Treadstone Associates
Case File · Property Management Operations

A compliance calendar catches a missed alarm test

Anonymised, illustrative composite. A property management firm tracked fire-alarm testing, WSIB clearances, elevator inspections and backflow testing across six buildings on one compliance calendar. The calendar had an entry for every building. What it didn’t have, for one of them, was a service report confirming the test had actually happened.

Treadstone Associates · Updated 2026

At a glance

  • • Ontario property management firm, six-building light-industrial and commercial portfolio.
  • • Compliance calendar tracked roughly 9 recurring items per building — about 54 tracked line items portfolio-wide.
  • • A quarterly cross-check, run separately from trusting the calendar itself, found one building’s fire-alarm service contract had lapsed.
  • • The vendor’s most recent service report could not be located — 1 of 54 tracked items (about 2%) without a confirming record on file.
  • • Closing the gap required one emergency vendor callout at $650 to reinstate coverage and get a fresh test on record the same week.

The situation

A compliance calendar is only as good as the assumption that an entry on it means the underlying thing actually happened. This portfolio’s calendar had a line for fire-alarm testing at every one of its six buildings, alongside WSIB clearance renewals, elevator inspection due dates, and backflow-preventer testing — roughly nine recurring items per building, 54 in total. Every entry looked current. What the coordinator’s quarterly cross-check actually does, deliberately, is not trust that.

The calendar itself had not failed in any visible way. Its fire-alarm line for the affected building showed the same recurring cadence as every other building’s, with a next-due date that looked entirely ordinary next to the other five. Nothing about the entry itself gave any indication that the vendor behind it had quietly stopped servicing the account — that fact only existed in a service report nobody had gone looking for.

The problem

The specific interval a fire-alarm system must be tested is set by the applicable fire code and the equipment’s own service standard — not by occupational health and safety law, and not by anything with a freely published figure this file can cite. That is worth stating plainly rather than guessing at a number: confirm the applicable testing frequency for a given building with the local fire authority or a licensed fire-protection contractor, because it is not a figure to assume from memory or from another jurisdiction.

What the cross-check found instead was a gap in the record, not necessarily in the equipment: the vendor servicing one building’s fire-alarm system could not produce its most recent service report, and nobody at the firm could independently confirm the date of the last completed test. CCOHS makes a related point about a different kind of safety record, but the logic transfers directly: “employers and supervisors should still keep records of safety talks, which can help demonstrate due diligence when they occur regularly.” A calendar entry that says a test is scheduled is not the same thing as a report confirming it happened — and the gap between those two only shows up if something checks for the second one specifically.

The numbers

One of 54 tracked items — about 2% of the portfolio’s compliance calendar — came back without a confirming service record at the time of the cross-check. Reinstating the vendor’s coverage and getting a fresh, documented test scheduled cost one emergency callout at $650, resolved inside the same week the gap was found.

The rule that decided it

Nothing here turns on a specific testing interval, because none is sourced in this file to state as fact. What decides an exposure like this is whether an employer can show a documented, regularly-kept record — not a reconstructed claim that testing “usually happens.” The stakes for getting that wrong are real: OHSA sets fines of up to $2,000,000 for a corporation and up to $1,500,000 for directors and officers, rising to a $500,000 minimum fine on a second or subsequent offence resulting in death or serious injury within a two-year period. A due-diligence defence rests on documented regularity, and a calendar entry with no underlying report behind it is not that.

The outcome

The vendor contract was renewed with an auto-renewal clause and a calendar reminder set 30 days ahead of the next expiry. The firm added a distinct quarterly step — confirm the service report itself is on file, not just that the calendar has an entry — on top of the calendar it already ran. See how a formal safety-audit standard like COR is scored, and for a related portfolio-wide compliance gap, see how the same firm found a lapsed vendor insurance certificate.

For how firms are using AI tools to keep incident and compliance records consistent, see AI-assisted incident and compliance record-keeping.

Takeaways

  • • A calendar entry documents that something is scheduled — it does not, by itself, confirm that the underlying test actually happened. Check for the report, not just the entry.
  • • Specific life-safety testing intervals are set by the applicable fire code and equipment standard, not by OHSA; confirm them locally rather than assuming a figure.
  • • Due diligence rests on documented regularity, not a reconstructed claim that a process usually runs on schedule — the record is what a due-diligence defence actually consists of.
  • • A small percentage of a large tracked portfolio — here, 1 of 54 items — is exactly the kind of gap that a periodic, independent cross-check catches and a passive calendar does not.
  • • A vendor going quiet mid-contract, without formally cancelling, can look identical to a vendor still servicing the account — until someone specifically asks for the service report.
  • • Run the cross-check as a distinct, scheduled exercise separate from day-to-day calendar maintenance, so it is not skipped simply because the calendar looked fine on its own.

See where AI pays off first in your business.

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