likelihood and consequence, not just "is this dangerous" — the next question is what to do about it, and that's governed by the hierarchy of controls: eliminate the hazard where you can, and where you can't, work down through substitution, engineering, administrative controls and PPE to bring the risk down as far as reasonably practicable. That last phrase carries specific legal weight in Victoria — see what "reasonably practicable" actually means under Victorian OHS law — and it only makes sense once hazard and risk have been correctly separated in the first place. You can't argue you've done what's reasonably practicable to control a risk you never properly assessed.
FAQ
Can something be a hazard but not a risk? Yes. A hazard with no exposure carries effectively no risk — a locked, isolated machine, or a chemical sealed and stored correctly. The hazard still exists (it hasn't been eliminated), but the risk of it actually harming someone is very low. The moment exposure changes — a door left unlocked, a container opened — the risk changes even though the hazard is identical.
What is a risk assessment? A risk assessment is the structured process of taking an identified hazard and working out the likelihood of harm occurring and the severity of that harm, so you can prioritise and select appropriate controls. It's the step between "we found a hazard" and "here's what we're doing about it."
Is a bigger hazard always a bigger risk? Not necessarily. A hazard with catastrophic potential severity (a pressure vessel failure, say) but very low likelihood due to strong engineering controls can carry a lower overall risk rating than a "smaller" hazard with high, constant exposure (a trip hazard on a busy walkway). Risk weighs both factors together.
Do WHS laws treat hazard and risk differently? Yes. Duty holders are required to identify hazards and then assess and control the risks arising from them — they're treated as sequential, distinct steps in both Safe Work Australia's model and Victorian OHS guidance, not interchangeable terms.
Why do people mix the two words up so often? Because in everyday speech "hazard" and "risk" are used loosely as synonyms — "that's risky," "that's a hazard" — with no technical distinction. In a WHS context the words carry precise, different meanings, and using them interchangeably in a risk assessment or incident report can genuinely muddy which step (identification vs. assessment) has actually been done.
If your hazard register hasn't been through a proper risk assessment in a while — or you're not confident your controls match the actual risk level rather than just the hazard list — book a safety audit with Ardmor. We work with Victorian manufacturers to turn hazard identification into a defensible, prioritised risk assessment and control plan.
This article is general information for manufacturers, not professional advice. Last updated: 20 September 2026.
Sources
Would you be ready if WorkSafe walked in tomorrow?
We walk your floor, check your records against what the OHS Act 2004 (Vic) actually requires, and hand you a live compliance score and a prioritised fix-list.
Not ready for that? Take the free JSEA template — three pages written for the OHS Act 2004, no email required.
