Working alone in Canada: what the law requires when help isn't nearby

A service tech on an evening call. A driver on a lease road at 2 AM. A cleaner closing a building. Working alone is normal across Canadian industry — and it’s specifically regulated, because when something goes wrong with nobody there, response time is survival time.

What “working alone” means legally

The trigger isn’t solitude — it’s assistance not being readily available in an emergency, injury or illness. In Alberta that’s the scope of OHS Code Part 28 (ss. 393–394). A worker in a busy shop with coworkers in the next bay isn’t working alone; the same worker doing a Saturday call-out in an empty building is. So is a driver out of cell coverage, even though dispatch “knows where they are.”

The two core duties (Alberta Part 28)

1. Assess and control the hazards of the work being done alone — the ordinary hazard assessment duty, sharpened by the question: what changes because nobody can help? A task acceptable with a partner (confined space entry, some energized work, high-risk lifts) may simply be prohibited alone.

2. Establish an effective communication system — the heart of the rule. The employer must ensure contact between the worker and someone capable of responding, at intervals appropriate to the hazard. Three words carry the weight:

• Effective — a phone in a no-coverage area is not a system. Neither is “she usually texts when she’s done.”
• Intervals — proportionate to risk: a high-risk task might warrant contact every 15–30 minutes; lower-risk lone work might be hourly or at defined milestones.
• Someone who responds — a check-in nobody monitors is theatre. The monitor must notice a missed check-in and act on a defined escalation path.

Other provinces run the same architecture with local variations — BC notably requires a written lone-worker procedure with check-ins — and federal undertakings handle it inside the CLC hazard-prevention framework. Confirm your own jurisdiction’s text if you’re outside Alberta.

What an effective system looks like in practice

1. A policy that classifies lone work by risk and sets the check-in interval per class — 15/30/60 minutes, not “regularly.”
2. A named monitor per session — a person or an automated system, with after-hours coverage. The gap in most SMB systems is 6 PM to 6 AM.
3. An escalation ladder with clocks: missed check-in → attempt contact (minutes, not “soon”) → supervisor → emergency contact → emergency services with the worker’s last known location. Written down, tested.
4. Location awareness — the escalation is only useful if the responder knows where to send help: site, unit, vehicle, GPS where appropriate.
5. Records — sessions, check-ins, misses and escalations logged. After an event, “the system worked” is a claim; the log is the proof.

Manual versions of this (whiteboards, buddy texts) decay within weeks — they depend on the least interruptible person having an uninterrupted day. This is one of the clearest cases for automation in all of OHS: Vigil’s work-alone module schedules the check-ins by risk level, notices the misses, and walks the escalation ladder automatically — monitoring contact, supervisor, emergency contact, emergency services — while logging everything.

Questions that find your gaps

• Who, by name, was monitoring your lone workers last Saturday night?
• How many minutes pass between a missed check-in and the first phone call?
• Would the responder know where the worker physically is?
• Can you show a log of last month’s check-ins if an officer asks?

If any answer is “not sure,” Part 28 is telling you what to fix — and the fix is a system, not a reminder to be careful.