Show Notes
Maintenance Is a Temporary Operating Mode
Planned maintenance is intended to reduce risk, but a service window can create disruption when the building is treated as a collection of independent systems rather than a shared operating environment. In this episode of Built, Wired & Secured, Alex Morgan speaks with Michael Harrington and James Rogers about the maintenance window problem: why a dashboard can be green while the building is still not ready for people.
The discussion begins with an early-morning scenario. A service team reports that work is complete, but the first employee’s badge is rejected and the lobby temperature begins to climb. The technical task may have been performed correctly, yet the building is not operating as occupants expect. That distinction is central to the conversation: equipment health is not the same as building readiness.
A maintenance window is defined as a period when a known change or service activity is permitted to affect normal operations under agreed conditions, with named people responsible for decisions. It is not simply a technician’s appointment. It is a temporary operating mode for the building, and someone must own that mode.
Why One Green Status Is Not Enough
Facilities, IT, security, and property operations each protect different outcomes:
- Facilities may focus on equipment state and environmental conditions.
- IT may focus on service availability.
- Security may focus on controlled access and alarm status.
- Property operations may focus on whether people can use the building as expected.
None of these perspectives is wrong. The risk appears when a team assumes that one healthy component or green dashboard confirms the full occupant-facing outcome. A controller may be reachable while an access sequence is wrong. A system can look healthy while a temperature schedule has not returned to normal. A technical restart can succeed while the person operating the building cannot complete an expected workflow.
Start With Dependencies and Decision Ownership
Before work starts, teams should ask what else could notice or be affected by the change. The conversation identifies several useful signals that a maintenance item should not be treated as a narrow work order:
- A shared service could be affected.
- An occupant-facing function could be affected.
- Recovery is uncertain.
If any of those conditions exists, the maintenance window needs a broader operating plan. That plan should identify who can proceed, who can pause the work, and who communicates if the expected condition changes. The authority to declare a window incomplete should be named before the work begins, not left to whoever first notices a problem.
The episode also emphasizes the importance of shift handoffs. A night engineer may understand the temporary operating condition, but the next shift can inherit a building that appears normal until a sequence, schedule, or system behavior changes. The clock matters, but the building’s actual rhythm matters more. Deliveries, cleaning, early shifts, patient care, research activity, attendance startup, and other real operating needs may shrink the recovery margin far more than the calendar suggests.
Handle Incomplete Documentation Honestly
Documentation is valuable, but the guests caution against treating perfect documentation as a prerequisite for every safe decision. When records are incomplete, uncertainty should be acknowledged and managed rather than ignored.
- Narrow the scope of the work.
- Increase observation during the change.
- Schedule discovery activity before attempting a broader change.
- Delay the work when an unknown could create a serious operating consequence.
- Make the uncertainty part of the approval decision.
Uncertainty does not disappear because it is inconvenient. It changes the conditions for proceeding. Someone with operational authority needs to understand the unknown and accept the consequence. Notification can tell people that work is planned, but it cannot substitute for ownership.
Think in Operating States, Not Task Names
Rather than turning routine work into an oversized procedure, the episode recommends thinking through operating states. Teams should define what must be normal before the window, what will intentionally be unavailable during it, what temporary condition is acceptable, what must happen first, what can happen in parallel, and what must wait until stability is confirmed.
The conversation also challenges the idea that “rollback” is automatically a plan. Teams need to know what condition they are returning to, how long recovery could take, and who has the authority to decide that continuing is less safe than stopping. A nominal finish time is not enough if the recovery margin overlaps with the next shift or the first occupants of the day.
Prove Recovery in Occupant-Facing Terms
A maintenance window is not complete when a task is marked complete. It is complete when the intended operating condition has returned, relevant behavior has been observed, and the people responsible for the building agree it is ready for the next operating period.
For a building opening at 6:00, useful recovery evidence could include:
- An authorized badge test.
- Confirmation that the expected temperature range is returning.
- Confirmation that alarm status is normal.
- A named person informing the next shift that the building is ready.
Occupants do not experience a successful restart as a technical accomplishment. They experience whether the door opens, whether the workplace is comfortable, whether a scheduled activity can begin, and whether someone can provide a credible answer when conditions change.
A Lightweight Coordination Framework
The episode closes with a practical framework that supports, rather than replaces, formal change control and documented approvals. Before work, ask what else could be affected, who owns the decision to proceed or pause, and what evidence will prove recovery. During work, keep the actual condition visible. Afterward, verify the experience, communicate if the plan changes, and capture what should improve next time.
Reliable maintenance is disciplined maintenance. The goal is not paperwork for its own sake; it is a short, focused conversation that exposes decisions people would otherwise make under pressure. Before the next service window, ask not only whether the technician finished the task, but whether the building is ready for the people who depend on it.
The Maintenance Window Problem Is Not Just a Scheduling Problem
Planned maintenance should make a building more reliable. Yet a poorly coordinated service window can interrupt access control, affect environmental conditions, create alarm uncertainty, or leave operations teams unsure whether the building is truly ready for people.
The problem is often not that a technician failed to complete a task. The uncomfortable reality is that the technical work may be completed correctly while the building is still not ready to operate. A dashboard can be green while a badge is rejected. A controller can be reachable while a schedule has not returned as expected. A system can report healthy while occupants arrive to an uncomfortable space.
That is the maintenance window problem: treating a temporary change as a narrow work order instead of a temporary operating mode for the entire building.
A Maintenance Window Needs an Owner
A maintenance window is more than a time on a calendar or a technician appointment. It is a defined period in which a known service activity is allowed to affect normal operations under agreed conditions, with named people responsible for decisions.
That definition matters because connected buildings depend on multiple teams protecting different outcomes. Facilities may be watching equipment state. IT may be watching service availability. Security may be watching access control and alarm status. Property operations may be focused on whether tenants, employees, visitors, and building staff can use the property as expected.
Each view is valid. The failure occurs when one team’s view is treated as the whole picture. A green status does not automatically mean the building is ready.
The first question is therefore simple: who has the authority to say that the window is complete, or that it needs to stop? The answer should be the person named before the work begins, not whoever happens to notice the first symptom. Without a named decision owner, the building is operating without a decision plan.
Follow the Building’s Rhythm, Not Just the Clock
A service window can look safe on paper and still provide too little time for recovery. Consider work scheduled from 4:30 to 5:30 for a building that opens at 6:00. The calendar may suggest a 30-minute buffer, but the actual recovery margin can be much smaller if the morning lead arrives at 5:15, the night engineer leaves before the work is complete, or early occupants begin arriving at 6:00.
The building’s rhythm matters more than the timestamp. Overnight work may overlap with deliveries, cleaning, early shifts, patient care, research activity, attendance startup, or other operational dependencies. Sunday mornings and holidays may also be critical periods, even when a conventional calendar suggests otherwise.
Before work begins, teams should ask what else could notice the change. Could the work affect access, alarms, environmental conditions, shared communications, or a tenant activity at the start of the day? If it could, the work should not be treated as a narrow technical task.
Use Three Signals to Expand the Conversation
A short coordination conversation is often enough to surface the risks that matter. Three signals indicate that a maintenance item deserves broader planning:
- A shared service may be affected.
- An occupant-facing function may be affected.
- Recovery is uncertain.
Any one of those signals should prompt the team to identify who can proceed, who can pause the work, and who communicates when the expected condition changes.
This is not an argument for unnecessary bureaucracy. It is an argument for making decision-making visible before people have to make those decisions under pressure. Formal change control and documented approvals remain essential in many environments. A lightweight coordination framework should support those practices, not replace them.
Incomplete Documentation Changes the Conditions for Proceeding
Many organizations have diagrams, work orders, and system records, but not a current picture of every dependency. Documentation remains valuable, but waiting for perfect records can become an excuse to avoid needed maintenance. At the same time, unknowns cannot simply be ignored because work is scheduled.
The practical response is to treat uncertainty as part of the decision. If records are incomplete, teams can narrow the scope, increase observation, schedule discovery activity, or delay a broader change when an unknown could create a serious operating consequence.
The important point is that uncertainty changes the conditions for proceeding. Someone with operational authority needs to understand the unknown and accept its consequence. A notification that work is planned does not fulfill that responsibility. Notification informs people; it does not create ownership.
Plan in Operating States
Teams do not need a 50-page procedure for every service activity. They do need clarity about the operating states around the work.
Before the window, define what must be normal. During the window, identify what is intentionally unavailable and what temporary condition is acceptable. Establish what must happen first, what can happen in parallel, and what must wait until stability has been confirmed.
Those questions convert a list of technical tasks into an operating plan. They also expose false confidence in finish times. A planned 5:30 completion may be irrelevant if the team needs additional time to validate access, temperatures, alarms, communications, or the handoff to the next shift.
Rollback deserves the same discipline. “We can roll back” is not a complete plan unless the team knows what condition it is returning to, how long that recovery could take, and who can decide that continuing is less safe than stopping.
Completion Means Recovery Has Been Demonstrated
A task marked complete confirms that an activity occurred. It does not prove that the building is ready. A complete maintenance window requires evidence that the intended operating condition has returned, that relevant behavior has been observed, and that the people accountable for the next operating period agree the building is ready.
For an early-morning opening, recovery evidence might include an authorized badge test, confirmation that the expected temperature range is returning, confirmation that alarm status is normal, and a named person communicating readiness to the next shift.
These are meaningful checks because they measure the experience people actually depend on. Tenants do not experience a successful restart as a technical achievement. They experience whether the door opens, whether the workplace is comfortable, whether a scheduled activity can begin, and whether someone can give them a credible update if the plan changes.
Communicate Changes and Preserve the Lesson
Several coordination failures are easy to recognize. One is late notice: affected groups learn about work after staffing or tenant communication decisions are already set. Another is missing stop authority: everyone assumes someone else can pause the work while symptoms accumulate. A third is missing follow-up: the building appears normal, so the team moves on without recording that recovery took longer, a handoff was unclear, or a check was skipped.
Preventive maintenance is most valuable when its lessons improve the next maintenance window. After work, record what changed, what took longer, what was observed, and what needs follow-up. If a temporary condition remains or the window runs long, provide a timely, specific update. People can work with uncertainty more easily when they know who is managing it.
Make Quiet Operations the Standard
The best day in building operations is often the one nobody notices. That result is not accidental. It comes from reviewing dependencies, naming decision owners, agreeing on recovery evidence, and making the actual condition visible throughout the work.
Before your next maintenance window, do not ask only whether the technician completed the task. Ask whether the building is ready for the people who depend on it. Bring facilities, IT, security, contractor, safety, and property operations voices into the plan as appropriate. Ask what else could be affected, who can proceed or pause the work, who communicates if conditions change, and what evidence will prove recovery.
For more practical conversations about the systems that keep commercial properties operating, listen to this episode of Built, Wired & Secured.