Treadstone Associates
Case File · Field Operations & Documentation

A stretched supervisor hands back two projects

Anonymised, illustrative composite. A hiring gap left one Ontario superintendent covering three concurrent sites instead of one. A worker’s stop-work refusal at the site he wasn’t at turned that arrangement from an inconvenience into a compliance problem.

Treadstone Associates · Updated 2026

At a glance

  • • Ontario GC, normally 3 superintendents across 5 concurrent mid-rise and light-commercial sites.
  • • A hiring gap left one superintendent covering 3 sites at once for roughly 6 weeks.
  • • A worker’s OHSA work refusal at Site B occurred while the superintendent was at Site A, about 35 minutes away in traffic.
  • • 95 minutes elapsed before the superintendent physically arrived to investigate — against OHSA s.43’s standard that the employer “must investigate the situation immediately, in the presence of the worker.”
  • • Two more refusals followed within five weeks at the other two sites this superintendent covered, with response times of 70 and 110 minutes.

The situation

The GC’s normal staffing model put one superintendent on each active site, sized to how much a person can actually be present for. A departure left a coverage gap that took longer to fill than planned, and for about six weeks one superintendent was formally responsible for three sites at once — not as a policy decision, but as a stopgap while recruiting caught up.

Nothing about the arrangement violated anything on paper. Each site still had a superintendent of record. What changed was how far that person actually was, physically, from any given site at any given moment.

The three sites were spread across the same metro area, close enough that a dispatcher looking at a map would call the coverage reasonable — typically 20 to 40 minutes apart depending on traffic. On a normal day, that spread never mattered: the superintendent moved between sites for scheduled walks and meetings, and a 30-minute gap between visits was simply part of the job. It only became a problem the moment something needed an immediate, in-person response and he happened to be at the wrong end of that spread.

The problem

OHSA s.43 gives a worker the right to refuse work believed unsafe, and puts a specific duty on the other side of that refusal: the worker must immediately tell the supervisor or employer, and the employer or supervisor “must investigate the situation immediately, in the presence of the worker” along with a worker-side representative — a JHSC member where possible. Until that investigation happens, the refusing worker “must remain in a safe place that is as near as reasonably possible to his or her workstation.” This is informally known as the first stage of a work refusal, and the statute does not carve out an exception for a supervisor who happens to be at a different site when it starts.

A worker at Site B refused a task over a condition he believed was unsafe. The superintendent responsible for Site B was at Site A. By the time he could reasonably drive over, 95 minutes had passed — a long way from “immediately” by any plain reading of the section, whatever the traffic.

The numbers

The affected crew of nine stood down for that 95-minute stretch; at a loaded rate of roughly $72 per hour, that is about $1,030 in idle labour for a single stoppage — the smaller cost, and not the one that mattered. Over the following five weeks, two more work refusals occurred at the other two sites this same superintendent covered, with response times of 70 and 110 minutes respectively. Three incidents, three delayed investigations, all traceable to the same root cause: one person physically covering ground that used to be covered by three.

The rule that decided it

The duty in s.43 attaches to presence, not to a job title or a headcount. Spreading one superintendent across three sites does not divide the statutory obligation into three thinner slices — each site still requires an immediate, in-person investigation the moment a refusal happens, in full, regardless of how many other sites that same person is also nominally responsible for. A superintendent covering three sites has not reduced the standard by two-thirds; he has simply made it structurally impossible to meet on two of them at any given moment.

The outcome

Following the third incident, the GC promoted a foreman into a fourth superintendent role and reassigned two of the three sites off the stretched supervisor, leaving each superintendent covering one or two sites within a short drive of each other. Response times on subsequent refusals fell to a typical 8–12 minutes, and no further stoppage has run anywhere near the length of the first. See how the work-refusal process itself unfolds and the underlying constructor-versus-employer duty split.

For a related case on how documentation carried an OHSA investigation, see how a site diary became the evidence that mattered.

Takeaways

  • • OHSA’s work-refusal investigation duty requires immediate, in-person presence — it does not scale down when one supervisor is covering more sites than usual.
  • • A staffing gap that looks manageable on a spreadsheet can be a compliance gap in the field the moment an incident lands at the site the supervisor is not currently at.
  • • Response-time data across repeated incidents, not just the first one, is what actually shows whether a coverage arrangement is workable.
  • • When a stopgap staffing arrangement is going to run more than a few days, price the response-time risk explicitly rather than treating it as a scheduling inconvenience.
  • • A coverage spread that looks reasonable for scheduled walks and meetings can still be unworkable for an event that requires an immediate, in-person response.
  • • Three incidents, not one, were needed before the pattern was treated as a staffing problem rather than a run of bad luck — track response times from the first one.

See where AI pays off first in your business.

A 30-minute call is enough to tell you whether AI pays for itself here.