Why Psychological Safety Can’t Be Fixed With Cupcakes, Posters, or Good Intentions

By Rose Byass

Across Australia, psychological safety has become a visible organisational priority. Policies are written, wellbeing calendars are circulated, diversity roles are established, and walls and intranets are refreshed with carefully chosen language about inclusion, respect, and safety. On paper, many organisations appear to be doing all the right things.

Yet for many workers, the lived experience tells a different story. Bullying continues because the person responsible is considered “too valuable” or “too entrenched” to challenge. Misconduct and fraud are quietly rationalised rather than addressed. New starters arrive with optimism and leave within months, often blamed for “not being the right fit.” Meetings are quiet, concerns are raised privately—if at all—and after incidents, silence replaces learning. When harm occurs, the explanation offered is not system failure, but individual resilience: people are told to toughen up, cope better, or manage their reactions.

This disconnect is not merely cultural or attitudinal. It is a governance failure. Psychological safety is not a mood, a benefit, or a branding exercise. It sits squarely within psychosocial hazard management, which Australian work health and safety regulators clearly identify as a category of workplace risk capable of causing serious harm. Safe Work Australia explicitly recognises psychosocial hazards—such as bullying, role ambiguity, poor support, and unsafe leadership behaviours—as hazards that must be identified, assessed, controlled, and reviewed like any other WHS risk.

When organisations treat psychological safety as a sentiment rather than a system, they under-invest in the controls that actually prevent harm. The result is a widening gap between what organisations say about safety and what workers experience in practice—one that carries human, legal, and financial consequences.

What psychological safety is (and isn’t)

Psychological safety is not:

people feeling “happy” at work,
comfort at all times,
the absence of accountability,
“everyone being nice,”
or a perk schedule.

In the research literature, psychological safety is a shared belief that it is safe to take interpersonal risks—speaking up, admitting errors, asking questions—without fear of humiliation or punishment. Amy Edmondson’s seminal work links psychological safety to team learning behaviours and performance. What leaders often miss is this: psychological safety is not created by intentions; it is created by systems, behaviour norms, and power dynamics.

Australian data: why this is now a board-level risk

If you want a board to listen, you talk in:

claims, time loss, cost, and
controls, assurance, compliance.

Safe Work Australia’s national reporting consistently shows that mental health condition claims are among the costliest forms of workplace injury:

In 2022–23, Safe Work Australia reports a median time lost of 35.7 working weeks for mental health condition serious claims, versus 7.4 across all serious claims; and median compensation of $67,400 versus $16,300.
Earlier releases show similarly elevated time loss and compensation for mental health condition claims (e.g., 2021–22 median time lost 37.0 working weeks; median compensation $65,400).

This isn’t “soft” risk. It’s expensive, persistent, and organisationally destabilising. And this sits alongside broader Australian workplace injury experience. The ABS estimates that in 2021–22, 497,300 people experienced a work-related injury or illness (3.5% of those who worked in the previous 12 months). Boards are increasingly accountable for ensuring psychosocial hazards are identified and controlled with the same seriousness as physical hazards—because the cost curve is not going down on goodwill alone.

The legal and governance frame: “psych safety” is part of WHS hazard management

Under the model WHS laws, PCBUs must manage psychosocial hazards. Safe Work Australia provides both high-level guidance and a Model Code of Practice that sets expectations for identifying hazards, assessing risk, controlling risk, and reviewing controls. This is why posters and cupcake days fail as “solutions.” They are not controls. They do not:

reduce exposure to bullying or harassment,
improve workload design,
clarify roles and decision authority,
strengthen complaint handling integrity,
or build manager capability to handle conflict and accountability.

In other words, they do not reduce the hazard or likelihood of harm in any reliable way. ISO guidance takes the same direction. ISO 45003 provides guidelines for managing psychosocial risk within an OH&S management system (as a companion to ISO 45001). The through-line is clear:

psychological safety should be treated as a managed risk system, not a morale initiative.

“This is what culture looks like”: a real-world example from an interview

A recent interview experience crystallised a point that many organisations still struggle to grasp. At the conclusion of an interview for a senior management role, a reasonable question was asked:

How does the organisation manage psychological safety?
How does it demonstrate to employees that they are psychologically safe?

The explanation offered was straightforward and evidence-based: psychological safety begins from the moment a person joins an organisation. It is shaped through effective onboarding, clear expectations, mentoring, and having the practical fundamentals in place — a workspace, system access, logins, direction, and support to do the job well.

This framing aligns with contemporary research and good governance practice.

What followed was unexpected.

A panel member responded abruptly that what had been described was merely “hardware,” and that the organisation was only interested in “software,” such as resilience — adding that if someone lacked resilience, they should “pull up their big girl pants.” The comment was alarming, not because it was offensive, but because it revealed a common and consequential organisational error: confusing individual coping capacity with organisational control. Resilience matters. But resilience is not a substitute for:

safe systems,
clear expectations,
fair and consistent processes,
competent leadership,
or reliable mechanisms to raise concerns.

No level of individual resilience compensates for poor systems or embedded organisational dysfunction. Equally, it is not a personal failing for a new employee to expect to be set up for success when they commence a role — it is a basic organisational responsibility.

When leaders default to “toughen up,” “sink or swim,” or similar narratives, two predictable outcomes follow:

silence, where people stop speaking up about risk, errors, or concerns; and
attrition or escalation, where people either disengage and leave, or the issue eventually emerges as a formal complaint, claim, or incident.

Safe Work Australia’s psychosocial hazard research consistently links hazards such as bullying, excessive work pressure, and poor organisational support to psychological injury claims and prolonged time away from work.

Notably, the chair of the interview panel later apologised for the comment. That acknowledgement mattered. It highlighted an important truth: senior leaders often recognise when a cultural signal is wrong — yet those signals still occur when organisations lack a shared, consistent understanding of psychosocial risk and psychological safety across all levels.

Culture is not what is written in policy or spoken in leadership statements.

It is revealed — often unintentionally — in moments like these.

The neuroscience reality: humans don’t “choose” compliance the way policies assume

A big reason organisations lean on posters and policies is that they assume behaviour is rational and fully voluntary:

“We trained it, therefore it will happen.”
“We wrote it, therefore it will be followed.”

But under pressure, humans default to threat responses: freezing, fawning (appeasing), withdrawing, avoiding, or becoming defensive. In workplaces, this shows up as:

not reporting bullying because “it won’t change”,
not raising safety issues because “it will come back on me,”
not challenging a respected bully because “they’ve been here forever,”
not questioning a manager because “it will affect my career.”

Psychological safety is the condition that reduces interpersonal threat enough for people to speak up and participate. That is why Edmondson’s research ties it to learning behaviours and performance: without safety, people protect themselves, not the system. This is the core misunderstanding:

A policy can describe expectations. Only a system can reliably change behaviour under pressure.

The myth of performative culture: cupcakes, lanyards, and posters

Let’s name the pattern. When culture is failing, many organisations respond with:

highly visible “culture” initiatives,
symbolic diversity activities,
poster campaigns,
and wellness events.

These can be positive add-ons, but they are often used as a substitute for doing the harder work:

addressing bullying,
lifting management competence,
redesigning workload,
improving reporting and investigations,
and holding people accountable consistently.

This isn’t cynicism; it’s risk logic:

If bullying continues while cupcakes increase, cupcakes are not a control.
If fraud continues while posters increase, posters are not a control.
If turnover continues while slogans increase, slogans are not a control.

It’s entirely possible to have a bright, values-filled intranet and a psychologically unsafe workplace.

What “risk managing psychological safety” actually looks like

If we strip away buzzwords, psychological safety is managed through the same logic as any WHS hazard:

1) Identify psychosocial hazards

Use multiple sources:

incident and hazard reports,
grievance and complaint data,
exit interview themes,
People at Work psychosocial risk assessments,
absenteeism and turnover hotspots,
worker consultation and HSR feedback.

Safe Work Australia promotes People at Work as a validated Australian psychosocial risk assessment survey.

2) Assess risk (likelihood × consequence)

Not every issue is equal. A few examples:

chronic bullying with a “protected” bully = high likelihood + high consequence
unmanaged workload in a critical team = high likelihood + high consequence
unclear role authority in high-stakes decisions = high likelihood + high consequence

3) Implement controls (not slogans)

Controls include:

role clarity and decision rights,
onboarding that actually equips people from day one,
workload design and staffing realism,
leader training in conflict, feedback, and procedural fairness,
reliable grievance processes with natural justice,
consistent discipline for misconduct (including “high performers”),
audit and assurance on control effectiveness.

The Model Code of Practice emphasises practical guidance for managing psychosocial hazards—this is not optional window dressing.

4) Review and improve

If control effectiveness isn’t monitored, the hazard returns. This is where boards should demand:

control testing,
trend analysis,
internal audit over psychosocial risk systems,
and proof that issues are resolved, not recycled.

Safe Work Australia’s reporting and snapshots make clear that psychosocial hazards like bullying and work pressure are not theoretical—they show up in real claims and serious time loss.

Why “resilience” is not the answer (and why it’s still part of the answer)

Resilience is valuable. But resilience is the seatbelt, not the brakes. A psychologically safe organisation:

builds resilience in individuals, and
designs systems that reduce predictable harm.

If you only sell resilience, you are effectively saying:

“We expect you to absorb the impact of our unmanaged hazards.”

That is not just ethically questionable; it is operationally foolish. In jurisdictions where psychosocial hazard management expectations are rising, this posture increases exposure.

The equity problem: when two people do the same job differently

Now a real life example—where organisations often get it wrong. You have:

Worker 1 who has adjustments (e.g., autism-related accommodations) and is given more flexibility around tardiness, communication style, or productivity variability.
Worker 2 who performs the job “as required” without adjustments.
Both are paid the same.

If this is handled poorly, it can damage psychological safety for both workers:

Worker 1 may feel scrutinised, othered, or resented.
Worker 2 may feel a sense of unfairness or “double standards.”
Managers may avoid addressing performance issues out of fear or confusion.
Teams may polarise into resentment and silence.

The principle that resolves this: equality of opportunity + clarity of standards

Australia’s legal and guidance environment is clear that “reasonable adjustments” are often required to remove barriers, allowing employees with disability to participate and work safely and productively—so long as it does not create unjustifiable hardship. Fair Work guidance similarly describes reasonable adjustments as changes to work processes, tools, or work areas to reduce barriers. So what does good practice look like?

How to protect psychological safety for both workers

Keep the performance standard consistent.
Adjustments are about how work is done, not the legitimacy of the role requirements (unless the role itself is redesigned).
Make adjustments transparent in principle, confidential in detail.
Teams don’t need private diagnoses. They do need to understand the organisation’s stance: we make reasonable adjustments so people can perform; standards remain clear; we manage performance fairly.
Train managers to talk about fairness properly.
Fairness is not “everyone treated the same.” Fairness is “everyone gets what they need to do the job safely and effectively.”
Use role design and job controls.
If the job is so tight that any adjustment looks like preferential treatment, the job design may be the hazard (workload, staffing, unrealistic KPIs).
Prevent scapegoating.
If Worker 2 is carrying unrecognised load, that is a workload and task allocation risk, not a personality issue.

Handled well, this approach increases psychological safety for everyone:

Worker 1 experiences dignity and capability.
Worker 2 experiences clarity, consistency, and fairness.
The team experiences trust in management decisions.

Why onboarding and “hardware” absolutely matter

Returning to that interview moment: dismissing onboarding, tools, and functional setup as “hardware” misses a critical hazard-control truth. If workers are not properly onboarded, they experience:

uncertainty,
reduced competence,
heightened threat response,
and elevated error likelihood.

In safety-critical environments, this directly affects incident risk. In any environment, it affects confidence to speak up and integrate. The idea that psychological safety is only “soft skills” is outdated. Psychological safety is an organisational system:

induction,
access,
clarity,
mentoring,
fair processes,
and leadership consistency.

What boards should demand: a practical governance checklist

If you’re presenting this to directors or executives, ask for evidence in four buckets.

1) Hazard identification and consultation

Are psychosocial hazards identified using structured methods (People at Work, surveys, consultation)?
Are workers consulted meaningfully about hazards and controls (not just informed)?

2) Controls and capability

Are controls focused on job design, workload, leadership behaviour, reporting integrity?
Are managers trained in conflict, feedback, and procedural fairness—not motivational slogans?

3) Reporting and assurance

Are board reports showing leading indicators (hazards, control effectiveness) and lag indicators (claims, time lost, turnover)?
Is internal audit testing psychosocial controls?

4) Consequence management

Is bullying addressed regardless of tenure or performance?
Are misconduct and integrity issues investigated and actioned consistently?

If a board cannot answer these questions with evidence, psychological safety is likely being treated as branding, not risk management.

The Australian picture: why this is urgent (and not going away)

Safe Work Australia’s reporting makes the urgency hard to ignore: mental health condition serious claims involve long time away from work and high compensation relative to other claims. Sector-specific insights add another layer. Safe Work Australia’s 2024 psychological health reporting (including People at Work survey insights) highlights exposure to hazards such as high emotional demands, bullying, and occupational violence in certain industries and points to elevated claim patterns in sectors including public administration and safety. That matters because when you connect:

high psychosocial hazard exposure,
high consequence claims,
and high public scrutiny,

you get a board-level risk that cannot be wished away with good intentions.

So what should organisations stop doing?

They should stop:

treating psychological safety as a marketing theme,
equating it with “being nice” or “being resilient,”
Stop thinking that free gym, yoga studios, and fresh fruit are psych safety
delegating it to a single role (without leader accountability),
and assuming a policy is a control.

And yes—organisations should retire the mindset that if someone raises a psychosocial concern, the correct response is to “pull up your big girl pants.” Because that is not leadership.

That is hazard denial.

What organisations should do instead: a clearer way to say it

If you want a simple line to anchor your message to executives, use this:

Psychological safety is the operational outcome of well-designed work, fair processes, and competent leadership under pressure.

Cupcakes can be nice.

But cupcakes do not control hazards. Systems do.

Closing: why this matters for leaders, not just workers

Psychological safety isn’t “soft.” It is:

risk,
continuity,
performance,
governance,
and reputation.

If you don’t manage psychosocial hazards, you don’t just lose staff.

You lose:

the truth (because people stop speaking),
early warnings (because people stop reporting),
and then, eventually, control (because issues surface through claims, crises, or scandal).

That is why psychological safety can’t be fixed with cupcakes, posters, or good intentions. It can only be fixed by risk management discipline—applied to human behaviour with the same seriousness we apply to physical hazards.