A single average close-time number blends triage speed, scheduling speed and vendor performance into one figure that cannot tell you which of the three is actually broken.
Key takeaways
“Average time to close a work order” is the metric almost every portfolio starts with, and it is the least useful one on its own, because it blends three stages that fail for completely different reasons: time from submission to acknowledgment (a triage and staffing problem), time from acknowledgment to a vendor being dispatched (a scheduling and vendor-capacity problem), and time from dispatch to verified completion (a parts, access, or vendor-performance problem). A building that looks slow on the blended number but fast on two of the three stages has a much narrower, much cheaper problem to fix than the blended number suggests.
Speed metrics get the attention, but a work order that closes fast and reopens within two weeks for the same issue has not actually been fixed — it has been counted twice while still costing the building the underlying problem. First-time-fix rate (the share of work orders resolved without a return visit for the same fault) and rework rate (the inverse, tracked over a fixed window such as 30 days) are better predictors of total maintenance spend than closure speed alone, because a fast, incomplete fix generates a second work order, a second dispatch, and often a second trip charge that never shows up against the original ticket’s reported closure time.
In a condominium, some repair work orders are not simply a maintenance cost — they are chargeback-eligible against a specific unit owner, and getting that classification right at intake saves a dispute later. Where an owner, tenant or guest causes damage through an act or omission, Condo Act ss.92 and 105(2) authorize the corporation to charge the owner “the cost of the repair or the condo corporation’s insurance deductible limit — whichever is less” Filing that work order under a generic “repair” category instead of a tracked “chargeback” field means the corporation has to reconstruct the paper trail after the fact, when a clean intake tag would have preserved it automatically.
Rework has a compliance dimension worth tagging too, not just a cost one. A work order that traces back to a vendor’s own faulty repair — rather than a new, unrelated failure — is potentially a claim against that vendor’s own commercial general liability coverage rather than a cost the building simply absorbs, since CGL is designed to cover claims that a vendor’s own operations caused property damage, which a defective repair can fall within depending on the policy wording Flagging rework work orders by suspected cause at intake preserves the option to pursue that recovery later, instead of the connection being lost once the ticket is closed and filed under a generic repair category.
Reactive work-order metrics tell only half the story if they are not read alongside preventive maintenance (PM) compliance — the share of scheduled PM tasks actually completed on time. A building with a falling reactive-work-order volume and a rising PM compliance rate is genuinely getting ahead of failures; a building with falling reactive volume and falling PM compliance is more likely simply deferring work that will show up as a spike later. Statistics Canada’s own description of what a property manager administers — contracts for property services including cleaning, maintenance and security — is a useful reminder that maintenance administration is core to the role being measured, not a side function the metrics can ignore.
Work-order volume is not evenly distributed across the year, and comparing a slow month to a busy month without adjusting for ticket count produces the same distorted read as comparing utility costs across buildings without adjusting for climate. A building that opens 40 tickets in a heavy month and closes 90% of them within target is often running a tighter operation than one that opens 15 tickets in a quiet month and closes 95% within target — the first number represents genuinely more work absorbed at a similar standard. Reporting a completion-rate percentage without the underlying volume next to it strips out exactly the context needed to judge whether a good-looking number reflects a genuinely well-run period or simply a quiet one.
Not every metric worth tracking belongs on the same cadence. A weekly operational dashboard is best kept to a handful of numbers a manager can act on immediately: open ticket count by stage (not just total open), tickets aged past their target in each stage, and any chargeback-eligible work still missing its documentation. First-time-fix rate, rework rate and PM compliance are better read monthly, since a single week’s sample is usually too small to mean much and reacting to weekly noise in those numbers tends to produce process churn rather than real improvement. The goal of the dashboard is to catch a stalled ticket before a tenant complains about it, not to relitigate the whole maintenance program every Monday morning.
Worked example — a falling average that was actually two problems (illustrative)
A 60-unit building’s average work-order close time drops from 6 days to 3.5 days over a quarter — a result that looks like a clear operational win on the summary dashboard.
Splitting the clock tells a different story: acknowledgment and dispatch speed genuinely improved, but first-time-fix rate fell from 88% to 71% over the same period. Tickets are being closed faster because they are being closed before the underlying issue is actually resolved, then reopened days later as a new ticket that resets the clock.
The blended close-time metric never would have surfaced this — each reopened ticket looks like a fast new resolution, not evidence of a rework problem. Tracking first-time-fix rate alongside close time is what catches a building trading real fixes for a better-looking summary number.
Related reading: the same split-the-clock approach applied to unit turnover, scoring the vendors generating the rework and setting a work-order baseline before a manager transition.
Because it blends three different stages — acknowledgment, dispatch and completion — that each fail for different reasons. A building can look slow overall while actually having a narrow, cheap-to-fix problem in just one stage.
The share of work orders resolved without a return visit for the same issue. It is a better predictor of total maintenance cost than closure speed, because a fast but incomplete fix just generates a second, often uncounted, ticket.
Sections 92 and 105(2) cap it at the lesser of the actual repair cost or the corporation’s insurance deductible — not the full repair cost by default.
Yes. Reading the two together is the only way to tell whether falling reactive volume reflects genuine improvement or simply deferred work that will surface later as a spike.
A 30-minute call is enough to tell you whether AI pays for itself here.
