Key Figures

1. A common question

Running the survey yourself is actually the better position to be in, whichever way the results go. In the Talison Lithium case, it was WorkSafe WA that ran the investigation, including a worker survey, and found excessive workloads, poor facilities and reported bullying and harassment. The hazards were already there. The only real question was who found them first.

If your own survey finds nothing significant, that's a dated record showing you looked and had a reasonable basis for your confidence. If it finds the same kind of issues WorkSafe found at Talison, and you act on them with a documented response, resourcing changes, consultation with the people affected, that's close to the textbook version of the "reasonably practicable" standard under WA's Work Health and Safety Act 2020. Either way, you're the one steering what happens next, rather than fixing problems on a regulator's timeline with a regulator's evidence instead of your own.

What's been publicly reported so far:

• In August 2026, WorkSafe WA issued an improvement notice against Talison Lithium, operator of the Greenbushes mine 240km south of Perth, the world's largest producer of lithium.
• Greenbushes employs more than 1,500 workers.
• WorkSafe's investigation, which included a worker survey, found excessive workloads without matching resourcing, inadequate facilities, and a pattern of reported bullying, intimidation, aggression and harassment.
• Workers described feeling tired, anxious, depressed, hopeless and worthless.
• Talison has until 20 November 2026 to show it has addressed the issues.

The Australian Manufacturing Workers Union's Steve McCartney called the company's response "a bit soft," arguing WorkSafe should have mandated interim measures rather than leaving the fix to a company self-review.

The angle most people miss: it's tempting to read this as "the survey caused the problem." It didn't. It found conditions that were already unsafe. The improvement notice responds to the underlying workloads, facilities and behaviour, not to the act of measuring them. Read the full case: skodel.com/blog/worksafe-wa-talison-lithium-psychosocial-hazards.

Source: ABC News, 7 October 2026

Let's look at some potential controls for a hazard like this

Job demands
Redesign workload allocation against current, not historical, output targets. Increase staffing or automate repetitive tasks where volume has grown faster than headcount.

Bullying
Give workers an escalation path that doesn't run through the person or team being complained about, and train managers to recognise and intervene in bullying and intimidation early.

You don't need a 1,500-person mine site for this hazard cluster to show up. Watch for:

• Treat a distress signal as information, not an indictment. A survey result, a complaint pattern or a leave spike is a diagnostic result. Investigate the cause before questioning why anyone measured it.
• Check whether resourcing has kept pace with growth. If output, headcount expectations or site scale have outpaced staffing, job demands risk is already building, named or not.
• Make escalation independent of the problem. Reporting pathways that run through the person or team being complained about will under-report by design, especially in remote or isolated settings.