Job Safety Analysis (JSA) Template and Guide
Most workplace injuries do not come from exotic, unforeseeable events. They come from routine tasks that nobody stopped to break down — a lift performed the same way for years until the day a back gives out, a guard removed for a "quick" adjustment, a confined space entered without anyone naming the atmospheric hazard out loud. A Job Safety Analysis (JSA) exists to interrupt that pattern by examining a task step by step before a worker performs it, and naming the hazards while there is still time to control them.
This guide gives you a JSA template you can use immediately, walks through the five-step method behind it, and shows five completed example analyses across construction, manufacturing, and warehouse work. By the end you will be able to produce a JSA that holds up to a supervisor's review and an OSHA inspection alike.
Turn your JSAs into a living safety system with WhyTrace Plus. A JSA is only useful if hazards get tracked, controls get verified, and the next incident gets connected back to the analysis that should have caught it. See how WhyTrace Plus links JSAs to investigations and corrective actions →
What a Job Safety Analysis (JSA) Is
A Job Safety Analysis is a structured technique that breaks a job into its sequential steps, identifies the hazards associated with each step, and determines the controls that eliminate or reduce those hazards before the work begins. OSHA uses the term Job Hazard Analysis (JHA) for the same technique; the two names are interchangeable in practice, with JSA more common in construction and JHA in OSHA's own guidance.
The method focuses on the relationship between four elements: the worker, the task, the tools, and the work environment. When those elements fall out of alignment — a worker in an awkward position, a tool used outside its design, a task performed in poor lighting — hazards appear. The JSA makes that relationship visible on paper before it becomes an injury.
According to OSHA's foundational guidance, Publication 3071 ("Job Hazard Analysis," originally issued in 2002 and periodically revised), a JSA centers on job tasks as the unit of analysis rather than on broad job titles or generic workplace conditions. That task focus is what distinguishes it from a general risk assessment.
A few distinctions worth holding:
| Term | Scope | When used |
|---|---|---|
| JSA / JHA | Single task or job, step by step | Before performing a defined job |
| Risk assessment | Broader activity, area, or process | Strategic hazard planning |
| Pre-task plan / SWMS | Task-level, often crew-facing daily | Construction, high-risk activities |
| Method statement | Detailed work procedure with controls | Complex or contracted work |
Which Jobs to Analyze First
You cannot write a JSA for every task in an operation at once, so the method requires prioritization based on injury potential and task characteristics. OSHA Publication 3071 directs employers to start with jobs that carry the highest consequences or the lowest tolerance for error.
Prioritize these categories first:
- Jobs with potential for severe or disabling injury — even when there is no history of past incidents. Absence of a prior accident is not evidence of safety.
- Jobs where a single human error could trigger a serious accident — tasks with no second chance, such as energized electrical work or crane lifts over occupied areas.
- Jobs new to your operation or recently changed — new equipment, revised procedures, or altered materials introduce hazards that established habits have not yet accounted for.
- Jobs with a history of incidents or near misses — your own incident data points directly at where analysis is overdue.
- Jobs complex enough to require written instructions — if the task already needs a procedure, it needs a hazard analysis to go with it.
A practical sequencing rule: rank candidate jobs by a simple product of injury severity and likelihood, then work down the list. Tasks that score high on both deserve a JSA before anything routine and low-consequence. This is the same logic behind a risk matrix for prioritization, applied at the task level.
The Five-Step JSA Method
The JSA method is a five-step sequence: select the job, break it into steps, identify the hazards in each step, determine controls, and review the analysis with the workers who perform the job. Each step has a specific output, and skipping any one of them produces a document that looks complete but fails in use.
Step 1: Select the job
Choose a defined task with a clear beginning and end — "replace a circular saw blade," not "carpentry." A job too broad cannot be broken into meaningful steps; a job too narrow produces a JSA that nobody will read. Aim for a task a worker performs as a recognizable unit of work.
Step 2: Break the job into steps
List the job's basic steps in the order they occur. A useful rule of thumb is that most jobs decompose into roughly 10 steps or fewer; if you exceed that, the job is probably too broad and should be split. Each step describes what is done, not how — "position the ladder," not "carefully and safely set up the ladder." Observe the job being performed if you can, rather than writing the steps from memory.
Step 3: Identify the hazards in each step
For every step, ask what could go wrong. Useful prompts: Could the worker be struck by or against something? Caught in, on, or between objects? Could they fall, overexert, or contact a harmful substance, temperature, or energy source? Name the hazard specifically — "blade contact with hand during removal," not "cut hazard."
Step 4: Determine preventive measures
For each hazard, define the control, working down the hierarchy of controls in order of effectiveness:
| Control level | Effectiveness | Example |
|---|---|---|
| Elimination | Highest | Remove the task or hazard entirely |
| Substitution | High | Replace with a less hazardous method or material |
| Engineering | High | Guards, ventilation, interlocks, isolation |
| Administrative | Moderate | Procedures, training, permits, signage |
| PPE | Lowest | Gloves, respirators, fall harnesses |
Always reach for the highest feasible control first. PPE is a last line of defense, not a substitute for engineering out the hazard.
Step 5: Review with workers
The workers who perform the job will spot steps and hazards you missed and will reject controls that are unworkable in the field. Reviewing the draft with them improves accuracy and builds the ownership that determines whether the JSA gets followed. Date the review and re-review when the job, equipment, or environment changes — or after any incident involving the task.
From paper JSA to closed-loop control. WhyTrace Plus stores your JSAs alongside the incidents and near misses that touch each task, so when something goes wrong the analysis is right there — and your corrective actions update the JSA automatically. Start free with WhyTrace Plus →
The JSA Template
A JSA template is a three-column working document — job step, potential hazards, recommended controls — anchored by a header that captures the job, the people involved, and the review dates. Below is a template you can copy directly into a document or spreadsheet.
Header fields:
| Field | Entry |
|---|---|
| Job / task title | _______________________ |
| Department / location | _______________________ |
| Analyzed by | _______________________ |
| Required PPE | _______________________ |
| Date prepared / reviewed | _______________________ |
| Reviewed with crew (name/date) | _______________________ |
Analysis table:
| # | Job step (what is done) | Potential hazards | Controls (hierarchy order) |
|---|---|---|---|
| 1 | |||
| 2 | |||
| 3 | |||
| ... |
Keep the language plain and the controls specific. A control that reads "be careful" is not a control. Every control should describe an action, a device, or a verifiable condition: "lockout breaker #14 and verify zero energy with meter," not "make sure power is off."
Five Worked JSA Examples
The fastest way to write a strong JSA is to study completed ones. Below are five examples spanning common high-risk tasks. Each shows a few representative steps rather than the full analysis, to illustrate how hazards and controls are phrased.
Example 1: Replacing a circular saw blade (manufacturing / carpentry)
| Step | Hazards | Controls |
|---|---|---|
| Unplug saw and confirm de-energized | Accidental start during blade change | Unplug at source; do not rely on switch (engineering/administrative) |
| Loosen and remove arbor nut | Blade contact laceration; pinched fingers | Use blade wrench with blade lock engaged; cut-resistant gloves (engineering + PPE) |
| Install new blade and tighten | Blade installed backward; loose nut on startup | Verify rotation arrow; torque to spec; test run away from body (administrative) |
Example 2: Working from a portable ladder (construction)
| Step | Hazards | Controls |
|---|---|---|
| Position and set up ladder | Ladder slip; contact with overhead power lines | Inspect ladder; 4:1 angle; secure footing; maintain 10 ft clearance from lines (administrative) |
| Climb to work height | Fall from height; overreaching | Maintain three points of contact; keep belt buckle within rails; no top-two-rungs (administrative) |
| Perform overhead task | Dropped tools striking workers below | Tool lanyards; barricade area below (engineering + administrative) |
Example 3: Manual lifting of materials (warehouse)
| Step | Hazards | Controls |
|---|---|---|
| Assess load before lifting | Overexertion; unexpected weight | Test weight; team lift or mechanical aid above set threshold (substitution/administrative) |
| Lift from floor to waist | Lower-back strain; awkward posture | Bend at knees, load close to body; use lift table where available (engineering + administrative) |
| Carry and place load | Struck-by from blocked vision; crushed fingers | Clear travel path; place hands clear of pinch points (administrative) |
Example 4: Forklift operation in aisles (logistics)
| Step | Hazards | Controls |
|---|---|---|
| Pre-operation inspection | Operating defective forklift | Complete daily checklist; tag out defects (administrative) |
| Travel through aisles | Pedestrian struck-by; tip-over | Segregated pedestrian routes; horn at intersections; speed limit posted (engineering + administrative) |
| Lift and place pallet at height | Load fall; rack collision | Verify capacity plate; spotter for high placements (administrative) |
Example 5: Confined space entry (general industry)
| Step | Hazards | Controls |
|---|---|---|
| Prepare for entry | Hazardous atmosphere; engulfment | Permit required; isolate and lock out lines; atmospheric testing before entry (engineering + administrative) |
| Enter space | Oxygen deficiency; toxic gas | Continuous monitoring; forced-air ventilation; retrieval harness (engineering + PPE) |
| Conduct work and exit | Attendant unaware of distress | Trained attendant stationed outside; communication protocol; rescue plan (administrative) |
These five share a pattern worth internalizing: hazards are named as specific mechanisms of harm, and controls climb the hierarchy rather than defaulting to "wear PPE and be careful." For task-specific deep dives, see our guides on construction safety and warehouse safety.
Common JSA Mistakes That Make Them Useless
The most common JSA failures are not gaps in the template — they are habits that turn the document into a compliance artifact nobody uses. Recognizing them is how you keep a JSA program from quietly dying.
- Vague hazards. "Cut hazard" or "fall hazard" without the mechanism gives the worker nothing actionable. Name how the harm happens.
- PPE-only controls. Reaching straight for gloves and glasses skips the more effective controls above PPE in the hierarchy. PPE is the last layer, not the first.
- Writing it once and filing it. A JSA that is never re-reviewed after equipment changes, process changes, or incidents becomes a record of how the job used to be done.
- No worker involvement. A JSA written at a desk and never reviewed with the crew will miss real steps and propose controls that don't survive contact with the field.
- Steps that are too broad. "Do the maintenance" is not a step. If a step hides multiple actions, it hides multiple hazards.
- Disconnection from incidents. When a near miss or injury occurs on an analyzed task and nobody updates the JSA, the analysis has failed at the one job it exists to do.
That last point is where JSAs most often break down. A JSA predicts hazards; an incident reveals which predictions were wrong or incomplete. Connecting the two — so every relevant incident triggers a JSA review — is what makes the program self-correcting. This is the same closed-loop discipline behind corrective action management: the analysis is only as good as the verification that follows it.
Frequently Asked Questions
Q. What is the difference between a JSA and a JHA?
There is no meaningful difference. Job Safety Analysis (JSA) and Job Hazard Analysis (JHA) describe the same technique — breaking a job into steps, identifying hazards, and determining controls. OSHA's own guidance uses "Job Hazard Analysis," while "JSA" is more common in construction and many corporate safety programs. Use whichever term your organization standardizes on, and apply the same five-step method either way.
Q. Does OSHA require a JSA?
OSHA does not have a single standard mandating a JSA for every job, but it strongly recommends the technique in Publication 3071 and effectively requires hazard assessment under several standards — most notably the PPE hazard assessment requirement in 29 CFR 1910.132 and the permit-required confined space program in 29 CFR 1910.146. A JSA is the practical method most employers use to satisfy these hazard-identification obligations, and inspectors view documented JSAs as strong evidence of a functioning safety program.
Q. How often should a JSA be reviewed and updated?
Review a JSA whenever the job, its tools, materials, or environment changes, and immediately after any incident or near miss involving the task. Many organizations also set a recurring review cycle — annually for stable jobs, more often for high-risk ones — to keep the analysis current even when no obvious change has occurred. The trigger that matters most is an incident on the analyzed task: that is direct evidence the JSA needs revisiting.
Q. Who should write the JSA?
A JSA is best written collaboratively. A supervisor or safety coordinator typically drafts it, but the workers who actually perform the job must review it before it is finalized. They identify steps and hazards an outside observer misses and flag controls that are unworkable in practice. A JSA written without worker input tends to be inaccurate and ignored.
Q. How many steps should a JSA have?
Most jobs break down into roughly 10 steps or fewer. If your analysis runs well past that, the job is probably too broadly defined and should be split into separate JSAs. Conversely, a JSA with only two or three steps may be glossing over distinct actions — and the hazards hiding inside them.
Key Takeaways
- A JSA breaks a job into steps, identifies the hazards in each, and determines controls before the work begins. JSA and JHA are the same technique under different names.
- Prioritize jobs by injury potential first: severe-consequence tasks, single-error tasks, new or changed jobs, and tasks with incident history come before routine work.
- Follow the five-step method — select, break down, identify hazards, determine controls, review with workers — and don't skip the worker review.
- Phrase hazards as specific mechanisms of harm and choose controls from the top of the hierarchy down; PPE is the last layer, not the first.
- A JSA only keeps its value if it stays connected to your incident data — every relevant near miss or injury should trigger a review of the analysis that should have caught it.
Related Resources
| Resource | Description | Best For |
|---|---|---|
| Construction Safety: Investigation and Prevention | Task-level hazards and controls for high-risk construction work | Building JSAs for ladders, scaffolds, and lifts |
| Warehouse Safety: Hazards and Controls | Forklift, lifting, and storage hazards with practical controls | JSAs for logistics and distribution tasks |
| Risk Matrix Prioritization | How to rank tasks by severity and likelihood | Deciding which jobs to analyze first |
Once your JSAs surface a recurring hazard pattern, the next step is structured cause analysis. For preventing field harm before it happens, pair your JSA program with proactive AI-driven hazard prediction and KY (kiken-yochi) activity support (AnzenAI), and build a near-miss reporting pipeline with frontline safety reporting and 4M analysis tooling (AnzenPost Plus). For comparing investigation and EHS platforms, the manufacturing safety and quality tool guides (GenbaCompass) cover selection criteria in depth.
Sources: