You have probably seen it, and if you are honest you have probably built it at least once: a hazard assessment that looks polished, covers every department, and checks every box, yet the one risk that could actually kill someone is either missing, minimized, or buried under “general precautions.”
That is not a paperwork problem, it is a risk-control problem, because a hazard assessment can be technically complete and still operationally wrong when “complete” means “we filled out all the fields” instead of “we identified the exposures that matter most and proved the controls are reliable.”
The good news is that you do not need to throw your whole program in the dumpster to fix this, because most organizations already have the right pieces, they just need a better process for spotting high-severity exposure, stress-testing assumptions, and verifying controls in real work.
Hazard assessments drift when the process rewards coverage over clarity, which means people list hazards by category, paste generic controls like “PPE required,” and move on, because everyone is busy and nobody wants to be the person slowing the job down.
The observation is simple: forms get completed fastest when they describe ideal work, not real work. The implication is that the assessment starts documenting what should happen instead of what actually happens, and that gap is where serious incidents live. The decision you need to make is to treat the hazard assessment like a living control tool that has to survive contact with real work, not like a document that wins points for being long.
There is also a psychological trap hiding in plain sight: familiarity. When a task happens every day and “nothing bad has happened,” people start believing the risk is low, even if the potential severity is massive, because humans confuse normal with safe, and safe with controlled.
The risks that hurt people the worst tend to share a few traits: they are low frequency but high consequence, they show up during non-routine work, and they rely on multiple controls working correctly at the same time, which means one small failure can cascade into a serious outcome.
Think about energy isolation during maintenance, vehicle and pedestrian interaction, work at heights, confined space entry, simultaneous operations, contractor interfaces, line breaks, troubleshooting, and those “just for a minute” tasks where people step outside the normal plan.
If your risk rating approach leans heavily on likelihood, you can accidentally down-rank these hazards because “it probably will not happen.” The implication is brutal: you spend most of your time managing frequent minor injuries while the fatality-and-serious-injury exposures sit in the middle of the page with bland controls. A better decision is to anchor on potential severity, then challenge likelihood using evidence, not confidence.
In Alberta, the expectation is not simply that hazards get listed, but that hazards are assessed before work starts, that workers are involved, and that controls are selected and documented in a way that actually reduces exposure, including when conditions change.
For federally regulated workplaces in Canada, the hazard prevention program approach pushes the same direction: have a methodology, identify hazards, assess risk, implement preventive measures, train people on what matters, and evaluate whether the program is working, because hazard prevention is a system, not a one-time worksheet.
The practical takeaway is not “do more paperwork.” The practical takeaway is that your hazard assessment should drive real decisions about controls and verification, and it should get updated when work changes, not just when the calendar says it is review season.
Normalization of deviance. The written standard stays clean, but shortcuts become normal, so the assessment describes ideal work while the field reality drifts, and everyone stops noticing because the drift is gradual.
Non-routine work gets treated like it does not belong. Shutdowns, maintenance, clearing jams, cleaning, startups, troubleshooting, after-hours callouts, and emergency response are exactly where high-severity exposures appear, yet they are often missing or lightly covered.
Interfaces and handoffs are not owned. Contractors, delivery drivers, visitors, multi-employer worksites, and simultaneous operations create gaps, because each party assumes the other party is managing the risk, and “assumed ownership” is not ownership.
Risk ratings are driven by confidence, not evidence. People rate likelihood low because they feel confident, not because they have verification records, inspection results, competency checks, or recent field observations proving controls are reliable.
Controls are listed, but not verified. “Training provided” and “procedure followed” are not controls unless you can show they are current, understood, and actually used when the job gets messy or the pressure is on.
If you want to find the most dangerous missing risk, do not start by rewriting the whole hazard assessment, because you will just create a longer version of the same blind spots.
Start by interrogating it like a skeptic, using real work scenarios and real control evidence, and focus on tasks where severity is high even if frequency is low.
Use these questions as a practical filter, and be honest about the answers, because the goal is not to look good, it is to be safe:
If your answers sound like “people know what to do” or “we remind them,” treat that as a red flag, because reminders are not controls, they are communication, and communication does not stop energy, gravity, vehicles, or toxic exposures.
A stronger hazard assessment is not necessarily longer, but it is sharper, because it uses better inputs, clearer assumptions, and control statements that connect directly to how work is actually performed.
Build a task inventory that includes non-routine work. If you only assess routine production tasks, you are only assessing part of your risk, so include maintenance, clearing jams, cleaning, changeovers, working alone, contractor work, deliveries, seasonal work, and emergency response activities.
Treat high-severity hazards like critical risks. When the potential outcome is fatality or life-altering injury, increase the rigor, which usually means defining critical controls clearly, assigning ownership, and verifying those controls more often.
Use the hierarchy of controls as your decision filter. Canadian Centre for Occupational Health and Safety lays this out well: elimination and substitution are generally more reliable than administrative rules and PPE, and engineering controls often provide the biggest reliability jump when human behavior is variable.
Separate “control exists” from “control works.” A guard is not a control if it is bypassed, a procedure is not a control if nobody can find it, and a training record is not a control if competence is not checked, so build your assessment so the control statement includes what “good” looks like and how it is confirmed.
Make worker participation real, not ceremonial. The fastest way to find hidden high-risk work is to talk to the people doing the job, especially the experienced workers who have seen drift happen over time, and then to validate those conversations with field observation.
Create review triggers that reflect reality. A hazard assessment should change when work changes, so treat new equipment, new chemicals, layout changes, staffing changes, new contractors, production pressure, incidents and near misses, and process changes as automatic triggers for review and update.
Make sure your “formal” hazard assessment process is actually formal. Energy Safety Canada has useful resources for formal hazard assessment thinking, including the idea that you assess tasks across the organization, not just the obvious high-risk ones, and you keep the assessment connected to what people really do.
If you are reading this and thinking, “We have the document, but I do not trust the document,” you are not alone, and that is a solvable problem, because what you need is not more paperwork, it is a better hazard identification and control verification process that supervisors and crews will actually use.
We can facilitate hazard assessment workshops that surface non-routine work, interface risks, and critical controls without turning the room into a blame session, and we can help you rebuild risk rating logic so high-severity hazards do not get artificially downgraded by “low likelihood” assumptions that are not supported by evidence.
We can also field-verify controls and help you tighten the link between hazard assessment, training, inspections, and corrective actions, because a hazard assessment that does not drive action is basically a fancy list, and lists do not prevent incidents.
If you want to see what we do or you want help pressure-testing your current hazard assessment process, start here: [https://calgarysafetyconsultants.ca](https://calgarysafetyconsultants.ca)
A hazard assessment that “looks complete” can be one of the most dangerous documents in your business, because it creates confidence without control, which means people stop looking for the exposure that will actually change a life.
If you want safer work and stronger due diligence, build your hazard assessment around real tasks, high-severity potential, and verified controls, because what you do not deliberately control is exactly what will eventually control you.
Connect with us here and let us help you improve your OH&S practices.
https://laws.justice.gc.ca/eng/regulations/sor-86-304/page-41.html
https://www.ccohs.ca/oshanswers/hsprograms/hazard/risk_assessment.html
https://www.ccohs.ca/oshanswers/hsprograms/hazard/hierarchy_controls.html
https://www.ccohs.ca/oshanswers/hsprograms/hazard/hazard_control.html
https://www.energysafetycanada.com/COR/Formal-Hazards-Assessment-Guide
https://www.hse.gov.uk/simple-health-safety/risk/steps-needed-to-manage-risk.htm
https://www.iso.org/standard/63787.html
A hazard assessment can look complete when it covers categories and checkboxes, but it becomes unsafe when it describes ideal work instead of real work, and when controls are listed without proof that they actually work in the field.
It is usually tied to high-severity exposures like hazardous energy, vehicles, work at heights, confined spaces, or simultaneous operations, especially during non-routine tasks like maintenance, shutdowns, troubleshooting, and startups.
Start by listing tasks that could cause a fatality or permanent disability if controls fail, then pressure-test the “likelihood” rating using evidence such as inspections, competency checks, verification logs, and field observations.
Calgary Safety Consultants is here to help you ensure compliance, enhance safety, and streamline your OH&S program. Don’t wait—fill out the form, and we’ll connect with you to discuss how we can support your business. Let’s get started!
Calgary Safety Consultants delivers game-changing COR Audits and OH&S consulting designed to keep your business safe, compliant, and accident-free. With 30+ years of expertise in high-risk industries like oil & gas, construction, and manufacturing, we understand the unique challenges you face. We offer tailored safety solutions that not only meet regulatory standards but also reduce operational risks and protect your workforce. From conducting thorough safety audits to implementing comprehensive safety management systems, we ensure your team is fully supported. Our hands-on, collaborative approach means we're with you every step of the way, helping you achieve a safer, more efficient workplace. Let us handle the heavy lifting so your team stays protected, and your business thrives!
