Matrix ratings are not reproducible
Without written descriptors, two competent people rate the same task differently. Defined bands make the score mean something.
A task-level risk assessment with a scoring method written out in full — so two assessors working the same task arrive at the same rating, and the reasonably-practicable judgement is on the page rather than in someone's head.
Instant download · Editable .docx
Most risk assessment templates hand you a five-by-five matrix and leave the definitions to intuition. The result is that "likely" and "moderate" mean whatever the assessor felt that day, ratings are not comparable between jobs, and the residual score cannot be defended when it matters. This template writes the likelihood and consequence descriptors out in full, requires the control level to be declared against the Regulation 6 hierarchy, and includes a written reasonably-practicable test — the element that actually carries the HSWA s36 duty.
Without written descriptors, two competent people rate the same task differently. Defined bands make the score mean something.
Dropping the rating after listing a control is not an assessment. The template requires the control level and the reason the residual score changed.
Section 36 turns on what was reasonably practicable. The template has a structured test — likelihood, degree of harm, what was known, availability and suitability of controls, and cost relative to risk.
This is the task-level assessment. For an ongoing record of hazards and controls, or a business-level risk view, use the registers.
Word — 14 pages (method + assessment forms + worked examples) · Formats: DOCX
Risk assessment is the mechanism by which the risk-management duty is discharged. Each section of the template maps to the part of the duty it evidences.
Written and reviewed by the occupational health and safety team at OH Consultant. The consequence descriptors are defined by health outcome rather than generic severity language, which matters for occupational health risks where harm accrues over years rather than in a single event — the category most safety-led risk assessments systematically under-rate.