A structured programme of a guided hypnotic process and breath practice, delivered to a cohort, in which no participant discloses a problem to anybody at any point. Not to us. Not to a manager. Not to a group. Not on a form.
The first two figures work themselves out from today's date every time this page is opened, so they are right on the day you read them and they keep counting on their own with nothing to maintain. The third is a design decision and it does not change.
Psychosocial hazard duties became enforceable in every Australian jurisdiction on 1 December 2025. In New South Wales the Codes of Practice became mandatory compliance benchmarks on 1 July 2026.
An organisation that controlled a hazard and cannot evidence it is, on paper, an organisation that did nothing.
Everyone in this room has already worked out what an honest answer would cost them. That calculation happens before a single word is typed, and no amount of survey design reaches around it.
Anyone answering a question about pressure, bullying, exhaustion or safety weighs that up first. They weigh the anonymity claim against how small their team is, how identifiable their writing is, and how the last person who spoke plainly was treated. That is not a flaw in the instrument. It is a rational assessment of their own career.
There is a second layer underneath that one. Plenty of people cannot answer honestly even when they want to, because they have not had a chance to process what they are carrying. Asking them to name it on a form makes them construct a story on the spot. Sometimes that story is accurate. Often it is the mind's best guess, and you then manage the guess.
Written from your chair rather than ours. If any of it is wrong for your organisation, say so in the first ten minutes and we will both save a morning.
The Code says risk is minimised by changing the design of work, the systems of work, the work environment and workplace interactions. Every one of those sits with operational leaders who own a budget and a delivery date. You are answerable for an outcome you cannot unilaterally cause.
Your data is self-report from people who have already weighed what an honest answer costs. The Code says the quiet part out loud: anonymous tools matter because they "protect workers from stigma or other adverse outcomes when reporting hazards or concerns."
Officers carry a due diligence duty in their own right under section 27 of the Work Health and Safety Act. That is a named individual, not a company.
Rosters, workload, role clarity and manager capability are the right answers and they take one to two years to land properly. The obligation started on 1 December 2025.
The Code's own examples include emergency response, investigating crimes, supporting victims, content moderation and delivering bad news. You cannot roster away exposure to trauma in a job that is made of it.
Change the roster tomorrow and the people who went through the last three years are still the people who went through the last three years. Almost nobody sells you a control for that.
This is the part of the duty with no obvious supplier, and it is the only part we are here for.
We do not do job redesign and we will not pretend to. The Code is explicit that risk is minimised by changing the design of work, the systems of work, the environment and workplace interactions, and every one of those is yours. Anyone selling you a programme delivered to individuals as though it eliminates a psychosocial hazard is selling you something that will not survive an inspector. What we do is the shaded part, and the shaded part currently has no supplier.
The Code tells you to eliminate first, then order the remaining controls from most to least effective. It also says "the greater the risks, the more that is required to be done to eliminate or minimise it," and that this "may mean using more than one, or a combination of control measures." Here is exactly where we sit in yours.
Remove the exposure. Stop the practice. End the arrangement causing it. The Code says you must aim here first.
Design of work, systems of work, environment and conditions, workplace interactions. The Code's own list, and the high-order controls.
The residual exposure that cannot be designed out, the one to two years before redesign lands, and the load already carried by the people in the building. Delivered without a single disclosure, which is why the people who most need it do not opt out.
The Code: "the least reliable controls," providing "the lowest level of health and safety protection." Consider these last and only in combination. A policy and a poster live here.
And it introduces no hazard of its own, which is not true of the alternatives. Interventions that require people to speak generate disclosures you are then obliged to act on, records that can be requested, and a mark on whoever took part. This one produces no free text, no individual response and no demographic data, so there is nothing to leak, nothing to subpoena and nobody to identify. For a risk committee that is a checkable statement rather than a promise.
Count them. The engagement survey. The psychosocial survey. The grievance process. The investigation. The intake call for the employee assistance programme. The return to work interview. The exit interview. Each one is a person being asked to go back and describe it.
Every one of those is well intentioned and several of them are legally required. None of them is free. Each retelling rehearses the account, sharpens it, and gives it a little more room in the sentence that starts "this is who I am." A person can arrive with something that happened to them and leave, several processes later, with something that defines them.
Fixing the system should not require condemning the people to relive it while you fix it.
That is the whole reason this is content free. Nobody is asked what happened, at any point, to anybody. Not on the questions, not in the sessions, not in the daily practice, not at the end. There is nothing to rehearse and nothing to sharpen, and the account never gets another airing.
Stated as a clinical design position and now with two pieces of published evidence underneath it. It is the reasoning the method is built on, held by a Registered Professional Clinical Hypnotherapist in practice since 2009. No outcome statistic is offered for it here, because none has been produced that we would stand behind.
At one end, a first responder pulling somebody out of a car. At the other, "by the time I get my coffee break the coffee is always cold." Nobody would call the second one trauma and nobody has to. They build the same way. Layer on layer of the same thing, or of similar things, until one ordinary Tuesday supplies the “last straw”.
The Code says exactly this and it is worth quoting to a risk committee: "Harm can be caused by a single instance or over time with repeated or prolonged exposure," and "in most circumstances, it will be a combination of psychosocial hazards which together may cause harm."
The Centrality of Event Scale measures how far an event has become a reference point for identity and for the meaning attached to everything else. A systematic review of 92 publications, with roughly 37,600 participants in its largest category, found it correlates .51 with post traumatic stress disorder measures.
The finding that matters: it contradicts the common assumption that failing to integrate a difficult memory is what causes the symptoms. Greater integration into the life story went with a worse picture, not a better one.
Gehrt, Berntsen, Hoyle and Rubin (2018). Psychological and clinical correlates of the Centrality of Event Scale: A systematic review. Clinical Psychology Review, 65, 57 to 80.
Its words: "Single session individual debriefing did not prevent the onset of post traumatic stress disorder (PTSD) nor reduce psychological distress, compared to control. At one year, one trial reported a significantly increased risk of PTSD in those receiving debriefing." And in the conclusions: "Compulsory debriefing of victims of trauma should cease."
In the two longest trials the gap between the debriefed group and the controls was widening over time.
Rose, Bisson, Churchill and Wessely (2002). Psychological debriefing for preventing post traumatic stress disorder (PTSD). Cochrane Database of Systematic Reviews, CD000560.
What these two do not show, said plainly. Neither studied grievance processes, engagement surveys or exit interviews. Debriefing after a critical incident is the closest thing anybody has studied, not the same thing. Both are correlational or preventive, so neither establishes that retelling causes harm. Centrality also correlates .55 with post-traumatic growth, higher than its .51 with distress, so an event becoming central is not simply bad. And trauma-focused therapy that a person chooses, which does involve recounting, is well supported for treating diagnosed post traumatic stress disorder. The narrow claim, and the only one made here: asking somebody to recount a distressing experience is not a neutral act, and you have eight processes that do it.
Your identification instrument and your control are usually two different purchases doing two different jobs. Here they are the same act, starting at the same moment.
Three or four fixed-choice questions, designed together against the hazard categories in the Code rather than picked off a menu, subject to acceptance and editing on both sides. They are placed inside an insight instrument that carries your organisation's branding, so it arrives as something your people were given rather than something an outside firm is running on them.
A survey takes something from the person filling it in and gives them nothing back. This one starts an internal dialogue on the way past. The questions are built to be answered honestly and privately, and the results that come back are drawn from that person's own answers, for them, on the day they gave them.
So identification and control are not two stages here. They are the same act. You get anonymous percentages for your file. They get the first movement of the thing they are about to be offered. Nobody has been asked what happened.
A structured problem-solving sequence in four phases, plus the Embodiment Map™ System as the daily practice that carries it into an ordinary working week. Personally directed, which means each person's own system selects what moves and at what pace, without supplying a single detail to anybody.
Access continues for as long as they want it. It does not expire when the cohort closes, and it does not expire when they leave you.
The Code requires you to review control measures and to record the process and its outcomes. Re-running the questions at an agreed interval gives you a second dated reading and shows movement or the lack of it, which is exactly what a review looks like on paper.
New starters, later cohorts and refresher cycles are quoted separately rather than assumed, so you are never paying for something you did not ask for.
What you hold at the end. A dated participation and completion record, anonymous percentages against questions you wrote, a second reading at review, and a workforce that was never once asked to explain itself to you. Numbers for the file. Nothing about a person.
Set out plainly, including where the answer is partial. A provider who claims all four outright is one you should test harder.
Can carry the whole limb, and how much of it depends entirely on the questions. The seeded questions return anonymous percentages from inside something each person is doing for their own benefit rather than for yours, which is why the answers come back straight. We advise on the question set before it is finalised, against the recognised psychosocial hazard categories, so the scope is decided deliberately rather than discovered afterwards. Ask a narrow set and you get a narrow read. Ask a well-built set and this reaches the hazards your survey has never once surfaced.
Gives you the ranking. You still own the assessment. Percentages against fixed-choice questions come back unweighted, which means the order is already visible: what proportion of your people chose each option, on each hazard you asked about. That is what tells you the priority actions. What we do not do is sign your risk assessment, apply your context, or replace your work health and safety adviser. The duty stays with you, and so does the judgement about your own workplace.
This is the limb it was built for. A structured programme delivered to a cohort, running daily rather than as a single event, designed to continue afterwards, and reaching the people who decline everything else because declining costs them nothing and taking part costs them nothing either. It is one control among the several you will need, and it sits at the human end rather than replacing job design, workload, rostering or management behaviour.
Directly delivers. A dated participation and completion record written for a file, plus whatever anonymous percentages you specified. It shows what was offered, to how many, over what period, and how many completed it. And it does that without creating a single document that could identify anybody, which means it adds to the evidence pile without adding to the disclosure pile.
The honest summary. Limb three and limb four are delivered directly. Limb one is delivered by design, and how much of it depends on how well the question set is built, which is why we advise on it. Limb two is directly informed, because unweighted percentages hand you the priority order. It does not by itself discharge a duty under work health and safety law, and nothing here is legal advice. We work alongside your work health and safety adviser rather than around them, and we would rather tell you that in writing than have you discover it later.
There is no point at which any participant is asked to name, describe, disclose or relive anything. That is a clinical design decision, and it is the whole control.
A workshop is an event. This is a daily practice with a fixed structure, run across the programme and designed to continue after it ends. The materials stay with each participant afterwards, with nothing further to pay for what they already have. Ongoing arrangements are available where you want them, for new starters, later cohorts or continuing support, and they are quoted separately rather than assumed.
This is what arrives at the end, and it is written for a file. Every line of it is a number. None of it is a name.
The floor is ten, and it sits in the contract rather than in a policy document. No percentage comes back unless at least ten people answered. At four respondents every answer lands on 25, 50, 75 or 100, and a manager who knows the team reads it straight off the page.
Point two is the one worth dwelling on in a procurement conversation, because it answers "identify" out of the same engagement that answers "control". You get honest answers to questions you chose yourself, because nobody answering them has anything to gain or lose by shading the truth. The questions sit inside something they are doing for themselves, and you never see the rest of it.
Five questions. Your answers are never sent to us and never saved anywhere. They stay on your own device and disappear when you close the page. That is the same principle the programme itself runs on.
Two points for each question you could answer yes to and evidence, one for a partial or undocumented yes, none for a no. It measures the paper trail a regulator would ask for, not how hard anybody has worked.
The method includes clinical hypnosis presented as hypnotic language, and we name it, in the proposal and in the participant consent, before anybody enrols. We lead with what it does rather than what it is called, which is ordinary in procurement, but we do not disguise it and an organisation should not want us to.
Anyone who discovers mid-programme that "mindset training" was clinical hypnosis has a legitimate complaint, and that complaint lands on the employer. What we say plainly is that nobody is asked to disclose, nobody loses control, nobody is made to do anything, and a participant can stop at any point.
It is worth saying what the state actually is, because the objection is almost always about the films rather than the method. It is a light state your people already enter several times a day: driving a familiar route and arriving without remembering the last twenty minutes, reading the same page three times over, the few minutes before sleep. Everyone is different and settles at a different level, from light through to deep, and each person reaches the level that is right for them at that time. Think of it as a power nap. Nobody is unconscious, nobody is under anybody’s control, and they can open their eyes, speak, move or stop at any point simply by deciding to.
What they already know is the state. What takes knowledge and skill is arriving there on purpose, and staying long enough for the work to be done, rather than surfacing out of it wondering where they have been.
Start with what the sessions actually contain. Everything a participant does unsupervised is down-regulating: coherent breathing, extended exhale, box breathing, the physiological sigh. Nothing that drives the breath hard for minutes at a time, and no breath held to the edge after over-breathing. The techniques that carry real contraindications are not in the programme at all. The risk is not managed. It is designed out, and that is checkable: ask us for the practice list.
The hypnotic component is a guided process much closer to a guided meditation than to anything in a film. Nobody is unconscious, nobody surrenders control, and anyone can stop at any point.
Every participant still completes a Health Declaration and informed consent before anything is released, and nothing is released until it is signed and returned. Not because the practices are risky, but because a hard gate catches the people a soft one misses. Those answers go to the practitioner and the employer never sees them.
You see numbers, never names. How many people it was offered to, how many started, how many completed, and the percentages against questions you wrote yourself.
The question we get asked here is whether you can at least see who did not start. The honest answer is that if you were to have the names it would cost you the thing you are buying. Somebody who declines is usually declining for exactly the reason the programme exists, and a follow-up from their own employer is the mark we removed. So we run the reminders instead. You get the uptake number, we chase the uptake, and nobody is named to you at any point.
That is a feature you are buying rather than a limitation you are accepting, and it is the reason the data you do get is worth having.
We do not publish outcome statistics, because we do not have primary evidence for them and will not invent any.
What we offer is a control with a completion record, optional anonymous data that you specify, and voluntary feedback from anonymous participants on their own experience. An organisation that wants published efficacy percentages should ask any provider offering them where the number came from.
The daily practice is designed to continue, and the materials stay with each participant afterwards with nothing further to pay for what they already have.
Ongoing arrangements are available where you want them. New starters joining after the cohort ran, a second or third cohort, refresher cycles, or continuing support on a term you set. They are quoted separately rather than bundled in by default, so you are never paying for something you did not ask for.
The bulk of the programme is recorded, so it does not depend on anybody's diary. Only the one to one work needs a person in the room, and that person is Graham or a Level 3 Master Practitioner trained and certified by Liminal Group.
We do not train your staff to deliver this, and that is deliberate rather than a missed upsell. Two reasons. It is Level 3 work, which is the top of a certification pathway rather than a workshop. And an internal deliverer can be leaned on, however unintentionally, to find out who said what, to add a name to a form, or to soften the content-free rule for one difficult case. The moment any of that happens the control stops working and the people it was built for stop taking part. Keeping delivery outside your organisation is what makes the promise to your people credible.
After that, either a pilot cohort or nothing, and both are acceptable outcomes. If it is not a fit we will both know inside ten minutes.
gc@liminalgroup.proFlat fee, never hourly. The cost is known before approval. One contract covers a cohort rather than an individual. Fifty per cent on commencement, fifty per cent on completion.
