80% of Serious Injuries and Fatalities (SIFs) occur due to poor contractor management

Executive message

Contractors are not peripheral to industrial risk. In IOGP member reporting for 2024, contractor employees accounted for 26 of 32 fatalities and 726 of 946 lost-work-day cases. Contractor Fatal Accident Rate was 0.84 per 100 million hours worked, compared with 0.57 for company employees. A separate HFACS-OGI study of 184 oil and gas accident cases found contractors involved in 86% of analysed accidents. In downstream and chemical industries, Concawe’s 2024 European data shows all reported fatalities were contractors, and contractors represented 55% of manufacturing lost-workday injuries.

Additionally, contractor exposure is a recurring high-risk pattern across other high-hazard industries. In US construction, contracted workers accounted for 52% of fatal injuries and 62.7% of fatal falls. In mining, contractor deaths accounted for 41% of US mine deaths in 2019 and 28% in 2020, while ICMM’s global mining/metals data reported contractors as 45% of fatalities in 2024. In steel, worldsteel reports contractor fatality frequency rates above employee rates.

These figures should not be read as a contractor-blame narrative. Contractor involvement is not the same as contractor fault. The stronger conclusion is more useful: the operator-contractor interface is where a disproportionate amount of operational risk is being created, transferred, diluted or controlled.

Most contractor HSE management systems still carry the architecture of procurement compliance: pre-qualification forms, document checks, annual audits and lagging-indicator dashboards. Those controls are necessary, but they are not sufficient for high-hazard work. They often describe the contractor before the work begins; they do not reliably verify whether the contractor is ready for the specific risk profile of the work being awarded.

The practical question for leaders is therefore simple: can the organisation identify, before mobilisation, which contractors carry the highest risk, what capability gaps remain open, who owns those gaps, and whether the controls are working in the field? If the answer requires a pause, the contractor HSE system is producing documentation rather than control.

1. What the evidence is really telling us

The most important pattern in the data is not that contractors appear in incident statistics. That is expected in industries where specialist contractors execute a large share of maintenance, turnaround, construction, shutdown, logistics and high-risk project activity. The important pattern is that many organisations still manage this exposure with controls designed for commercial qualification rather than operational risk governance.

2. Why conventional contractor HSE systems underperform

Contractor HSE programs rarely fail because leaders do not care about safety. They fail because the system is built around the wrong unit of control. It asks whether the contractor can pass a generic qualification process, when it should ask whether the contractor can safely execute a specific scope under specific site, schedule, interface and competency conditions.

Across high-hazard program, five structural failures recur. They are predictable; therefore, they are designable.

3. What a redesigned system looks like

A contractor HSE management system that controls operational risk needs three sequential layers. The order matters. Organisations often jump to dashboards, forums or audit programs before they have risk classification, ownership and verification discipline in place. That creates activity without control.

The redesigned model changes the operating question. Instead of asking whether the contractor has been approved, the organisation asks whether the contractor is verified as ready for the risk profile of the work and whether that readiness is being maintained during execution.

4. Staged contractor HSE management cycle

The three-layer design logic becomes operational through a defined contractor management cycle. In the case implementation, this cycle was aligned with IOGP Report 423 and structured around twelve control points that carry contractor HSE from initial ownership and risk definition through to mobilisation and site verification.

The point of the cycle is not to add twelve administrative steps. It is to prevent contractor risk from moving through the contract lifecycle without a named owner, a risk-based pathway, verified capability, and field-level assurance. Each stage answers a specific control question and creates evidence that the next stage can rely on.

Figure 1. The 12-stage contractor HSE management cycle implemented aligned with IOGP Report 423, showing control points across the pre-award and post-award phases.

The cycle at a glance

5. The shift from static assessment to dynamic control

Traditional contractor assessments often create a false sense of control. A desktop review, annual audit, pre-start document submission or occasional site visit may confirm that a contractor is compliant at a specific point in time. However, contractor risk does not remain static. Scope changes, supervisors rotate, schedules become compressed, equipment condition changes, and field behaviors evolve throughout the contract lifecycle.

This is why contractor management must move beyond static assessment and become a dynamic control system.

The real weakness in many contractor management systems is not the absence of procedures, registers or scorecards. These usually exist. The weakness is the disconnect between data and decisions. Information is collected but not always converted into timely action. KPIs are tracked, but not always used to challenge performance. Audits are completed, but findings do not always influence mobilisation, supervision intensity or leadership intervention.

A dynamic contractor assessment system closes this gap by continuously connecting contractor data with operational decision-making. It gives Contract Holders, HSE teams and management a live view of contractor performance, not only during pre-qualification, but across mobilisation, execution, assurance and close-out.

Instead of asking, “Has the contractor passed the assessment?”, the better question becomes:

Is this contractor’s risk profile improving, deteriorating or remaining stable — and what decision should we take now?

In this model, contractor performance is assessed through multiple indicators that reflect both compliance and real field behaviour:

The purpose of dynamic assessment is not to collect more data. It is to collect better data and use it to support better decisions. Every indicator should help leaders understand contractor capability, risk exposure, behavioural trends, action discipline and readiness to perform work safely.

This shift changes the role of contractor management. It moves the process from administrative compliance to proactive risk governance. High-risk contractors receive deeper analysis, stronger monitoring and mandatory field verification. Lower-risk engagements are managed with proportionate oversight. Performance conversations become evidence-based, and escalation decisions become clearer.

The broader lesson is simple: static assessments document compliance; dynamic contractor control improves safety performance. In high-risk operations, the question is no longer whether the contractor management process exists. The real question is whether it is helping the organization see risk early, act faster and make safer operational decisions.

6. What this means for senior leaders

7. Cases in point: the model under live operating pressure

The three cases below apply the same contractor HSE architecture under different operating conditions: a high-density mega-turnaround, a portfolio-wide contractor improvement program, and a digital/AI enablement layer. They are deliberately presented as separate cases, but the thread is the same: risk-based classification, named ownership, verified readiness, field assurance and visible performance management.

7.1 How to manage contractors in a mega-Turnaround?

In a highly challenging mega-turnaround, zero Serious Injuries and Fatalities (SIFs) achieved:

The challenge

A mega-turnaround is not a larger version of an ordinary turnaround. It is a different problem. At the densities involved here — millions of man-hours compressed into weeks, thousands of people moving through site, more than a thousand permits issued a day at peak — the clients’ HSE team cannot be everywhere it needs to be.

3.2 million man-hours were executed across 52 days. Six to seven thousand people were onboarded and trained inside a compressed window. Sixty-plus contractors and vendors operated simultaneously, fourteen of them carrying out high-risk activities — and at peak, the permit-to-work system was issuing around 1,300 authorizations a day, on its way to a total of roughly 30,000 over the event.

A permit-to-work system operating at that volume stops being a control in the engineering sense. It becomes a queue. Supervisors who should be deciding whether a job is safe end up signing forms; toolbox talks risk collapsing into procedural routine; and Job Hazard Analyses (JHAs), written to surface specific hazards, begin to read like compliance paperwork rather than thinking.

The arithmetic of the workforce made one conclusion inescapable. Fourteen high-risk contractors covered approximately 85% of the on-site workforce, and 84% of all additional personnel mobilised for the event would be working under one of those fourteen banners.

The approach

The response rested on three principles — reduce complexity, sharpen accountability, and move risk control to the frontline — delivered through five practical moves: a front-loaded capability assessment of each of the fourteen high-risk contractors; specific gap-closure plans with named ownership across 42 audits, 1,722 audit questions and 290 owned actions; a three-stage rolling verification (pre-mob, mob, site); a senior engagement loop with joint HSE walkabouts; and frontline risk control through permit simplification, Last-Minute Risk Assessment-anchored toolbox talks, and Life-Saving Rules deployed with proportionate consequence management. The methodology behind these moves follows the framework set out above. What follows is what that approach produced — and where the value landed.

What it produced

The event delivered zero Serious Injuries or Fatalities across the full 52-day window — across 3.2 million man-hours, 6,000–7,000 personnel, and fourteen high-risk contractors operating concurrently. The figure is more meaningful read against the load it withstood than as a number in isolation.

Readiness managed at the front end, not under pressure. Contractors closed on average 72% of their pre-mobilisation and mobilisation requirements before turnaround start. The value of that figure lies in what it prevented: the gaps that did remain were small, known and assigned to named owners — not discovered under execution pressure, when the cost of correction multiplies.

Accountability that moved. Across the fourteen contractors, 290 owned gap-closure actions replaced what had previously been a reactive oversight cycle. The shift was not procedural. Audits had functioned as evidence of attention; the named-ownership model converted them into evidence of accountability — and follow-through moved from the client’s HSE team to the firms whose work had created the exposure. That change is the durable component of the result.

A structured mechanism. The combination of front-loaded assessment, staged verification, owned gap closure, simplified permits and visible senior engagement produced a system that was load-bearing rather than fortunate. The next mega-turnaround does not start from scratch: the contractor management model is now a repeatable asset to be recalibrated each cycle, not rebuilt.

7.2 Zero red-banded contractors: how our client reached the target and sustained it

The challenge: an inherited contractor population

In May 2024, contractor HSE management across the client was reactive, inconsistent and unstandardised. Pre-qualification, mobilisation, capability assessment and verification varied by site and, in some cases, did not exist at all. The system was driven by legal and regulatory minimums; the client did not have a consolidated view of contractor HSE performance.

The approach: build the architecture before expecting the outcome

Between May 2024 and December 2025, the client together with SnSD built the architecture the dashboard now reflects. Every contractor in scope was evaluated against a risk-based model. 78 Capability Assessments were completed during pre-award; 102 verifications replaced the previously variable, site-by-site entry practices; 28 contractor audits generated structured action items. 14 Resource Sheets translated risk-based contract HSE clauses into instructions tied to specific contract scopes.

The infrastructure that scales the model was embedded alongside the activity. Contract Holders inside the client organisation were trained and assigned for Job Authority, putting contractor HSE leadership inside the client’s own organisation. A set of contractor-HSE KPIs now feeds Business Performance Review meetings. The first Contractor HSE Leadership Forum brought 250+ participants from 60+ contractor companies into a shared space.

Eleven contractors had been formally red-banded — capability scores below the client’s HSE threshold. None lost the work; each entered a structured Gap Closure Plan with named owners and milestones. By April 2026, every one of the eleven had cleared red — by improvement, not by replacement.

The sustaining mechanism: live performance management

The achievement was eighteen months of discipline. The sustaining platform was the live operational view shown below: contractor segmentation, assurance activity, score-category movement, action ownership and closure performance were visible as management information, not hidden in trackers.

Figure 2. Example screenshot of contractor HSE management dashboard used to track contractors, capability assessment coverage, site verification activity, risk-matrix distribution, red/amber/green banding, annual trends and HSE plan compliance.

What sustained the result

A defining feature of the program was that contractor HSE management was not implemented as a static procedure set, but governed through real-time dashboards that made contractor risk, capability, readiness and follow-up actions visible at portfolio level. These dashboards converted the 12-stage cycle from a documented process into a live management system, enabling leaders to track contractor segmentation, assessment coverage, site verification activity, action closure and ownership discipline in near real time.

The summary dashboard provided management with a consolidated view of the contractor portfolio. It showed not only how many contractors were in scope, but how they were distributed by risk and mode, how much of the contractor base had been assessed, which firms fell into high-risk bands, and how assurance activity was trending over time. This made it possible to differentiate assurance intensity rather than applying the same controls to every contractor. It highlighted red-category contractors, tracked open actions and findings, showed closure performance, and linked contractor development to accountable ownership. In practice, this allowed the team to focus management attention where exposure was highest and to verify whether gap closure plans were progressing rather than merely existing on paper.

For a thought leadership perspective, the significance of these dashboards is not the software or visual design itself. It is the management behavior they enable: the ability to move from reactive reporting to active contractor risk governance. That governance discipline is what underpinned the program outcomes reported below.

Recorded outcomes in the first 18-months of implementation

Outcome note: outcomes are based on SnSD program records.

7.3 AI-powered contractor HSE management: 83% faster assessment

A strong contractor management process is necessary, but it is not sufficient. With 267 contractors moving through different stages of readiness, assessment and mobilisation, even sound procedures still depend on human coordination — and human coordination does not scale. The process had to be digitally enabled, data-driven and visible in real time.

The challenge: a strong process that still struggles to scale

After establishing the full contractor management cycle, a critical leadership question remained: how does an organisation hold control, consistency and visibility when hundreds of contractors are moving at once through readiness, assessment, mobilisation, execution and follow-up — each at a different stage?

For SnSD, the question was practical. Monitoring the status and risk profile of 267 contractors through traditional manual methods was challenging to scale. It depended on fragmented communication, dispersed documentation, repeated follow-ups and individual interpretation of assessment evidence. The procedures were sound; the system underneath them still ran on human coordination. Contractor management had become a strategic safety capability — it determines how quickly and reliably an organisation can mobilise third parties without compromising assurance — and a strategic capability cannot rest on chasing information.

The digital layer: Safetybud^[12] as the single source of truth

To address this, SnSD developed a ‘Contractor HSE management’ module under Safetybud, a mobile HSE platform that connects frontline execution with management visibility. Safetybud digitises the full contractor management lifecycle and creates a single source of truth for contractors, Contract Holders, HSE supervisors and management teams.

Contractor information, capability-assessment inputs, mobilisation-readiness evidence, field-verification findings, corrective actions and performance follow-up are all captured and governed through one integrated digital layer. Through dashboards, management can monitor the full lifecycle at once — contractor status, risk levels, pending actions, mobilisation readiness, verification results and follow-up performance — which is what moves the organisation from reactive contractor administration to proactive contractor risk governance.

The AI use case: capability assessments at decision speed

The most impactful use case was AI-assisted contractor capability-assessment scoring. In the conventional model, assessing one contractor’s capability could take almost two weeks — more than 50 communication touch-points, ten to twelve hours of review, and no centralised document database. The result was not only slow execution but variability in judgement, limited traceability and delayed mobilisation decisions.

With Safetybud, contractor and contract data, assessment records, evidence documents and communication history are consolidated in one place. AI then supports the review of submitted evidence and generates a capability-assessment score within seconds, with more than 90% alignment to expert human review. This shifted HSE and Contract Holder teams from chasing information to making informed decisions — it reduced contractor assessment effort by 83%, improved scoring consistency, and gave process owners a live view of contractor readiness and risk.

Crucially, Safetybud reframes AI as an assurance enabler, not a replacement for professional judgement. It does not remove human accountability from the process; it strengthens it — through faster evidence review, more consistent banding, better documentation traceability and clearer escalation points. Digitalization creates the visibility; AI creates the decision velocity.

References and sources

[1] International Association of Oil & Gas Producers (IOGP). Safety Performance Indicators – 2024 Data. IOGP Report 2024s, published 2025.

[2] Nwankwo, C. D. et al. Analysis of accidents caused by human factors in the oil and gas industry using the HFACS-OGI framework. International Journal of Occupational Safety and Ergonomics, 2022.

[3] Hallowell, M. R. et al. The Statistical Invalidity of TRIR as a Measure of Safety Performance. Professional Safety, ASSP, April 2021.

[4] Interstate Natural Gas Association of America (INGAA). Implementation of Standardized Leading Safety Indicators, 2023.

[5] International Association of Oil & Gas Producers (IOGP). HSE Management: Guidelines for Working Together in a Contract Environment. IOGP Report 423, 2017.

[6] UK Health and Safety Executive. Managing Contractors: A Guide for Employers. HSG159.

[7] CPWR – The Center for Construction Research and Training. Fatal and Nonfatal Falls in the U.S. Construction Industry, 2011–2022. Data Bulletin, March 2024.

[8] U.S. Department of Labor, Mine Safety and Health Administration. U.S. Department of Labor Reports 29 Mine-Related Deaths in 2020. News Release, January 13, 2021.

[9] International Council on Mining and Metals (ICMM). Safety Performance: Benchmarking progress of ICMM company members in 2024. ICMM Safety Performance Report, 2025.

[10] World Steel Association. Safety and health in the steel industry: Data report 2026. worldsteel, 2026.

[11] Concawe. European downstream oil industry safety performance: Statistical summary of reported incidents 2024. Concawe Report No. 9/25, 2025.

[12] Safetybud. Link is available at: https://www.youtube.com/watch?v=wTyhkRmJhAM