SHARE Knowledge Hub
Sensible Risk Management and Risk Assessment Programmes
Sensible risk management identifies the work that could cause real harm, chooses proportionate controls and checks that they work. A risk assessment programme links organisation-wide priorities to specialist, task and point-of-work decisions; the document is useful only if it changes the work. Assessments need enough detail for the risk and the applicable Great Britain or Northern Ireland duty.
Understanding Sensible Risk Management and Risk Assessment Programmes
Sensible risk management means concentrating effort on significant work-related risks, taking proportionate precautions and checking that they work. It is neither risk elimination at any cost nor a light-touch excuse for leaving serious hazards uncontrolled. A risk assessment should change how work is designed, resourced, supervised and reviewed, not just produce a completed form.
Why this matters at work
Failure modes and effects analysis (FMEA) examines how an item or function can fail, what local and wider effects follow, how the failure is detected and what prevents or limits harm. It is useful for machinery, controls, utilities and equipment where component behaviour matters. Define the system boundary and functions first. A pump can fail to start, fail to stop, leak, reverse flow or deliver insufficient flow; each failure has different consequences.
Although this is an assurance process, every useful metric should trace to harm. A machine guard failure can permit entrapment and amputation; a lift defect can cause crush injury; poor LEV can allow isocyanate inhalation and occupational asthma; mixed warehouse traffic can lead to fatal struck-by injury. In petrochemical or energy work, loss of containment, pressure or electrical energy can cause burns, poisoning, electrocution or off-site harm. The monitoring question is which initiating event and barrier state could lead to these outcomes, and what evidence would reveal deterioration before injury.
How organisations manage this in practice
HSE presents five connected steps: identify hazards, assess risks, control risks, record findings and review controls. The cycle begins with seeing actual work and asking workers, then considers who could be harmed and how. It must lead to a justified control decision and verification. If existing controls are suitable, sufficient and effective, record that conclusion and maintain them. Where further controls are needed, implement and verify them; a completed form alone is insufficient.
Close the decision loop with an owner, date, resources, completion evidence, effectiveness test, feedback to affected people and explicit review trigger. Review is triggered by change, new knowledge, control failure, incident, near miss, symptom, worker concern, monitoring result, audit finding or evidence that the assessment is no longer valid. Calendar review alone is not enough. Another competent reviewer should be able to reconstruct why the decision was proportionate, what evidence supported it, what remained uncertain and what would cause work or the programme to stop and reopen.
Record the decisions, owners, resources, deadlines, interim restrictions and verification plan. Review the effect later, including whether a fix shifted risk to another task. Challenge unexplained green results, repeated deferred actions and differences between sites. If indicators no longer explain risk, change them. A review is complete only when actions affect work, not when minutes are approved. Keep health and long-latency risks visible alongside immediate accidents.
Turn the roof-service request into an owned precaution
Risk assessment should change the work, not merely fill a form. Follow the real task from observation and affected people to stronger options, implementation and a check in use.

Observe real work
Understand the actual roof-service need, access, weather, equipment and interfaces with workers. Start with whether access can be avoided or redesigned.
People and harm
Identify who could be affected, including contractors and people below or nearby. Describe credible harm and uncertainty rather than relying on a numerical score alone.
Effective precautions
Apply established good practice and compare feasible stronger measures. Assess new risks from the chosen method rather than accepting a generic control list.
Owner and completion
Assign authority, resources, implementation and evidence. A recommendation that has not been put into practice does not control the task.
Check in use
Verify that the precautions work under representative conditions with the people using them. Resolve gaps before relying on the arrangement.
Change triggers review
Revisit the assessment when the work, people, layout, equipment or evidence changes. Keep the record useful and proportionate to the actual risk.
Read the complete controls: ERICPD planning prompts
Discipline verifies and sustains all the controls; it is not a separate substitute for them.
| Prompt or level | Complete task-specific example |
|---|---|
| Eliminate | Assess moving the roof service point to ground level. |
| Reduce | Assess a suitable fixed platform instead of a portable ladder. |
| Isolate | Provide designed guarded access and secure openings. |
| Control | Use a task-specific permit and briefing where required. |
| PPE | Assess residual fall protection against the actual task. |
| Discipline | Maintain access and stop for missing guards or changed roof layout. Read the accompanying source table for the framework distinctions. |
Read the complete controls: ESEAP effectiveness order
Compare stronger feasible controls while preserving the actual task, checks and release requirements.
| Prompt or level | Complete task-specific example |
|---|---|
| Elimination | Assess moving the roof service point to ground level. |
| Substitution | No separate measure is specified in this example. |
| Engineering controls | Assess a suitable fixed platform instead of a portable ladder. Provide designed guarded access and secure openings. |
| Administrative controls | Use a task-specific permit and briefing where required. Maintain access and stop for missing guards or changed roof layout. |
| PPE | Assess residual fall protection against the actual task. Read the accompanying source table for the framework distinctions. |
The request provides no roof-access design or permission to work at height. Apply the actual site and task requirements.
What are HSE’s five risk assessment steps?
Follow a depot team as it assesses a shared forklift crossing, changes the layout and checks the controls during deliveries.
Identify hazards
Walk through the job with the people who do it. Look at routine work, cleaning, maintenance and changes. Check near misses for ways people have already come close to harm.
At the depot. At the depot, the team finds that forklifts and pedestrians use the same crossing during unloading.
Assess the risks
Identify who might be harmed and how. Consider workers, contractors, visitors and anyone who needs different arrangements. Judge how likely harm is and how serious it could be, taking account of controls already in use.
At the depot. Depot staff and visiting drivers cross the route. A forklift strike could cause serious or fatal injuries. Markings and briefings still leave people and vehicles sharing the crossing.
Control the risks
First consider removing the hazard or avoiding exposure. If risk remains, choose controls that act at source and protect everyone affected. Put the measures into use and brief the people who rely on them.
At the depot. Reroute pedestrians away from unloading where possible. Where a crossing is still needed, fit barriers and choose a crossing point with clear sightlines. Restrict access until the arrangement is ready.
Record your findings
Record the significant hazards, who may be harmed, the controls in place and any further action. Name an owner, a due date and the evidence needed to close the action. In Great Britain, employers with five or more employees must record significant findings.
At the depot. The depot supervisor records the temporary restriction, who owns the permanent route changes, when they are due and who will check them. The shift briefing explains the change to staff and visiting drivers.
Review the controls
Check that the controls work during real shifts. Review when controls may no longer work, after changes, when workers raise problems, or after accidents and near misses. Update the assessment and brief affected people.
At the depot. Observe a busy delivery after installation. Check that people use the crossing, sightlines stay clear and barriers are intact. If the route is blocked or bypassed, deal with that before relying on the assessment.
Download worked SHAREintelligence assessments
Compare two complete fictional teaching reports exported from the SHAREintelligence assessment tools. Each includes four hazard lines, evidence, actions, responsibilities and review arrangements.
Scored assessment · Conveyor maintenance
The conveyor example records 3 × 5 = 15 (High) with isolation still unverified. The proposed 1 × 5 = 5 (Medium) depends on verified isolation and other controls. Intrusive work stays on hold in the example.
Download scored assessment PDFQualitative assessment · Cleaning-product preparation
The cleaning example explains splash and skin-contact risks in words. It prioritises stopping unprotected decanting, reviewing safer methods and checking the chosen control before release.
Download qualitative assessment PDFWhat does sensible risk management really mean?
Start with a credible harm pathway, choose controls that change the work, and keep programme, task and point-of-work decisions connected.
The programme should join risk profiling, task assessment and field checks. A site manager can accept a routine, known task only after checking local conditions; a serious or unfamiliar exposure needs competent specialist assessment and an owner for any engineering change. HSE risk assessment steps.
Sensible risk management is sometimes misunderstood as doing less. It is actually about doing the right things well. Low-risk everyday activities do not need elaborate documents, but high-energy, high-exposure, complex or poorly controlled work needs competent assessment, clear controls, strong supervision and evidence that those controls remain effective.
A suitable and sufficient assessment identifies the significant hazards, who may be harmed, how harm could occur, what controls are already in place, what further controls are needed, who owns them and when review is required. It should be specific enough to guide the work, but practical enough for the people using it.
The best assessment programmes are useful in the field. A worker can understand the control. A supervisor can check it. A manager can resource it. An auditor can sample the evidence. A director can see whether serious-risk controls are working.
| Principle | What it means | Common failure |
|---|---|---|
| Focus on significant risk | Spend most effort on hazards that can cause real harm, serious loss or legal exposure. | Treating trivial hazards and fatality risks with the same paperwork process. |
| Proportionate control | Match control depth to likelihood, severity, uncertainty, exposure and vulnerability. | Using a generic checklist where a technical assessment or design control is needed. |
| Suitable and sufficient assessment | Cover the actual task, people, environment, equipment, abnormal conditions and review triggers. | Copying an old assessment that does not match work as done. |
| Hierarchy of control | Remove, reduce or engineer out risk before relying on PPE, warnings or behaviour. | Jumping straight to training, signs or PPE when the task could be redesigned. |
| Worker involvement | Use the knowledge of people who do the work and see shortcuts, constraints and weak signals. | Writing assessments in an office without observing the job. |
| Clear ownership | Each control, action and review should have a named owner with authority and resources. | Actions sit in a spreadsheet with no real accountability. |
| Evidence of control | Keep evidence that controls are present, working, maintained and understood. | Assuming a control exists because it is written in a document. |
| Review after change | Review assessments after incidents, change, new information, audit findings or work drift. | Annual review becomes a date change without looking at the work. |
How can the hierarchy of controls guide proportionate and effective workplace precautions?
Control classification: A designed fixed platform/guarded access is an engineering measure; it is not an administrative permission. Avoid work at height, then prevent falls, then minimise distance/consequences, preferring collective protection. A separate genuine substitution requires its own explanation. Read this correction with the source table. Retained overview and stage artwork requires the matching classification update.
Choose a control that interrupts the stated harm route and verify it where the work happens.
For roof-plant access, first ask whether the service point can be moved to ground level and the climb eliminated. If access remains essential, consider a designed engineering access measure such as a fixed platform rather than a portable ladder; design guarded access and secure openings before relying on a permit, briefings and fall-protection equipment. The designer records why relocation is feasible or not, facilities commissions the access, and the supervisor checks the guard and access route before each job. A missing guard or changed roof layout stops the task. ERICPD (Eliminate, Reduce, Isolate, Control, PPE, Discipline) and the HSE five-level hierarchy are teaching prompts for control selection; neither replaces task-specific engineering judgement. Here, Discipline means maintaining, supervising and learning from the selected controls. HSE five-level hierarchy of controls HSE hierarchy of controls HSE risk assessment steps.
HSE’s work-at-height guidance prioritises avoiding the climb, then preventing falls, then limiting the consequences of a fall; collective protection is considered before personal protection. For this roof job, the assessment must also address safe access, fragile surfaces, falling objects and rescue. Use the separate NI duty where the work is in Northern Ireland. HSE roof-access control sequence.
How should a risk assessment programme be structured?
A strong programme uses different assessment layers: organisational risk profile, specialist assessments, task assessments, dynamic review and assurance.
A risk assessment programme is more than a folder of assessments. It is the organised way a business decides what needs assessment, who is competent to assess it, how controls are approved, how actions are closed, how workers are briefed and how leaders know whether the programme is working. HSE managing health and safety.
The programme should start from the organisation's risk profile. A leisure centre, construction project, laboratory, warehouse and fuel terminal do not need the same depth of controls. Each needs a structure that covers its significant risks, legal assessments, high-risk tasks, vulnerable groups, contractors, change and emergency conditions.
The programme should also prevent duplication. A COSHH assessment, fire risk assessment, DSEAR assessment, manual handling assessment and task RAMS may all refer to the same job, but each should have a clear purpose. The aim is integrated control, not multiple documents saying different things.
| Layer | Purpose | Examples | Output |
|---|---|---|---|
| Organisational risk profile | Identify the top risks that need leadership attention, resources and assurance. | Fire, workplace transport, violence, asbestos, construction, process safety, occupational health. | Risk profile, board dashboard and assurance plan. |
| Legal/specialist assessments | Meet specific legal or technical requirements with competent input. | Fire risk assessment, COSHH, DSEAR, noise, vibration, manual handling, DSE, legionella, pressure systems. | Assessment, control plan, specialist evidence and review date. |
| Task assessment | Translate risk into the safe way of doing a job. | Maintenance, cleaning, lifting, roof access, chemical decanting, lone working, public events. | SSOW, method, controls, briefing and supervisor checks. |
| Project or contractor RAMS | Coordinate site-specific work, interfaces, permits, temporary controls and handover. | Refurbishment, scaffold, hot work, excavation, plant movement, shutdown work. | Approved RAMS, permits, inductions and interface controls. |
| Dynamic review | Check whether conditions today still match the assessment. | Weather, congestion, new people, failed equipment, public presence, changed sequence. | Stop, adapt, escalate or proceed with confirmed controls. |
| Assurance and review | Test whether assessments lead to real control in the workplace. | Inspections, audits, action verification, incident learning, worker feedback. | Updated assessment, closed actions and management learning. |
How does a sensible assessment move from hazard to decision?
A good assessment describes a credible harm pathway and then selects controls strong enough to break that pathway.
Plan a roof-plant maintenance job
Task material and setting: Replace a fan motor on a leisure-centre roof where the access route crosses an unguarded edge; the outcome is a repaired plant and a safe route.
People and prerequisites: A facilities manager owns the job; a competent engineer and contractor plan the method, protect colleagues and people below, and confirm roof condition, weather, load, access and rescue arrangements.
Before: Consider relocating the service point or working from below. If roof access is essential, specify and commission guarded access or a safe platform, then brief a task-specific method and check the workface.
During: The supervisor checks the physical guard and access against the approved plan before entry, monitors changed weather and other work, and keeps the area below segregated.
Hold or stop: Hold for missing guard, unknown roof condition, adverse weather, changed contractor route or unverified rescue arrangements. A signed generic form is not a release.
Completion and hand back: After the repair, inspect reinstated guards and openings, clear the area and have the facilities owner accept the safe plant and route. Escalate relocation as a longer-term design action.
Evidence: Keep the scenario assessment, option comparison, guard commissioning and pre-use check, contractor briefing, defect record, action owner and hand-back.
At a leisure-centre roof plant, the facilities manager first defines the job: a contractor must replace a fan motor, and the existing route crosses an unguarded roof edge. The people at risk are the contractor, a colleague carrying parts and anyone below. The manager checks the existing assessment against this exact route, weather, load and rescue plan before issuing work. The highest-value option is to relocate serviceable equipment or perform work from below; if that cannot be achieved for this job, the engineer specifies guarded access and a secure work platform. The contractor's method, supervision and fall-protection arrangements address only residual risk. Work stops if the guard, platform or weather condition differs from the approved method. After completion, the manager verifies reinstatement and records the design improvement for future jobs.
| Assessment step | Weak assessment | Sensible professional decision |
|---|---|---|
| Hazard description | Work at height. | Engineer could fall from roof edge or ladder during roof plant inspection, especially in wind, rain or poor light. |
| People affected | Employees. | Maintenance engineers, contractors, escorts, emergency responders and people below exposed to dropped objects. |
| Existing controls | Be careful and use ladder safely. | Check fixed access, edge protection, fragile surfaces, lighting, weather limits, permit route and rescue arrangements. |
| Control selection | Harness available. | First consider relocation, remote monitoring, permanent guarded access, fixed edge protection and safe isolation before relying on fall arrest. |
| Action quality | Brief staff. | Install missing guardrail, set weather stop rule, update access permit, brief contractors and verify after completion. |
| Evidence | Risk assessment signed. | Photographs, access inspection, permit, maintenance record, contractor briefing and supervisor verification. |
| Review trigger | Annual review. | Review after plant change, roof work, weather event, access defect, near miss or contractor feedback. |
What does suitable and sufficient look like?
Suitable and sufficient means the assessment is good enough to identify significant risk, guide controls and support a defensible decision.
A suitable and sufficient assessment is not judged by length. It is judged by whether it understands the work, identifies the people affected, recognises foreseeable harm, selects appropriate controls and leaves a clear evidence trail. A short assessment can be enough for simple work. A longer technical assessment may be needed where risk is high, uncertain or specialist. HSE risk assessment steps.
Competence matters. A supervisor may be competent to assess a routine housekeeping task, but may need specialist help for DSEAR, asbestos, radiation, complex lifting, structural temporary works, occupational hygiene or process safety.
| Quality test | What good looks like | Warning sign |
|---|---|---|
| Scope | The assessment covers the actual work, location, people, equipment and interfaces. | Generic title with no site, task or boundary. |
| Hazard mechanism | It explains how harm could occur, not only the hazard name. | List of hazards without scenarios. |
| People affected | Includes employees, contractors, visitors, public, vulnerable workers and emergency responders where relevant. | Only says 'staff'. |
| Controls | Controls follow hierarchy and are specific, practical and assigned. | Controls say 'take care', 'wear PPE' or 'follow training' only. |
| Residual risk | Residual risk is considered after controls and uncertainty are addressed. | Risk rating reduced without explaining why. |
| Actions | Further controls have owner, priority, date and interim arrangements. | Open actions with no authority, date or verification. |
| Communication | People doing the work can understand and apply the assessment. | Document signed but not briefed. |
| Review | Clear triggers exist for change, incident, audit, new guidance or work drift. | Annual review only, regardless of change. |
Where is the line between proportionate control and bureaucracy?
Proportionate risk management reduces real risk without burying people in documents that do not improve control.
Poor risk management can fail in two opposite ways. One failure is under-control: a dangerous task is treated casually because it is familiar. The other is over-bureaucracy: low-risk work is buried in forms, so people stop reading the important controls.
A professional system makes the difference visible. It uses simple checklists for simple, repeatable work; task-specific assessments for meaningful risk; and specialist assessment where hazards are technical, high consequence or legally specific.
| Situation | Proportionate approach | Over- or under-control to avoid |
|---|---|---|
| Office kettle use | Basic welfare and electrical condition checks. | A lengthy risk assessment for ordinary hot drinks. |
| Minor shelf-stacking | Simple manual handling guidance and supervisor check. | Complex paperwork where loads are light and work is low risk. |
| Repeated heavy lifting | Manual handling assessment, redesign, aids, staffing and early reporting. | Only telling people to lift correctly. |
| Contractor roof work | Site-specific RAMS, access control, edge protection, rescue and supervision. | Accepting a generic contractor document. |
| Hot work near combustible material | Permit, isolation, fire watch, area check, handback and emergency controls. | A generic 'hot work risk assessed' statement. |
| Chemical decanting | COSHH assessment, SDS, ventilation, spill plan, PPE/RPE and training. | Using PPE without assessing exposure route. |
| High-hazard process change | MOC, technical review, HAZOP/LOPA where needed, approval and start-up review. | Treating change as a routine maintenance job. |
How should ALARP and SFARP judgement be used?
ALARP and SFARP judgement asks whether further risk reduction is reasonably practicable, considering risk, control benefit and sacrifice.
So far as reasonably practicable is not a licence to do the cheapest thing. It requires a reasoned judgement about the level of risk, what more could be done, how much risk reduction would be achieved and whether the time, trouble and cost would be grossly disproportionate to the benefit. HSE health and safety law.
For low-risk tasks, simple controls may be enough. For serious risks, the balance shifts. If a control could prevent fatal or life-changing harm, a much higher level of sacrifice is likely to be expected before it can be rejected. The judgement should be recorded in plain language, especially where a higher-order control is not implemented.
| Decision question | What to consider | Evidence expected |
|---|---|---|
| What is the credible harm? | Fatality, life-changing injury, occupational disease, major accident, public harm or business-critical loss. | Risk scenario and affected people. |
| What controls are possible? | Elimination, substitution, engineering, isolation, procedural control, competence, PPE and emergency measures. | Options considered, not only chosen control. |
| How much risk reduction is achieved? | Does the option remove the pathway, reduce exposure, improve reliability or only rely on behaviour? | Control effectiveness rationale. |
| What is the sacrifice? | Money, time, trouble, operational disruption and technical feasibility. | Cost/time estimates where needed. |
| Is rejection defensible? | For serious risk, rejection needs strong reasoning and evidence. | Recorded ALARP/SFARP judgement and approval. |
| What interim controls are needed? | Temporary controls while permanent action is planned or installed. | Interim risk acceptance, owner and review date. |
| Who approves? | Competent manager, technical authority or senior leader depending on risk. | Decision record and escalation route. |
What are the different levels of risk assessment?
Risk assessment is a layered system, from organisational risk profiling down to point-of-work checks and assurance.
A mature risk assessment programme connects decisions at different levels. The board needs to understand the organisation's biggest sources of harm, managers need site and specialist assessments that define control standards, supervisors need task assessments and permits that match work as planned, and workers need a final point-of-work check before conditions change under their feet.
The layers should not be treated as substitutes for each other. A last-minute risk assessment must not replace planning, and a strategic risk register must not become detached from work as done. The strongest programmes can trace a serious risk through this chain: organisation risk profile -> premises/specialist assessments -> task assessment -> permit/briefing -> last-minute check -> assurance.
Fire, DSEAR, COSHH, asbestos, noise, vibration and QRA are not simply longer versions of a normal workplace risk assessment. They require specific competence, evidence, assumptions, legal duties, technical data and review triggers. A general risk assessment may identify that one of these topics is present; it should then signpost the need for the correct specialist assessment rather than trying to absorb everything into a generic form.
| Level | Purpose | Examples | Output |
|---|---|---|---|
| Organisational risk profile | Identifies the biggest risks across the business. | Top 10 company risks, strategic risk register, board risk profile. | Leadership priorities, risk owners and assurance plan. |
| Site / premises assessments | Controls risks linked to a location, asset, building, landlord interface or operating environment. | Fire risk assessment, DSEAR assessment, asbestos, legionella, workplace transport, security and welfare. | Site controls, inspection regime, action plan and accountable premises owner. |
| Specialist technical assessments | Used where competence, law or complexity requires deeper review. | QRA, COSHH exposure assessment, DSEAR zoning, noise, vibration, pressure systems and temporary works. | Technical basis, assumptions, control standard and specialist evidence. |
| Quantitative / semi-quantitative assessment | Gives structured scoring, modelling or risk comparison where useful. | QRA, LOPA, 5x5 risk assessment, bowtie and risk ranking. | Risk comparison, tolerability judgement, priorities and decision evidence. |
| Activity / task risk assessment | Controls a specific job, work sequence, interface or temporary activity. | RAMS, method statement, lifting plan, permit-to-work and maintenance task assessment. | Task controls, briefing, supervision and stop triggers. |
| Dynamic / last-minute risk assessment | Final point-of-work check before or during work. | Take 5, point-of-work risk assessment and supervisor pause check. | Confirm conditions match the plan, or stop and escalate. |
Distinguish the decision level from the analytical method: bowtie commonly communicates qualitative barriers; a 5×5 matrix provides ordinal ranking; LOPA is commonly semi-quantitative; QRA requires quantitative evidence. Drawing a bowtie does not itself produce numerical risk evidence.
Which level of assessment should be used?
Choose the level that matches the decision, the harm potential, the legal duty, the competence needed and the uncertainty in the work.
The practical question is not 'do we have a form?' It is 'what decision are we trying to make, and what evidence is needed to make it responsibly?' A supervisor may only need a point-of-work pause when a delivery route is temporarily blocked, but a company deciding whether a fuel transfer system is adequately protected needs technical assessment, operating limits, maintenance evidence and senior assurance.
Dynamic risk assessment is not permission to improvise around poor planning. If the job is outside the assessment, outside the method, outside the permit conditions or outside the competence of the people present, the correct decision is to stop, make the area safe and escalate to the right owner.
| Question | Likely level | What the decision should produce |
|---|---|---|
| Is this one of the organisation's biggest sources of fatality, major accident, occupational disease, legal exposure or business interruption? | Organisational risk profile | Named risk owner, control strategy, assurance plan, leadership review and resource decision. |
| Is the risk tied to a site, premises, asset, public interface or landlord / tenant boundary? | Site / premises assessment | Site controls, inspection regime, action plan, communication and review after layout or occupancy change. |
| Does the topic have specific legal, technical or competence requirements such as fire, DSEAR, COSHH, asbestos, noise, vibration or QRA? | Specialist technical assessment | Competent assessment, assumptions, measurement or modelling where needed, control standard and review trigger. |
| Is the risk part of a defined job, work sequence, contractor activity or temporary project? | Activity / task assessment | Task controls, RAMS or method, briefing, supervision, permit links and stop-work conditions. |
| Does the task involve energy, isolation, breaking containment, height, lifting, confined space, live electrical work or multiple interfaces? | Permit, specialist authorisation or redesign | Authorised conditions, isolations, emergency arrangements, handover and close-out evidence. |
| Are conditions at the job different from the plan because of weather, congestion, failed equipment, new workers, public presence or changed sequence? | Dynamic / last-minute risk assessment | Proceed only if controls still match the plan; otherwise stop, make safe and escalate. |
| Are controls claimed to be in place, but not visibly reliable? | Assurance / proactive monitoring | Verification evidence, defect action, repeat-finding review and management escalation where needed. |
Which types of risk assessment are commonly used?
Different assessment types serve different purposes, and a mature programme knows when each is needed.
Many organisations use the phrase 'risk assessment' as though every assessment is the same. In practice, a programme may need generic assessments, site-specific assessments, task assessments, legal/specialist assessments, dynamic assessment, quantitative analysis and critical-control standards.
The wrong assessment type creates false confidence. A generic warehouse assessment may help with routine housekeeping, but it will not be enough for a new loading-bay layout, lithium battery charging area, high-rack maintenance job or contractor lifting operation.
| Assessment type | Best use | Where it fails |
|---|---|---|
| Generic assessment | Simple, repeatable low-risk work with little variation. | Fails where site, people, equipment or interface risk changes. |
| Specific/task assessment | A defined task, location, group of people or work sequence. | Fails if not briefed, supervised or reviewed after change. |
| Dynamic assessment | Final check before or during work where conditions can change. | Fails if used as an excuse for poor planning. |
| Legal/specialist assessment | COSHH, fire, DSEAR, noise, vibration, manual handling, DSE, legionella, asbestos or radiation. | Fails if completed by someone without technical competence. |
| Quantitative or semi-quantitative assessment | Major hazard, reliability, exposure or complex risk comparison. | Fails if numbers hide uncertainty or replace judgement. |
| Critical-control assessment | Serious risks where certain controls must not fail. | Fails if controls are listed but not verified. |
| Project/RAMS assessment | Contractor or temporary work needing method, sequence and interface control. | Fails when copied across projects without site review. |
Why is a risk matrix not risk management?
A risk matrix can help prioritise, but the real work is understanding harm pathways, choosing controls and verifying effectiveness.
Risk matrices are useful when they support conversation. They are weak when they become a scoring ritual. Two assessors can score the same task differently, and a neat number can hide uncertainty, high-consequence exposure or weak controls. HSE risk assessment steps.
The matrix should never be the evidence of safety. Evidence is found in design decisions, engineering controls, maintenance records, competent people, observed behaviour, monitoring results and verified action close-out.
For the roof job, a low probability score may reflect that no one has fallen yet, while the fall consequence remains severe. A matrix can help decide which assessor and approver to involve, but it cannot turn an unguarded edge into an acceptable route. Record assumptions about frequency, weather and exposure, then test the physical control and the worker's ability to use it. If those facts are unknown, the score is uncertain and the task waits.
| Matrix limitation | Why it matters | Better practice |
|---|---|---|
| False precision | A 3 x 3 or 5 x 5 score can look more scientific than it is. | Use narrative risk scenarios and uncertainty notes. |
| Actual versus potential harm confusion | A low-consequence near miss may have high fatality potential. | Assess credible worst-case and high-potential events. |
| Control reliability ignored | Same score may be used whether controls are strong or weak. | Assess control criticality, condition and assurance evidence. |
| Bias and pressure | Scores can be lowered to make work proceed. | Require review for high-risk or contested decisions. |
| Action quality hidden | Score changes may be claimed after weak actions. | Check whether controls remove or reduce the pathway. |
| No ownership | Score exists but no one owns the control. | Assign owners for controls, actions and verification. |
How does sensible risk management apply across sectors?
The principles are common, but the examples, controls and evidence change by sector and risk profile.
A credible page needs to show that risk management is not abstract. In an office it may focus on DSE, fire, stress, access and contractor work. In construction it may focus on work at height, excavations, lifting, dust, plant and temporary works. In process safety it may focus on containment, ignition, pressure, operating envelopes and critical barriers.
The same logic applies in every case: understand the work, identify the credible harm pathway, select proportionate controls, brief the people, verify the controls and review after change.
For a distribution yard, the site manager first maps visiting vehicles, pedestrian crossings and loading times with the drivers and warehouse team. A physical route separation and controlled loading zone address the collision path; a sign alone leaves people exposed. The supervisor checks the route at the start of each shift and stops a delivery when a barrier is missing or the unloading area is occupied. This is an illustration of the assessment steps in HSENI workplace-transport guidance for Northern Ireland; for a Great Britain site, use the applicable HSE workplace-transport assessment guidance.
In a care setting, the goal may be to enable an ordinary activity while preventing significant harm. The assessor involves the person receiving care and those supporting them, chooses practical precautions, records who will check them and reviews the plan as needs change. The decision cannot be imported unchanged into a construction or major-hazard task: the people, harm route, available controls and governing duties differ. HSE sensible risk assessment in care settings gives the Great Britain care context.
| Sector | Typical significant risks | What good risk management looks like |
|---|---|---|
| Office and hybrid work | DSE, stress, fire, slips, visitors, lone work and contractors. | Simple DSE process, clear reporting, fire arrangements, workload review and contractor controls. |
| Warehouse and logistics | Vehicle-pedestrian interface, loading, racking, manual handling, fatigue and visiting drivers. | Segregated routes, loading rules, defect reporting, driver briefing and proactive monitoring. |
| Construction | Work at height, excavation, plant, lifting, dust, noise, HAVS and temporary works. | Task-specific planning, competent appointments, permits, inspections and daily change review. |
| Manufacturing | Machinery, energy isolation, maintenance, ergonomics, chemicals, noise and shift handover. | PUWER controls, lockout, maintenance planning, LEV, competence and defect escalation. |
| Leisure and public-facing premises | Fire, slips, pool plant, public interface, events, safeguarding and contractors. | Public risk review, emergency plans, plant-room controls, inspection and visitor communication. |
| Process and high-hazard sites | Loss of containment, fire, explosion, pressure, reaction hazard and control-system failure. | MOC, HAZOP/LOPA, operating envelope, critical-control assurance and board-level indicators. |
An illustrative multi-site programme can trace the same decision across different tasks: a warehouse crossing needs a site-specific vehicle/pedestrian assessment, while office screen work needs its own workstation and routine assessment. The programme owner records each task owner, further action and verification method. The supervisor checks the real work before release; the programme reviewer samples whether the action changed exposure. A common form can organise the records, but cannot establish that both tasks have the same controls or that an unresolved serious hazard is acceptable. HSE task-specific assessment process.
How should a risk assessment programme be governed?
Programme governance makes sure assessments are owned, current, useful, briefed, actioned and assured.
A common weakness is allowing hundreds of assessments to accumulate with no clear owner or review logic. Mature organisations maintain a register, apply risk-based review frequencies, assign competent assessors, track actions, link assessments to procedures and sample whether controls work in practice. HSE managing health and safety.
Governance should also include change. New equipment, layout changes, new substances, contractor activities, incidents, audit findings, legal updates and worker feedback should trigger assessment review before the next annual date.
A programme owner should keep a live register of assessments, significant actions, site owners, expiry or change triggers and evidence that controls have been installed. A monthly review of only document dates misses whether the roof guard was built or whether workers use it. Sample a high-consequence task on site, ask the worker how the control is checked, compare the answer with the assessment and escalate any gap to the budget holder. Close the action only after field verification.
| Governance element | What it controls | Example evidence |
|---|---|---|
| Assessment register | What assessments exist, who owns them and when review is due. | Register with scope, owner, date, status and links. |
| Competence standard | Who may write, review or approve different assessment types. | Assessor matrix and specialist escalation rules. |
| Approval route | Which decisions need supervisor, manager, specialist or senior sign-off. | Approval thresholds and exception records. |
| Action management | How further controls are prioritised, funded, tracked and verified. | Action tracker with verification evidence. |
| Communication | How controls are briefed to workers and contractors. | Briefing records and worker understanding checks. |
| Assurance plan | How assessment quality and control effectiveness are sampled. | Audit schedule, inspection results and trend review. |
| Management review | How leaders review top risks, overdue actions and repeated weak signals. | Board pack or senior review minutes. |
What does a strong learner answer include?
A strong answer moves from definition to application, control quality, evidence and review.
For safety and professional learning, the strongest answers do not simply say 'complete a risk assessment'. They explain the purpose of the assessment, the legal duty, the people affected, the hazard pathway, the hierarchy of control, competence, communication, monitoring and review.
A weak answer often relies on generic phrases such as 'train staff' or 'wear PPE'. A stronger answer explains why a control is needed, how it reduces risk, who owns it and how the organisation will know it remains effective.
| Question | Weak answer | Stronger professional answer |
|---|---|---|
| What is sensible risk management? | Doing risk assessments and keeping records. | Focusing effort on significant risks, proportionate controls, competent judgement, evidence and review. |
| How do you control a high-risk task? | Train staff and issue PPE. | Eliminate or redesign where possible, engineer/isolate the hazard, define SSOW, train, supervise and verify controls. |
| How do you decide if risk is acceptable? | Use the risk matrix score. | Consider credible harm, legal duties, available controls, ALARP/SFARP judgement, uncertainty, vulnerable people and assurance evidence. |
| What makes an assessment suitable? | It has been signed and dated. | It matches the task, identifies affected people, explains harm routes, selects controls and includes actions and review triggers. |
| How should assessments be reviewed? | Every year. | After change, incident, near miss, audit finding, new information, worker feedback or evidence that controls are weak. |
When should a risk be escalated, redesigned or accepted?
Escalate when risk is serious, controls are uncertain, competence is missing, legal duties are specific or decisions need resources beyond the supervisor.
A sensible system gives supervisors permission to stop and escalate. If a task cannot be done within the agreed controls, the answer is not to make the paperwork fit the job. The work should stop, be redesigned, receive competent review or be authorised with clear temporary controls. HSE risk assessment steps.
Risk acceptance should be explicit for serious risks. It should identify what risk remains, why further controls are not reasonably practicable at that point, who has authority to accept the risk, what interim controls apply and when the decision will be reviewed.
The supervisor may stop an unfamiliar job and arrange the approved safe method, but a decision to reject a guarded platform because of cost belongs with a competent dutyholder. That person should compare credible harm, recognised practice, the available design options and evidence of actual risk reduction. If the decision depends on specialist engineering or a disputed legal duty, obtain that advice before release rather than treating a signed form as permission.
| Trigger | Decision | Who should be involved |
|---|---|---|
| Control missing or defective | Stop or restrict work until control is restored or temporary control approved. | Supervisor, manager, competent person and affected workers. |
| Task has changed from assessment | Pause and review assessment before continuing. | Supervisor, task owner and workers doing the job. |
| High-consequence hazard | Require specialist assessment or senior approval. | HSE adviser, technical specialist, manager or dutyholder. |
| Reasonable control requires budget/design change | Escalate with risk, options, cost, benefit and interim controls. | Manager, asset/project owner and senior leader. |
| Worker raises credible concern | Investigate respectfully and check work as done. | Supervisor, worker representative and competent adviser where needed. |
| Contractor RAMS do not match site | Reject, revise or coordinate before work starts. | Contract manager, principal contractor/site manager and contractor supervisor. |
| Residual risk remains high | Hold inadequately controlled work. Consider redesign, genuine substitution and additional engineering; formal acceptance records a lawful, adequately controlled residual-risk decision and cannot replace the controls. | Senior manager and specialist adviser. |
Formal risk acceptance records a lawful and adequately controlled decision; it does not replace a control or waive applicable duties. Keep work stopped where effective protection cannot be demonstrated, regardless of the seniority of the signatory.
Where do risk assessment programmes commonly fail?
Failure usually comes from generic documents, weak ownership, poor action closure, lack of worker involvement or no link to real control.
A failing programme often looks busy. There may be hundreds of assessments, all reviewed annually and signed by managers, while the workplace still contains obvious uncontrolled risks. The problem is not lack of paper; it is lack of useful thinking, ownership and verification.
The deepest failures are cultural and organisational. Assessments are copied because time is tight. Supervisors sign documents they have not read. Workers are briefed on controls that are impossible in the real job. Actions stay open because nobody owns the resource decision.
| Failure or weak practice | How it develops | What good looks like |
|---|---|---|
| Generic assessment | Same wording used across sites, tasks or contractors. | Assessment reflects the actual task, people, location and interfaces. |
| Paperwork without control | Document says controls exist but workplace evidence is weak. | Controls are observable, maintained, supervised and verified. |
| Matrix-led thinking | Risk score drives discussion more than harm pathway. | Scenario, controls and uncertainty drive decisions. |
| Poor action closure | Actions marked complete after email or briefing only. | Actions are verified in the workplace with evidence. |
| No worker input | Assessment written without task observation or worker involvement. | Workers help identify practical constraints and weak signals. |
| No change trigger | Assessment remains current despite layout, equipment or people changes. | MOC and review triggers are embedded. |
| Assessment overload | Too many documents reduce attention to critical controls. | Programme is rationalised around significant risks and usable controls. |
| Contractor disconnect | Contractor RAMS approved but not checked against site conditions. | RAMS reviewed, briefed, supervised and adjusted for interfaces. |
What evidence shows good risk management?
Evidence should show that assessments are suitable, controls are implemented, people understand them and the organisation learns from monitoring and change.
The strongest evidence connects the whole chain: risk profile, assessment, control selection, communication, action tracking, inspection, incident learning and management review. A reviewer should be able to pick a significant risk and trace how the organisation understands it and controls it.
For serious risks, evidence should go beyond the assessment form. Look for inspection records, maintenance records, competence checks, permits, monitoring data, worker feedback, action verification and assurance reports.
| Evidence item | What it should show | Assurance question |
|---|---|---|
| Risk assessment register | Assessment title, scope, owner, review date, risk rating, critical controls and linked actions. | Can leaders see the assessment population and overdue reviews? |
| Control standard or procedure | How the control should be designed, used, maintained and supervised. | Does the written control match work as done? |
| Action tracker | Action, owner, due date, priority, interim control and close-out evidence. | Are high-risk actions verified in the workplace? |
| Training and briefing records | Who was trained, on what, when, by whom and how understanding was tested. | Can people explain the controls and stop conditions? |
| Inspection and assurance records | Checks on control presence, condition, use, defect response and repeat findings. | Are checks focused on things that prevent serious harm? |
| Incident and near-miss learning | Weak signals, high-potential events, causes, actions and verification. | Has learning changed assessments, controls or supervision? |
| Assessment quality review | Sample checks against task, control hierarchy, legal duties, actions and review triggers. | Does quality improve over time? |
| Risk profile and board pack | Top risks, critical controls, overdue actions, incidents, weak signals and decisions required. | Are leaders seeing real risk, not just activity counts? |
| MOC records | Change screened for risk impact before implementation. | Were assessments updated before controls changed? |
| Contractor RAMS review | Site-specific assessment of interface risks, permits, competence and supervision. | Does contractor paperwork match the actual job? |
How should sensible risk management be assured?
Assurance should test whether the programme improves control of significant risks, not only whether documents exist.
First-line assurance belongs close to the work: supervisors checking that controls are present and understood. Second-line assurance should sample assessment quality, action closure and risk-based coverage. Third-line assurance should test whether governance, competence, legal compliance and critical controls are effective across the organisation.
A strong assurance question is: if we select one significant risk, can we trace it from the risk profile to the assessment, control standard, briefing, inspection, incident learning and management review?
| Assurance line | What to test | Example test |
|---|---|---|
| First line verification | Supervisors check controls before and during work. | Observe a task and ask workers to explain the top three controls and stop points. |
| Second line assurance | Managers/HSE advisers sample assessment quality, action closure and topic coverage. | Review five high-risk assessments against work as done and action evidence. |
| Third line audit | Independent or specialist review tests the programme, competence and governance. | Audit whether the risk profile, assessments, indicators and board review align. |
| Worker voice | Workers can raise concerns and see action. | Ask whether assessments reflect the real job and whether feedback changes controls. |
| Contractor assurance | Contractor assessments are site-specific and supervised. | Sample RAMS, permits, inductions and field checks for one contractor task. |
| Management review | Senior leaders review risk quality, overdue actions and weak signals. | Trace one major risk into resource decisions and follow-up. |
Which duties and guidance apply in Great Britain and Northern Ireland?
Select the legal instrument for the work location and the actual risk, then identify the specific duty that informs the control decision.
The assessment should show the decision and control, not just cite a statute. GB workplace duties; NI workplace duties.
| Jurisdiction and route | Direct instrument or guidance | Supporting source | Decision supported |
|---|---|---|---|
| Great Britain · GB general duties | Health and Safety at Work etc. Act 1974 | HSE health and safety law | Protect workers and others so far as reasonably practicable; decide controls from the actual risk. |
| Great Britain · GB management rules | Management of Health and Safety at Work Regulations 1999 | HSE risk assessment steps | Make suitable and sufficient assessments and record significant findings if five or more employees. |
| Northern Ireland · NI general duties | Health and Safety at Work (Northern Ireland) Order 1978 | Management of Health and Safety at Work Regulations (Northern Ireland) 2000 | Use the separate NI employer duty and assessment regime. |
The five-employee recording threshold concerns general GB Management Regulations duties; all employers still assess and act. Specialist duties can differ. In England and Wales, the Fire Safety Order requires the full fire risk assessment to be recorded regardless of headcount following the October 2023 change; use separate Scottish and NI fire routes.
How should management check this work?
Use a defined plan, carry out the controls, check performance and correct what does not work.
For sensible risk management and risk assessment programmes, assign a competent owner to set the work controls, check their use and review them after change, failure or new evidence. HSE managing health and safety
A management-system label or certificate is not approval of a site limit, procedure or individual work decision. Apply GB and NI duties to the correct location.
What should each role learn and demonstrate?
Competence is evidenced by a person making and explaining the right decision for a realistic task, then responding to change.
A record should show the assessed scenario, observer, action taken and the trigger for reassessment; attendance alone is not proof of performance.
| Role | Learning | Demonstrated competence | Assessor and record | Reassessment trigger |
|---|---|---|---|---|
| Workers and contractors | Recognise a changed task, failed barrier and stop route. | In a roof-access drill, point out the missing guard and refuse access until an approved control is in place. | Supervisor observation and briefing record | New task, change or near miss |
| Supervisors | Translate assessment into checks and named action owners. | Walk the job, explain the harm route and verify the specified control before release. | Operations manager sampling real jobs | Defect, new contractor or control change |
| Assessors and managers | Select stronger controls and judge significant findings and escalation. | Produce a task-specific assessment with options, owners, evidence and review trigger. | Competent H&S lead peer review | High-consequence risk or failed assurance |
| Senior dutyholders | Fund priority controls and test programme assurance. | Challenge a rejected engineering control and close a sampled high-risk action. | Board or independent audit evidence | Risk profile, law or resource change |
What do the acronyms and key terms mean?
These terms explain the decisions on this page.
| Acronym or term | Full expansion | Meaning here |
|---|---|---|
| ALARP | As low as reasonably practicable | Risk-reduction expression used in technical decisions. |
| SFARP | So far as reasonably practicable | Qualified legal duty; risk is weighed against the sacrifice of further control. |
| HSE | Health and Safety Executive | GB regulator and guidance publisher. |
| HSENI | Health and Safety Executive for Northern Ireland | NI regulator. |
| Residual risk | Risk after controls | Needs an owner and a review trigger. |
| AI | Artificial intelligence | Automated response tool; use the verified source and task facts. |
| COSHH | Control of Substances Hazardous to Health | Substance-control regime; GB and NI instruments differ. |
| DSE | Display screen equipment | Screen-work risk and work-routine context. |
| DSEAR | Dangerous Substances and Explosive Atmospheres Regulations | GB fire and explosion risk framework; NI has a separate instrument. |
| GB | Great Britain | England, Scotland and Wales for the stated legal route. |
| HAVS | Hand-arm vibration syndrome | Health condition from harmful hand-transmitted vibration. |
| HAZOP | Hazard and operability study | Structured process deviation review. |
| LEV | Local exhaust ventilation | Captures airborne contamination near its source. |
| LOPA | Layers of protection analysis | Method for examining independent protective layers. |
| MOC | Management of change | Review of risk before a planned change. |
| NI | Northern Ireland | Separate legal and service route in this article. |
| PPE | Personal protective equipment | Last-line protection for remaining exposure. |
| PUWER | Provision and Use of Work Equipment Regulations | GB work-equipment regime; NI has a separate instrument. |
| QRA | Quantitative risk assessment | Numerical risk analysis where warranted. |
| RAMS | Risk assessment and method statement | Linked record of risk and task method. |
| RPE | Respiratory protective equipment | Breathing protection selected for residual exposure. |
| SDS | Safety data sheet | Product-specific chemical hazard and response information. |
| SSOW | Safe system of work | Agreed practical method with controls and stop points. |
| UK | United Kingdom | Geographical umbrella; GB and NI law can differ. |
| ERICPD | Eliminate, Reduce, Isolate, Control, PPE, Discipline | Teaching framework for control selection and sustained supervision; not a statutory rule. |
Unlock another resource
You have used this month’s two free downloads. Publish a post in your own words to unlock one more resource, or choose SHARE Individual for 10 downloads per month.
You decide whether to share. If you do, use the relevant platform disclosure tools and do not make claims you do not genuinely hold.