Why More Documentation Is Not Always The Answer
Australian work health and safety law, among many things, requires persons conducting a business or undertaking to identify hazards, manage risks, and maintain records. In regulated industries, this obligation typically generates a substantial documentary footprint: policies, risk assessments, Safe Work Method Statements, inspection checklists, training registers, and incident reports. Yet in 2024, 188 workers were fatally injured at work across Australia, and 146,700 serious workers’ compensation claims were lodged in 2023–24 alone (Safe Work Australia, Key Work Health and Safety Statistics Australia 2025). Behind every one of those numbers is a person, a family, and, in many cases, a set of documents that was supposed to prevent the very thing that happened.
The conventional response to this observation is to argue that the documents were not good enough: that the risk assessment missed something, that the procedure was not detailed enough, that the Safe Work Method Statement was generic rather than site-specific. The remedy, according to this logic, is better documents. More documents. More detailed documents. More frequently reviewed documents.
This article seeks to challenge that assumption. The evidence from safety science, from recent Australian prosecutions, and from three decades of consulting practice across high-hazard industries suggests that the paper-to-practice gap is not primarily a documentation failure. It is, in many cases, a documentation consequence. The volume and complexity of safety paperwork has itself become a mechanism that drives the gap between what organisations describe and what organisations do.
The bureaucratisation of safety
In 2014, Professor Sidney Dekker of Griffith University described a phenomenon he termed the ‘bureaucratisation of safety’: the progressive expansion of documentation, procedural requirements, and compliance activities within safety management systems, to the point where the paperwork becomes an end in itself rather than a means of preventing harm (Dekker, S., ‘The bureaucratization of safety’, Safety Science, vol. 70, 2014, pp. 348–357).
Dekker’s observation was not that documentation is inherently bad. It was that the safety yield of further bureaucratisation is declining or plateauing in many industries, and that predictability, standardisation, and control carry secondary effects that are rarely examined. The suppression of surprise, diversity, and professional judgement that comes with ever-more-detailed procedural specification can itself create fragility. When workers are trained to follow documents rather than to understand and manage risk, the organisation becomes dependent on the document’s capacity to anticipate every operational scenario. No document can do this.
This argument has been developed with greater specificity by Rae, Provan, Weber, and Dekker, who in 2018 introduced the concept of ‘safety clutter’: the accumulation of safety procedures, documents, roles, and activities that are performed in the name of safety but do not contribute to the safety of operations (Rae, A.J., Provan, D.J., Weber, D.E. and Dekker, S.W.A., ‘Safety clutter: the accumulation and persistence of “safety” work that does not contribute to operational safety’, Policy and Practice in Health and Safety, vol. 16, no. 2, 2018, pp. 194–211). The authors identified three mechanisms by which clutter accumulates: duplication (the same risk managed through multiple overlapping systems), generalisation (activity designed for one context applied wholesale to another), and over-specification (procedures written at a level of detail that bears no relationship to the risk being controlled). The mechanisms are asymmetric. It is always easier to add a safety procedure than to remove one. Each incident, each audit finding, each regulatory change creates pressure to introduce new documentation. Almost nothing creates equivalent pressure to retire documentation that has outlived its purpose.
The consequences of clutter are not merely administrative. Rae and colleagues argued that clutter produces cynicism and ‘surface compliance’: the organisational behaviour where workers complete paperwork to demonstrate formal compliance without engaging with the underlying risk management intent. When a worker signs a pre-start checklist without reading it, or a supervisor acknowledges a SWMS without reviewing it against actual site conditions, the organisation has substituted a record of compliance for an act of risk management. The two are not the same thing.
Work as imagined, work as done
In 2012, David Borys published an ethnographic study of how managers and workers interpret and use Safe Work Method Statements on two commercial construction sites in Australia (Borys, D., ‘The role of safe work method statements in the Australian construction industry’, Safety Science, vol. 50, no. 2, 2012, pp. 210–220). His findings exposed three distinct gaps. The first was a compliance gap between the safe behaviours specified in the SWMS and the behaviours required of workers on site. The second was an adequacy gap between the scenarios described in the SWMS and the actual task demands encountered during construction. The third was a competence gap between the procedural knowledge encoded in the document and the practical expertise required to manage variable, unpredictable conditions.
The study’s most confronting finding was not that these gaps existed. It was that, in the minds of those on site, there were no unresolved gaps. Workers and supervisors had adapted their practices to bridge the distance between the document and the task. They used professional judgement, social interaction, and informal problem-solving to manage the situations that the SWMS could not anticipate. The formal system did not account for this adaptive capacity. In many cases, it could not even see it.
This distinction, between ‘work as imagined’ (the task as described in the management system) and ‘work as done’ (the task as actually performed by workers in context), has become central to contemporary safety science. It was articulated by Erik Hollnagel as part of the Safety-II framework and has been extensively developed by Dekker, Borys, and others working in resilience engineering. The paper-to-practice gap, from this perspective, is not an aberration. It is an inevitable feature of any system that attempts to govern complex, variable, sociotechnical operations through pre-specified written procedures. The question is not whether the gap will exist. The question is whether the organisation can see it, understand it, and work with it constructively.
A case study in systemic failure
The consequences of failing to see this gap were demonstrated with tragic clarity in March 2026, when Mastermyne Crinum Operations Pty Ltd was found guilty of industrial manslaughter in the District Court of Emerald, Queensland, following the death of 62-year-old underground miner Graham Dawson in September 2021. A jury returned a guilty verdict by a margin of 11 to 1. On 1 May 2026, the company was fined $7 million with $300,000 in court costs and a recorded conviction.
The sentencing judge observed that the death was avoidable and that there had been an alarming lack of consultation about the company’s decision to change its methods, along with clear warning signs about the risk. This was Queensland’s first successful industrial manslaughter prosecution in the mining sector since the offence was extended to the resources industry in 2020.
Mastermyne was likely not an organisation without a range of well thought out documents for managing its safety. It was an organisation where the relationship between the documents and the operational reality had broken down across at least six dimensions: the approved strata support method was reportedly abandoned without a formal management of change process; geotechnical specialists were not informed of changed underground conditions; precursor events indicating instability were observed but not escalated; the documented system described one method while the workforce executed another; specialist expertise was engaged episodically rather than embedded in ongoing operations; and the specific risks of recommissioning a decommissioned tunnel through unstable geological formations were not re-assessed. The paperwork existed. The connection between the paperwork and the people it was supposed to protect clearly did not.
You might like to read What makes great WHS Policy & Procedures
The compliance trap
If the argument ended here, the implication might be that documentation is an obstacle and that organisations should abandon formal safety management systems in favour of worker autonomy and adaptive expertise. This is not the argument.
The WHS Act 2011 (applying in Queensland, New South Wales, and Commonwealth jurisdictions) requires persons conducting a business or undertaking to ensure, so far as is reasonably practicable, the health and safety of workers. The WHS Regulations prescribe specific documentation requirements for high-risk construction work, hazardous chemicals, plant, confined spaces, and (increasingly) psychosocial hazards. Codes of practice, particularly in New South Wales following the 2025 amendments, now function as enforceable minimum standards rather than merely persuasive guidance. An organisation that abandons documentation abandons its capacity to demonstrate due diligence.
The challenge is more subtle than ‘documents good’ or ‘documents bad’. The challenge is what Dekker, in his 2025 work Safety Theater, described as the tendency for organisations to pursue procedural perfection while losing authentic contact with the reality of work as it is actually performed. The desire for a perfectly documented safety system, where every risk is assessed, every control is specified, every worker has signed every acknowledgement, produces a kind of institutional performance. The system performs safety to itself and to its regulators without necessarily producing it at the operational level.
This is not an abstract concern. Safe Work Australia data shows that the four leading mechanisms behind serious workers’ compensation claims (body stressing, falls, being hit by moving objects, and mental stress) account for 84% of all serious claims. Vehicle incidents and falls from height remain the dominant fatality mechanisms, year after year. These are not novel hazards. They are hazards for which extensive documentation already exists in one form or another across many organisations. The persistence of harm despite the presence of procedures is not evidence that the procedures are inadequate. It may be evidence that procedural adequacy is the wrong question.
Toward a different relationship with documentation
The practitioner’s task, then, is not to choose between documentation and practice but to change the relationship between them. This requires several important shifts in organisational thinking.
The first is to treat documents as hypotheses, not instructions. A SWMS, a risk assessment, or a procedure represents the organisation’s best current prediction of how a task will unfold and what controls will be effective. Like any prediction, it may be right or wrong. The document’s value lies not in its existence but in the discipline of testing it against observed conditions, revising it when conditions change, and trusting workers to exercise professional judgement when reality diverges from the written plan. Borys found that the construction workers in his study were already doing this. The formal system had simply not been designed to recognise or support it.
The second is to apply the same critical thinking to safety activities that organisations apply to operational activities. Every procedure, every form, every checklist, every sign-off should be subject to the question: does this contribute to the safety of operations, or does it contribute to the appearance of safety? Rae and colleagues suggested that organisations rarely ask this question because the mechanisms that generate clutter are self-reinforcing. Incidents produce new procedures. Audits identify ‘gaps’ that are closed by adding new documentation. Regulators issue improvement notices that require documented responses. Each addition is individually rational. The cumulative effect is a system where the volume of safety work expands while the connection to operational safety erodes.
The third shift is to invest in the conditions that make adaptive, competent risk management possible rather than investing exclusively in the documents that describe it. This means genuine consultation under s 47 of the WHS Act, not as a box-tick but as a source of intelligence about how work is actually performed. It means training that develops judgement and hazard recognition, not merely familiarity with procedures. It means leadership that values honest reporting of deviations, near misses, and operational surprises over clean audit scores and low injury rates.
The fourth, and perhaps most difficult, is to resist the organisational reflex to respond to every adverse event with additional documentation. When something goes wrong, the instinct to write a new procedure is powerful. It feels productive. It creates an artefact that demonstrates action. But if the problem was not a gap in documentation, the new procedure does not address the problem. It addresses the anxiety produced by the problem. This distinction matters because, as the safety clutter literature makes clear, each new procedure carries an opportunity cost: the time and cognitive load required to comply with it is time and cognitive load not available for the situational awareness and professional judgement that might have prevented the next incident.
Where does this leave the practitioner?
None of this clearly absolves organisations of their legislative obligations. A PCBU must still identify hazards, assess risks, implement controls, and maintain records. A principal contractor must still ensure SWMS are prepared for high-risk construction work. An officer must still exercise due diligence. The regulatory framework is clear, it is tightening across multiple jurisdictions, and the Mastermyne prosecution demonstrates that enforcement carries real consequences.
But the practitioner who understands the bureaucratic drift literature is better equipped to build safety management systems that function rather than merely exist. The practitioner can distinguish between documentation that serves a genuine control function (a permit-to-work system that requires active verification of atmospheric conditions before confined space entry, for example) and documentation that serves a compliance function without contributing to operational safety (a generic pre-start checklist signed in the crib room before workers have seen the site conditions for the day). The practitioner can advocate for fewer, better, more operationally connected documents rather than more, longer, more procedurally exhaustive ones.
Patrick Hudson’s Safety Culture Maturity Model describes a five-stage progression from pathological to generative safety culture. The organisations that close the paper-to-practice gap most effectively are not those with the largest document management systems. They are those that have progressed beyond the calculative stage, where data and documentation are abundant but remain a management concept, to the proactive and generative stages, where safety is owned by the people doing the work and information flows horizontally as well as vertically. Documents support this culture. They do not substitute for it.
The gap between paper and practice will never be fully closed. Complex work in variable environments will always produce conditions that written procedures cannot fully and wholly anticipate.
The measure of a mature safety management system is not the absence of that gap, but the organisation’s capacity to see it, discuss it honestly, and respond to it in real time. If the response to a gap between paper and practice is always to produce more paper, the gap will only widen.
References
Borys, D. (2012). The role of safe work method statements in the Australian construction industry. Safety Science, 50(2), 210–220. https://doi.org/10.1016/j.ssci.2011.08.010
Dekker, S.W.A. (2014). The bureaucratization of safety. Safety Science, 70, 348–357. https://doi.org/10.1016/j.ssci.2014.07.015
Dekker, S.W.A. (2025). Safety Theater: How the Desire for Perfection Drives Compliance Clutter, Inauthenticity, and Accidents. Routledge.
Hudson, P. (2007). Implementing a safety culture in a major multi-national. Safety Science, 45(6), 697–722. https://doi.org/10.1016/j.ssci.2007.04.005
Rae, A.J., Provan, D.J., Weber, D.E. and Dekker, S.W.A. (2018). Safety clutter: the accumulation and persistence of ‘safety’ work that does not contribute to operational safety. Policy and Practice in Health and Safety, 16(2), 194–211. https://doi.org/10.1080/14773996.2018.1491147
Safe Work Australia (2025). Key Work Health and Safety Statistics Australia 2025. https://data.safeworkaustralia.gov.au
Safetysure is an ISO 9001/45001/14001 accredited workplace health, safety, and occupational hygiene consulting firm. We work across Queensland, New South Wales, Victoria, Western Australia, the ACT, and Commonwealth jurisdictions. For more information, contact us at safetysure.com.au.