September 2026 delivered a stark reminder to the construction industry that personal accountability for health and safety failures is not simply a theoretical risk.
Within six days, two separate excavation cases resulted in individuals receiving custodial sentences, one suspended and one immediate. One also resulted in a ten-year director disqualification.
The circumstances were serious, and the outcomes should not be interpreted as meaning that every health and safety breach will lead to somebody going to prison.
But they do underline something that every construction company owner, director, contracts manager and senior leader should understand:
Health and safety responsibility is judged by what you knew, what you controlled and what you did and not simply by whether a risk assessment, method statement or construction phase plan existed.
Excavation work is a particularly alarming example because when the controls fail, there is often very little margin for error.
A trench can collapse without warning. Ground that appeared stable yesterday may not be stable today. Nearby structures can be undermined. Plant, spoil heaps, water and changing ground conditions can all affect stability.
And where people are working below ground level, the consequences of getting those decisions wrong can be catastrophic.
Two cases, six days apart, and one clear warning
On 9 September 2026, two companies and a self-employed contracts manager were sentenced following the death of Gheorghita “Geo” Arsene, aged 33, at a site in Banstead, Surrey.
Geo was replacing a damaged pipe at the bottom of a deep trench when the unsupported sides collapsed. Around two tonnes of soil buried him.
The investigation found that there were no supports in the trench and that excavated soil had been deposited immediately beside it, increasing the loading on the excavation sides.
Anthony O’Connor, a self-employed contracts manager working for T Vaughan Limited, had seen another worker inside the same unsupported excavation earlier that day.
He knew measures were required to prevent a collapse but failed to ensure they were put in place.
He pleaded guilty to breaching Section 7 of the Health and Safety at Work etc. Act 1974 and received ten months’ imprisonment, suspended for 18 months, together with 150 hours of unpaid work.
T Vaughan Limited was fined £650,000 and ordered to pay £40,000 in costs. Principal contractor Appledorn Developments Limited was fined £500,000.
Six days later, Peter Clifford was sentenced at Ipswich Crown Court following the death of 30-year-old Dale Baker during renovation work in Woodbridge.
A trench had been excavated dangerously close to a conservatory, undermining its foundations. The trench was inadequately supported and subsequently collapsed, followed by the conservatory foundations.
Clifford received eight years and ten months’ imprisonment for gross negligence manslaughter and was disqualified from acting as a company director for ten years.
These were serious cases with tragic consequences.
But there is a very practical lesson running through them:
Excavations do not become dangerous at the moment they collapse.
The conditions that allow the collapse to occur usually exist beforehand.
That is where planning, competent supervision and intervention matter.
Excavation safety starts before anybody enters the ground
HSE guidance is clear that excavation precautions need to be considered before digging starts.
That means thinking about more than simply how deep the excavation will be.
The ground conditions matter.
So does groundwater.
Nearby buildings and structures matter.
So do underground services, vehicle movements, plant, materials, access and the effect of placing excavated spoil close to the edge.
Before an excavation is opened, somebody competent should have considered what measures will be required to prevent collapse and whether support, a proprietary trench system or suitable battering will be necessary.
The equipment required to make the excavation safe should also be available before workers are exposed to the risk.
HSE guidance specifically warns against working ahead of excavation support or removing it prematurely. It also expects work involving excavation support to be directed by competent supervision.
That sounds obvious.
Yet the prosecution cases show why the obvious still needs managing.
“It looked stable” is not an excavation control
One of the dangerous assumptions around excavation work is that ground which appears solid will remain that way.
It may not.
Ground conditions vary enormously and can be affected by weather, water, vibration, adjacent loads, previous disturbance and the excavation itself.
The weight involved is also easy to underestimate.
When a trench collapses, the person inside is not simply dealing with loose soil around their legs. They can be struck, buried or crushed by tonnes of material within seconds.
That is why unsupported excavation should never rely on somebody looking at the side of a trench and deciding that it “seems fine”.
Where support is required, it needs to be properly considered, designed where necessary, installed and maintained.
Where battering is being used to control collapse, the angle has to be suitable for the ground conditions.
And where the excavation could affect nearby walls, foundations or structures, that interaction must be assessed before the work progresses.
Temporary works are not an administrative exercise.
They are often the thing preventing a structure — or the ground itself — from failing.
Spoil heaps, plant and loading at the trench edge
The Banstead case also highlights another basic excavation principle.
What happens around the excavation can affect what happens inside it.
Excavated material piled beside a trench adds loading to the ground.
So can vehicles, excavators and other plant.
Construction teams therefore need to consider not simply whether the excavation has been supported, but what additional loads are being introduced close to its edge.
This is exactly why excavation safety cannot be reduced to a generic RAMS document prepared before a job starts.
Conditions change.
Spoil moves.
Plant moves.
Weather changes.
The excavation gets deeper.
Different trades arrive.
The work progresses.
Effective site management means continuing to ask whether the controls remain suitable for the work that is actually taking place.
Inspections are not optional paperwork
The Construction (Design and Management) Regulations 2015 require excavations to be inspected by a competent person at the start of the shift in which work is to be carried out, after events likely to have affected stability and after material unintentionally falls or becomes dislodged.
Work should only proceed where the competent person is satisfied it can be carried out safely.
And where an inspection identifies something that means work cannot safely continue, the problem needs to be satisfactorily remedied before work restarts.
That distinction matters.
An excavation inspection is not simply a form somebody completes because there is a box in the site file marked “inspection”.
Its purpose is to make a decision:
Is this excavation safe for people to work in now?
If the answer is no, or nobody competent can confidently answer yes, nobody should be entering it.
The person who sees the problem matters too
The Banstead prosecution contains another important lesson for construction managers and supervisors.
Anthony O’Connor had seen somebody working in the unsupported excavation earlier in the day.
He knew controls were required.
But they were not implemented before Geo subsequently entered the trench.
That is different from a hazard nobody had spotted.
Once an unsafe condition has been identified, there is an opportunity to intervene.
That might mean stopping the work.
It might mean getting the trench support installed.
It might mean bringing in somebody with the necessary temporary works expertise.
It might mean reviewing the method.
It might simply mean saying:
Nobody goes back into that excavation until we know it is safe.
What it cannot mean is noticing a significant risk and carrying on regardless.
In construction, people with supervisory or management responsibilities need to understand that their role is not limited to observing work.
They are there to influence what happens next.
Another director sentence just months earlier
The September cases did not occur in isolation.
In June 2026, HSE reported the sentencing of James Harper, sole director of Cavendish Basements Limited, after a worker suffered life-changing injuries when the side of a three-metre-deep excavation collapsed.
The worker had been levelling the ground. It had been raining, and excavated material had been piled beside the excavation.
HSE subsequently found failures in the planning, management and monitoring of the work, including issues relating to excavation stability, work at height and supervision.
The incident had also not been reported to HSE when it should have been.
Harper pleaded guilty to an offence under Section 37 of the Health and Safety at Work etc. Act 1974 in connection with a breach of CDM 2015.
He received a six-month custodial sentence, suspended for two years.
Three excavation cases in a matter of months should make the construction industry take notice.
Not because directors need to stand beside every trench.
But because serious construction risks must be properly planned, managed, supervised and monitored.
Where does temporary works management fit?
For many excavation jobs, particularly complex or deep excavations or work affecting neighbouring structures, temporary works management becomes critical.
Shoring, propping and excavation support need to be suitable for the conditions they are intended to control.
That may require input from an appropriately experienced temporary works designer or engineer.
The design then needs to translate into what actually happens on site.
That means correct installation, inspection, maintenance and, importantly, controlled alteration or removal.
A perfectly good temporary works design provides little protection if the installation differs from it or somebody removes support to make the job easier.
HSE guidance on temporary structural support reinforces the need for competent design and for temporary supports to be installed, checked, inspected, maintained and dismantled in accordance with the design.
For contractors, the practical question should therefore not simply be:
“Have we got a temporary works design?”
It should be:
“Is the excavation on site actually constructed and being managed in accordance with it?”
That is a very different level of assurance.
Who is actually controlling the excavation?
Construction projects frequently involve several organisations.
There may be a client, principal contractor, groundwork subcontractor, specialist temporary works provider, engineer, contracts manager, site manager and supervisors.
That creates obvious potential for assumptions.
The groundworker thinks the principal contractor is dealing with temporary works.
The principal contractor assumes the subcontractor has designed the support.
The site manager assumes the contracts manager has checked it.
The director assumes the site team has everything under control.
And meanwhile somebody climbs into the excavation.
This is why responsibilities need to be explicit.
Everyone involved should understand who is making decisions about excavation stability, who is inspecting it, who can authorise changes and — equally importantly — who has the authority to stop the work.
Where responsibilities become blurred, risk tends to sit in the gaps.
A practical excavation check
Before somebody enters an excavation, a construction business should be able to demonstrate that it has considered the following:
- The ground conditions and potential for collapse;
- The depth, shape and nature of the excavation and whether shoring, trench support or battering is required;
- Nearby buildings, foundations, walls and other structures that could be undermined;
- Underground and overhead services;
- Groundwater, surface water and changing weather conditions;
- Plant, vehicles, spoil and other loads close to the excavation edge;
- Safe access, edge protection and the risk of people, plant or materials falling in;
- The competence of the people planning, supervising and inspecting the work;
- Any required temporary works design and whether what is installed on site matches it;
- Inspection at the required stages and evidence that defects are being acted upon; and
- Clear authority for somebody to stop the work if conditions become unsafe.
That is construction control.
The paperwork should support it — not substitute for it.
What would HSE find after a collapse?
This is where the technical and leadership issues come together.
Following a serious excavation incident, investigators are unlikely to be interested only in whether there was a risk assessment somewhere in the project file.
They will want to reconstruct how the work was planned and managed.
What information was available about the ground?
Who decided how the excavation would be supported?
Was temporary works input required?
Who supervised the work?
Who inspected the excavation?
Were those people competent?
Was spoil being placed too close to the edge?
Were weather or changing ground conditions taken into account?
Did the excavation undermine another structure?
Were workers going into unsupported ground?
Had anybody raised concerns?
Had anybody seen unsafe practices previously?
If they had, what did they do?
And why was work allowed to continue?
That is why good construction safety management has to leave an evidence trail.
Not because paperwork prevents accidents, but because good records often reflect good decisions.
Section 37 and personal accountability
Section 37 of the Health and Safety at Work etc. Act 1974 is particularly relevant to directors and senior managers.
Where a company commits a health and safety offence and that offence occurred with the consent or connivance of, or was attributable to neglect by, a director, manager, secretary or similar officer, the individual may also face prosecution.
That does not mean every director becomes personally liable whenever an employee or subcontractor makes a mistake.
Nor does it mean directors must possess the technical knowledge of an excavation engineer or temporary works designer.
It does mean leadership cannot simply disappear behind an organisational chart.
You can delegate technical tasks.
You can appoint a competent site manager.
You can use subcontractors.
You can appoint temporary works specialists.
You can bring in a health and safety adviser.
But you cannot delegate leadership oversight.
Construction businesses still need suitable arrangements, competent people, adequate resources, meaningful supervision and a way of knowing whether significant risks are being controlled.
A RAMS pack is not the same as control
Construction has become very good at producing documents.
Construction phase plans.
Risk assessments.
Method statements.
Permits.
Inspection sheets.
Temporary works registers.
Toolbox talks.
All of those can serve an important purpose.
But a file full of immaculate documents will not hold up an unsupported trench wall.
This is where businesses sometimes confuse documentation with assurance.
If your RAMS say the excavation will be supported but workers are entering it unsupported, you do not have control.
If your inspection sheet says the trench has been checked but nobody competent looked at it, you do not have control.
If your temporary works design specifies a particular support arrangement but site conditions are different, you do not have control.
If a contracts manager sees unsafe excavation work and does nothing about it, you do not have control.
The document and the work have to tell the same story.
The question construction leaders should be asking
The lesson from these cases is not that directors should become excavation specialists.
It is that serious construction risks need a chain of competent decision-making that actually works.
A director should be able to ask:
Who is controlling this risk?
Are they competent to do it?
What are the critical controls?
How do we know they are in place?
What happens when they are not?
And for higher-risk work, somebody should be able to answer those questions clearly.
That is what effective health and safety governance looks like on a construction site.
Not a board sitting in an office discussing policy in the abstract.
It is leadership creating the conditions in which the right technical decisions are made on site — and unsafe work is stopped before somebody gets hurt.
The final question
If HSE arrived on one of your projects tomorrow and walked towards an excavation, what would they see?
Would the support arrangements match the method and temporary works design?
Would spoil and plant be appropriately controlled?
Would the excavation have been competently inspected?
Would the site team understand who is responsible for it?
Would a supervisor challenge somebody entering an unsafe trench?
Would your managers know they have the authority to stop the job?
And if HSE then followed the evidence back through your organisation, would it show that the risks had been properly planned, managed and monitored?
Or would it show that everybody assumed somebody else had it covered?
Those are very different positions to defend after an accident.
At accuSafe, we work with construction businesses to make sure health and safety arrangements do more than exist on paper.
That includes risk management, contractor and subcontractor arrangements, audits, management systems, director and manager training and ongoing competent-person support.
The aim is straightforward:
To make sure the controls described in your documents are the controls people actually see on site.
Because in excavation work, finding the gap after the ground has collapsed is already too late.
