A tenant on eleven emails the property manager at 4:40 on a Friday: the men's room on their floor has a leak under the sink. The manager is on a lease call. The email is read at 6:10, forwarded to the engineer at 6:12, and the engineer sees it Monday at 7. By then the tenant has told three colleagues that nothing happens when you report something in this building.

Nothing in that story is a hard problem. No part of it requires a better plumber. It is a routing failure, and routing is what work order management for commercial property is for.

The anatomy of a request that closes on time

Watch a request that goes well and it has the same five stages every time: it is captured with enough detail to act on, routed to whoever owns that trade, accepted by a human being, worked, and closed with the requester told. Skip any one and the request does not fail immediately. It just gets slower, and slowness is what tenants experience.

Most buildings have all five stages. What they lack is a place where the stages happen in sequence without a person carrying the request from one to the next by memory.

Capture: the detail you get is the detail you asked for

A request that says "it's cold" costs two visits. A request that says "Suite 1104, north side conference room, cold since Monday, thermostat reads 68" costs one.

The difference is the form. A free-text box gets free-text answers. A short structured intake, issue type, location, and one description field, gets you an engineer who arrives with the right part. Three fields, chosen well, are worth more than a fourteen-field ticket nobody completes.

The other half of capture is where it happens. If the fastest route for the tenant is an email to the manager they already know, they will email the manager, and every routing rule you built is bypassed on the first request of the day.

Routing: the step buildings skip

Routing is the difference between a request system and a to-do list. It means a plumbing issue in the Office Block reaches the plumbing vendor for that building, at that hour, without a manager choosing.

Three rules cover most of it: issue type to trade, building to vendor, and hour of day to whoever is covering. The fourth rule is the one that matters after hours, which is what happens when nobody accepts inside a set window.

An unaccepted request is an invisible request, and invisible requests are how buildings find out about problems from the tenant's lawyer.

Duplicates are useful information

Three tenants reporting the same leak is one problem with three people watching, and it should collapse into one dispatch with three subscribers.

Systems that cannot do this create a specific kind of waste: three engineers dispatched, three sets of notes, three closures, and a tenant on the fourth floor who never hears anything because their duplicate was deleted.

What good work order management looks like

Structured intake with three fields, not fourteen. Issue type, location, description. Everything else can be inferred or asked later.
Automatic routing by type, building and hour. With an explicit rule for what happens when nobody accepts inside the window.
Status the tenant can see without asking. The single largest source of "chasing" is a requester who has no way to know whether anything is happening.
Duplicates collapse and carry their subscribers. One dispatch, everyone told at closure.
Closure requires a note. "Completed" with no detail is how the same fault gets fixed four times in a year without anyone noticing it is the same fault.

The numbers worth watching, and the one worth ignoring

Two measures tell you almost everything: time to acceptance and time to closure. Acceptance is the one you can fix this month, because it is a process problem instead of a labour problem. Closure follows the trade, the part and the schedule.

Volume, on its own, is the number worth ignoring. A building with rising request volume may be failing, or it may finally have tenants who believe reporting something works. Read volume next to acceptance time or do not read it at all.

Recurring work is not the same work

A quarterly filter change and a burst pipe do not belong in the same queue with the same urgency, and a system that treats them identically will bury the pipe under the filters.

Planned work should generate itself on a schedule, carry its own procedure, and sit in a separate lane that gets worked when reactive volume allows. Buildings that mix the two end up doing reactive work only, then wondering why equipment fails early.

After hours, and the escalation nobody tests

Every building has an after-hours procedure. Fewer have one that has been tested since the last time the answering service changed hands.

The failure is rarely the first call. It is the second. The tenant reaches someone, someone reaches the on-call engineer, the on-call engineer does not pick up, and the process has no defined next step, so it ends with a voicemail and a building that finds out at seven in the morning.

Write the escalation as a chain with times attached: who is called first, how long before the second name is tried, who is woken at the third step and for which categories of fault. Water, smoke, no heat in January and a stuck lift belong in a different tier from a jammed door.

Then test it twice a year at an inconvenient hour. A procedure that has never been run at 2am is only a document. The test costs one uncomfortable evening and it is the only way to find out that the number on the sheet belongs to somebody who left in March.

The part nobody enjoys: closing the loop

Ask a property manager what tenants complain about and "nobody told me it was done" comes up more than the fix itself. Closure notification is the cheapest tenant experience improvement available to any building, and it is the one most often left switched off because someone worried about noise.

Send it. A tenant who is told their request closed will report the next thing they notice, which is the whole point.

One more habit is worth the ten minutes it costs. Once a week, read the requests that took longest to accept, not the ones that took longest to close. Closure time is usually a part on a truck somewhere. Acceptance time is almost always a rule that is wrong, a vendor who has stopped answering, or a category nobody owns, and all three are fixable by the person reading the list.

Where BuildingHub fits

Work orders in Operations Hub capture a structured request, route it by issue type and building, collapse duplicates into a single dispatch and keep the requester informed to closure, with preventive maintenance running in its own lane so planned work does not compete with a leak. If you want the arithmetic on what routing delay costs across a portfolio, we will work it through with you, and the difference between a portal and a platform is worth reading before any vendor demo.

The work order that closes on time: anatomy of a good request