There’s a pattern playing out across safety social media. Someone posts a punchy line.
“Paper won’t save you.”
“Procedures don’t prevent incidents.”
“Compliance isn’t safety.”
It sounds tough. It sounds practical. It sounds like someone cutting through nonsense. The likes roll in. The algorithm rewards it. The nuance disappears.
And somewhere down the line, someone makes a real decision based on a slogan they half-remember from LinkedIn.
This isn’t about one person or one post. It’s about what happens when legitimate concerns get abstracted into ideas that sound clever but quietly undermine how safety actually works.
The problem that really exists
Let’s start with the part that’s true.
- Safety documentation often becomes disconnected from purpose.
- Tick-box permits.
- Generic safe system off work or method statement copied and pasted between jobs.
- Traffic-light dashboards that tell you nothing about whether controls are actually working.
A 47-page safe system off work or method statement for building a wall is not unusual. It’s also useless.
That kind of paperwork deserves criticism. It adds friction without control. It becomes compliance theatre. People sign it, file it, and work around it.
Calling that out is fair. Necessary, even.
But that is not the same thing as saying documented systems of work are the problem.
Where the abstraction starts
Social media doesn’t reward careful distinctions. It rewards slogans.
- A nuanced argument about bad documentation design gets flattened into “paperwork is bad”.
- A critique of compliance theatre becomes “procedures don’t work”.
- A legitimate frustration with bloated systems turns into a blanket dismissal of documented safe systems of work.
That’s the abstraction.
Once that happens, the message starts travelling without its context. And that’s where it becomes dangerous.
What gets lost when everything becomes “paper”
The system disappears
Physical controls do not exist in isolation.
An isolation point, a barrier, a guard, or rescue equipment only works properly when there is a system governing it.
That system answers basic questions.
- Which control is required for this task?
- Who authorises it?
- How is it verified?
- What happens at handover?
- What changes if conditions change?
That is not bureaucracy. That is control.
A permit to work is not paper in opposition to a physical control. It is the mechanism that makes that control reliable.
Remove the system and the hardware becomes assumption-based.
The evidence disappears
Fatal investigations tell a remarkably consistent story.
- The hardware was present.
- The barrier existed.
- The isolation point was there.
What failed was identification, authorisation, verification, handover, or change control.
Piper Alpha did not kill 167 men because isolation valves did not exist. It killed them because the state of a permit was not properly communicated at shift handover.
Those failures live in the system of work, not in the steel.
The law disappears
In many jurisdictions , high-risk construction work requires documented safe systems of work. In the UK, documentation is how duty holders demonstrate that risks were identified and controlled so far as is reasonably practicable.
The law does not create a hierarchy where physical controls trump documented systems. It expects them to function together. When something goes wrong, regulators do not ask “was there a barrier?”
They ask “how was it specified, authorised, verified, monitored, and managed when conditions changed?”
If you cannot show that, the assumption is simple. It was not controlled.
Control verses control systems
This distinction matters and is often missed.
A physical control is not the same thing as a control system.
- A guard can exist without being under control.
- An isolation point can exist without being correctly identified or verified.
- Rescue equipment can exist without anyone being able to use it in time.
A control system is what determines whether controls are the right ones, in place at the right time, working as intended, and withdrawn safely.
Most serious incidents occur not because controls were absent, but because the system governing them failed.
The real mistake behind “paper won’t save you”
What usually sits behind these posts is experience with bad paperwork.
- Bloated permit to work systems
- Unreadable safe systems of work and method statements.
- Forms that exist to cover backsides rather than manage risk.
The mistake is concluding that the concept is the problem, rather than the execution.
Bad documentation is not proof that documentation doesn’t work. It is proof that the document has failed as a control system.
The answer to dysfunctional paperwork is not no paperwork. It is paperwork that actually does its job.
- Clear.
- Task-specific.
- Used.
- Verified.
- Owned.
That takes more competence, not less.
Why this matters now
There is a commercial appeal to the “ditch the paperwork” message.
“Focus on critical controls, not paperwork” sounds empowering. It promises simplicity. It fits neatly on a slide.
The uncomfortable truth doesn’t sell as well. Documented systems of work are how critical controls are implemented, verified, and kept reliable over time.
Strip that nuance away and you create permission. Permission to quietly downgrade permit systems. Permission to treat SWMS as optional overhead. Permission to assume that hardware will somehow manage itself.
That looks progressive right up until something goes wrong and the investigation asks why the system of work wasn’t followed. Or worse, why it no longer existed.
What actually stands up
Physical controls matter. Nobody serious disputes that.
Those controls need systems to function reliably. That is not theory. It is what investigations, regulators, and courts repeatedly demonstrate.
Bad documentation should be challenged. Ruthlessly.
- Name the document.
- Explain what it fails to do.
- Redesign it so it governs the controls it is supposed to govern.
Do not abstract the problem into a slogan that dismisses the very mechanisms that stop people being killed.
Controls fail when the system governing them fails. That isn’t a trope. It’s how safety actually breaks.
Author – Phil Douglas is a Chartered Safety and Health Professional with more than three decades of frontline and leadership experience in high-risk industries. He is the founder of Oracle Safety Associates Ltd. He writes critically on safety theory, regulation, and accountability, with a particular focus on the gap between language-led interventions and controls that actually prevent harm.
