What Is a Safe Method Variation

A Safe Method Variation (SMV) is the difference between the approved or expected safe method of work (work-as-planned) and the method actually used in practice (work-as-done). These variations are not always negative. In many cases they reveal how frontline teams adapt procedures to improve safety, efficiency, or practicality.

Best captured using a planned task observation (PTO), SMVs are measured as the difference between work as done and work as planned: SMV = WAD − WAP. Every difference is something worth learning from, and the classes below decide what kind of learning it is.

SMVs provide valuable insight into the real-world execution of safety procedures and help organisations understand the gap between documented processes and operational reality. In ISO 45001 aligned Occupational Health and Safety (OH&S) systems, analysing these variations supports learning, corrective action, and continual improvement.

The SMV Equation

SMV = WAD − WAP. Work as planned (WAP) is the documented intention: the approved method, its risk assessment, permits and sequence. Work as done (WAD) is what actually happens under live conditions: the crew, tools, weather, pressures and judgement of the day. The SMV is the difference between the two.

The equation has three properties worth internalising. First, there is always a reading. Conditions shift and crews adapt, so every observed task returns a difference of some kind, and the equation resolves to one of the four classes below, never to nothing.

Second, the size of the difference carries no sign. A large variation can be the best finding of the year or the worst, and a small one can hide a bridged interlock. The four classes below are how the difference gets its sign, and the risk assessment is how the sign is read.

Third, the equation only resolves under observation. WAP can be read at a desk; WAD exists only in the field while the task runs, which is why the planned task observation, watching a full task cycle in real time, is the instrument of measurement.

Every handled SMV closes its difference in one of two directions: a validated positive moves the plan to match the work, and a corrected negative moves the work back to match the plan. Either way the goal is agreement, and the honest expectation is that new differences will keep arriving.

The As-Built of the Task

Construction has run this discipline for decades at project level. As-built drawings record what was actually constructed, as against what the design drawings intended, because everyone accepts that built reality drifts from drawn intention, and that all future work on the asset depends on knowing the reality rather than the intention.

The SMV is the same discipline applied at task and activity level. Where the as-built captures the difference between design and structure, the SMV captures the difference between method and practice, and for the same reason: everything that follows, training, planning, the next revision of the procedure, depends on the true record rather than the intended one.

The analogy also carries the right attitude. Nobody reads an as-built as an admission that the project failed; it is the honest record reality demanded. An SMV log deserves the same standing: the mature artefact of an organisation that knows work as done drifts from work as planned, and prefers to know by how much.

Categories of Safe Method Variations

Not all deviations carry the same level of risk. Safe Method Variations are classified according to their impact on safety and operational performance.

• Increased Opportunity (SMV-IO): the variation reveals an opportunity for a better system: safer, and better beyond this task, faster, more cost-efficient or newly capable, and the class most programmes never record.
• Enhanced Resilience (SMV-ER): the variation reinforces the current system: margin, capacity or an alternate path added at the same cost, the same task spent more safely.
• Increased Vulnerability (SMV-IV): the method creates localised risk: the worker or team becomes more vulnerable to harm, and the exposure is bounded by the task, largely ending with it.
• Decreased Reliability (SMV-DR): the harm goes systemic: a general control is degraded or a risky behaviour is taught, so the added risk survives the task and spreads elsewhere in the system.

The two positive classes mark where reality has overtaken the procedure; the two negative classes map where the approved method is losing its contest with pressure. Both pairs divide by reach: an SMV-ER reinforces the system as it stands while an SMV-IO points beyond it, just as an SMV-IV is bounded by the task while an SMV-DR outlives it.

In every case the class is read from the risk assessment rather than the rulebook: what the change does to exposure and controls decides it, never the motive behind it.

How to Handle an SMV

• Record it as found: capture what varied, the conditions around it, and the crew's reasons, which usually matter more than the variation itself
• Classify by effect: place it in one of the four classes based on what it does to exposure and controls
• Validate before judging: risk-assess the variation formally, with gains and losses confirmed under normal conditions rather than the best day anyone remembers
• Act by class: adopt and standardise validated positives, crediting the crew that found them; correct negatives, and fix the pressure that taught them
• Update the method: close every SMV by making the record and the work agree again, in whichever direction the evidence points

Handled this way, the variation record becomes something rare: an account of how work is actually done that people are willing to keep honest. A record with no positive findings is rarely describing a workforce with no good ideas; it is usually describing a programme that never asked.

Why Safe Method Variations Matter

Documenting SMVs allows organisations to treat deviations as learning opportunities rather than simply violations. This approach helps safety teams understand how work is actually performed.

• Identify practical safety improvements developed by frontline teams.
• Reveal where procedures are being bypassed or adapted and why.
• Improve training materials so they reflect real working conditions.
• Detect unsafe shortcuts before incidents occur.

This approach becomes particularly powerful when used alongside Planned Task Observations (PTOs), where inspectors observe an entire task process to compare planned methods with actual practice. Together, PTOs and SMVs shift safety management from reactive enforcement toward proactive learning.

SMVs and ISO 45001

ISO 45001:2018 promotes continual improvement through operational feedback and worker participation. SMVs provide a structured way to capture and analyse that feedback.

• Clause 5.4: Worker participation in hazard identification and decision making.
• Clause 6.1: Identification of hazards and evaluation of risk.
• Clause 10.2: Corrective action and learning from nonconformity.

By recording and analysing Safe Method Variations, organisations can convert informal workarounds into structured safety improvements while strengthening the effectiveness of their OH&S management system.