What Does “Reasonably Practicable” Actually Mean for a Blue-Collar Employer with 12 Workers?

It’s the most important phrase in WHS law. Most employers have heard it. Almost none of them can tell you what it actually requires.

Every employer in Australia operates under the same core legal obligation: ensure, so far as is reasonably practicable, the health and safety of workers. That phrase — “so far as is reasonably practicable” — appears in the WHS Act more than any other qualifying standard. It is the measure against which every employer’s duty is tested.

And yet, when you ask a small contractor running a crew of twelve what it actually means, you’ll usually get one of two answers. The first is “do your best.” The second is “do everything.” Both are wrong — and the gap between what employers think the phrase means and what it actually requires under the law is where most compliance failures begin.

It’s not a vague standard — it’s a five-factor test

Section 18 of the WHS Act defines “reasonably practicable” as that which is, or was at a particular time, reasonably able to be done to ensure health and safety, taking into account and weighing up all relevant matters including five specific factors:

1. The likelihood of the hazard or risk occurring. How probable is it that someone will be harmed? Not in theory — on this site, with this crew, doing this work. The more likely the risk, the more weight this factor carries when the regulator tests what you should have done about it.

2. The degree of harm that might result. If the risk does materialise, how serious is the outcome? A minor inconvenience and a life-altering injury are assessed very differently. When the potential harm is severe, the standard for what you should have done rises sharply.

3. What the employer knows, or ought reasonably to know, about the hazard and the ways to manage it. This is the factor that defeats the most common employer defence: “we didn’t know.” The test is not what you actually knew. It is what a reasonable person in your position should have known. In 2026, with psychosocial hazards a nationally publicised enforcement priority and the Code of Practice freely available, an employer cannot credibly claim they were unaware of the hazard or that tools to manage it do not exist.

4. The availability and suitability of ways to eliminate or minimise the risk. Are there controls that work for this hazard, in this context, for a business of this size? If suitable controls exist and are available, the question becomes why they were not implemented — not whether the employer knew about them.

5. The cost of eliminating or minimising the risk. Cost is the last factor in the list, not the first. It only enters the assessment after the other four factors have been weighed. And the threshold for rejecting a control on cost grounds is high: the cost must be “grossly disproportionate” to the risk. Not just expensive. Not just inconvenient. Grossly disproportionate.

That ordering matters. An employer who starts with cost — “we can’t afford it” — has the test backwards. The law requires you to assess the risk first, assess what controls exist, and only then weigh cost. If the risk is high and the control is available, cost alone will rarely be enough to justify not acting.

What this looks like for psychosocial hazards

Now run those five factors against psychosocial risk management for a small blue-collar employer.

Likelihood. One in four construction workers show high symptoms of depression or anxiety. Presenteeism rates across Australian workforces sit above 40%. Psychosocial claims have risen 161% in a decade. The likelihood of a psychosocial hazard being present on a blue-collar worksite is not speculative. It is statistically near-certain.

Severity. The average psychosocial injury claim in NSW now costs $288,542. These claims take almost five times longer to resolve than physical injury claims. Only 50% of workers with a psychological injury have returned to work after a full year. The degree of harm is severe — to the worker and to the business.

Knowledge. Psychosocial hazards have been a nationally publicised WHS priority since 2022. The Code of Practice has been available since 2021 and became legally enforceable in NSW on 1 July 2026. SafeWork NSW added 51 inspectors in March 2026, 20 dedicated to psychosocial risk. An employer in 2026 ought reasonably to know that psychosocial hazards exist on their worksite and that tools to manage them are available.

Available controls. Daily check-in systems, structured flag escalation processes, toolbox talk programmes covering psychosocial topics, supervisor training, and documented consultation processes all exist and are suitable for blue-collar worksites of any size. The controls are available. They are not experimental or unproven.

Cost. A daily psychosocial safety system for a small crew costs a fraction of a single employee’s annual wage. Set that against a $288,542 average claim cost, and the cost of the control is not remotely disproportionate to the risk — let alone grossly disproportionate.

When you run through all five factors honestly, the answer is clear. The risk is high. The harm is severe. The employer knows or should know. Controls exist. The cost is small. There is no version of the section 18 test where doing nothing is a defensible position.

“Reasonably practicable” is proportionate — not optional

This is the part that most small employers get wrong. They hear “reasonably practicable” and assume it means they only need to do what they can reasonably afford. It does not. It means they need to do what a reasonable person in their position could reasonably be expected to do — which is a much higher standard.

But it is also a proportionate standard. A 12-person crew does not need an enterprise compliance platform with dashboards, analytics, and a full-time safety coordinator. That would not be proportionate to the size of the business. What a 12-person crew does need is a system — any system — that identifies hazards, consults workers, implements controls, and documents what was done.

The scale of the system can match the business. The obligation itself does not scale down. A sole trader with three workers has the same duty to manage psychosocial risk as a national contractor with three hundred. The difference is in how much system is reasonably practicable given the resources of the business — not whether a system is needed at all.

For a small crew, “reasonably practicable” might look like daily check-ins on their phones, a flag process for raising concerns, and a record that shows the system was running. That is proportionate. That is affordable. That meets the five-factor test. And critically, it generates the documented evidence the employer will need if a claim, an inspector, or a court ever asks what they were doing to manage the risk.

The “we didn’t know” defence is dead

Factor three — what the employer knows, or ought reasonably to know — deserves its own section, because it is the one that has changed most dramatically in the past two years.

Before 2024, an employer in a small regional town running a crew of twelve could plausibly argue that psychosocial hazards were not something they were aware of, that the obligations were unclear, and that no tools existed for a business their size. That argument was always weak, but it had some basis.

In 2026, it has none. The regulatory framework has been publicised nationally. The Code of Practice is freely available online. SafeWork NSW’s enforcement priorities are published and widely reported. Industry bodies, insurers, and training providers have all communicated the obligation. Tools designed specifically for small blue-collar employers exist and are accessible.

The “ought reasonably to know” standard is objective — it asks what a reasonable person in the employer’s position would know, not what the employer actually knew. A reasonable employer running a construction crew in NSW in 2026 would know that psychosocial hazards are a top enforcement priority, that the Code of Practice sets out specific obligations, and that daily check-in systems and structured consultation tools are available.

Claiming ignorance is no longer a question of credibility. It is a legal impossibility under the section 18 test.

What an inspector is actually testing

When a SafeWork NSW inspector assesses a worksite, they are not asking whether the employer had good intentions. They are running the section 18 test: given the likelihood and severity of the risk, given what the employer knew or should have known, given the controls that were available, and given the cost — did the employer do what was reasonably practicable?

No incident is required for this assessment. A regulator does not wait for someone to be harmed. They test the controls against the standard, and if the controls are missing, a penalty notice follows regardless of whether any worker has been injured.

For a 12-person crew with no psychosocial risk management system in place — no daily check-ins, no documented consultation, no flag process, no evidence of ongoing hazard identification — the section 18 test produces a predictable result. The likelihood was high. The severity was high. The employer knew or should have known. Controls existed and were affordable. Nothing was in place.

That is a breach. Not a close call. Not a grey area. A straightforward failure of the five-factor test.

What “reasonably practicable” looks like on a real worksite

For a blue-collar employer running a crew of twelve, meeting the “reasonably practicable” standard for psychosocial risk management does not require a large investment or a complex system. It requires a system that runs, is documented, and is proportionate to the business.

SOV·32 is built to sit at exactly this point. Daily check-ins that workers complete on their own phones in under a minute. A flag system that surfaces concerns and documents the response. A 32-week daily programme that builds psychosocial awareness into the daily rhythm of work. And a GPS-timestamped compliance trail that proves the system was running every shift.

It is proportionate — designed for small crews without desk access, not for enterprise compliance teams. It is affordable — a fraction of a single employee’s wage. And it generates the documented evidence that answers every question the section 18 test asks.

Because “reasonably practicable” does not mean perfect. It means demonstrable. And for a 12-person crew in 2026, the bar for what is demonstrable is lower than most employers think — and higher than what most employers are currently doing.

SOV·32 enters its first live beta in mid-November 2026, with general availability early 2027. Businesses can register their interest now to secure early access and priority onboarding. Book a demo at sovsystems.com.au.