It is a Sunday night when the message arrives: the online portal for a hangar-side property, where tenants check leases, submit repair requests and pay fees, has stopped responding. A tenant is locked out of the access-code page, another cannot reach the maintenance form, and an owner is asking whether her paperwork is safe. Anyone who manages property or aviation facilities knows that moment of scattered questions. The questions below are the ones people most often ask when they first meet IncidentOps, the organised way of handling an unplanned service disruption, and the answers are written for the aviation-and-property world. This is general information, not legal or financial guidance.
What is IncidentOps in one honest sentence?
It is a set of habits for the period between “something is wrong” and “we have learned from it”: declaring the problem, assigning roles, communicating on a rhythm, applying a quick relief measure, and reviewing afterwards. One caveat comes from the reference entry itself. The Ops Atlas entry (read it here for the full outline of IncidentOps) labels the term a casual nickname for incident management, not a field with its own accreditation. The habits are the valuable part.
When does a glitch become an incident?
When a written rule says so. Technology operators tend to use three tripwires. Help from a second team is needed, outsiders can see the damage, or an hour has gone by with no fix. For an estate or aviation property office, the equivalent might be a tenant-facing outage, a vendor needing to be called, or a problem lingering beyond the first hour of a working day.
People worry that declaring too soon looks dramatic. The numbers favour speed. A flag that proves small costs a few minutes; a late one can cost hours of confusion. A story from a much larger world illustrates it: in a published cloud case, a cluster-creation failure lasted more than six hours with forty-one people in chat, and the review judged that an earlier rollback would have helped long before noon.
Who should be in charge when the portal goes down?
Not necessarily the most senior person. The Atlas points out that the commander is a hat, not a title, and that the role should go to someone trained to coordinate, with the job explicitly not being to repair the fault. In a property business, the managing partner might be the best voice for owners, while a trained duty coordinator runs the response and a technician makes the changes.
The people with permission to alter the live system should be a deliberately small group. Once a coordinator would have more than roughly five direct contacts, carve the group into sub-teams, each with a lead of its own. A separate person writes the messages, and another records what was tried and when. For a firm that prides itself on a personal touch, our piece on delivering a personalised client experience in upscale property describes the tone that these messages should carry.
What do we tell tenants and owners, and how often?
Quickly, plainly and on a schedule. Published guidance sets a tempo. Send the first note around minute five, add a note on scope soon after, and then refresh at least once every twenty minutes until two hours have gone by. The sign-off should say plainly whether any records disappeared. One open-source response guide adds a line worth printing: clear is better than concise.
Public status pages usually apply four labels in order: investigating, identified, monitoring, resolved. Tenants who see those words understand where you are without a phone call.
Where should the status page live?
Away from the system it describes. A famous 2017 storage outage showed that a service-health dashboard hosted on the failing service itself was unusable for part of the morning. The same logic applies to your own paging route: it should not depend on the portal that is down. If the status notice and the emergency contact path share a fate with the portal, you lose both at once.
Do we have to know what went wrong before we act?
No. This is the question people find most surprising. Quick, reversible moves (undo the latest update, point visitors at a healthy copy, turn off something new, lighten the load) call for a rough sense of where the trouble lies and nothing more. In a property context that could mean putting up a temporary form, taking requests by phone or switching to a standby page while the technician investigates. The reason comes later, in the review.
What does a fair review look like?
It starts with a named owner before the response call ends and takes place within days. Someone who sat out the response chairs the meeting, which keeps the talk candid. What leaves the room is a brief set of tasks, every one with a single name and a single date beside it. Three traps are common. Some tasks only dull the symptom, some are too fuzzy to ever be called finished, and some get “closed” by being demoted to low priority.
The Atlas cites a safety researcher who observed that hindsight biases judgements of human performance. That is a gentle caution against writing “staff error” and moving on. Ask what conditions made the error likely, then change those. The article on how complex property decisions can be simplified shares a similar spirit: reduce the number of moving parts that can go wrong.
How can we tell whether the routine is paying off?
With care. The common metric for mean time to repair can refer to at least four different phases, so always say which one you mean. Incident lengths are lopsided: a few long events dominate. Its worked sample covers eleven events. The average time to mitigate came out above an hour and a quarter while the typical case took barely half an hour, and removing one severe event would have cut the average by three-quarters of an hour. A better approach is to report separate distributions for detecting, acknowledging, mitigating and resolving, plus the share of review actions completed on time.
How many people does an on-call rota need?
More bodies than a small office usually has to spare. Imagine six people sharing two full-day slots every week. Each ends up on duty roughly one-third of the time, while big operators regard about one-quarter as the most a person can sustain. That works out at about eight people. For a small estate office, this is a good argument for sharing a rota with a partner firm or using a simpler schedule that does not rely on a few overworked individuals.
Is there a sensible place to begin on Monday?
Yes, and it is smaller than people fear. You do not need new software, a consultant or a budget line. You need an afternoon, a printed page and the willingness of two or three colleagues to try it once before judging it.
One page of triggers, a short list of hats written on cards, two message channels (one for responders, one for onlookers), a status page hosted elsewhere, and one thirty-minute practice session using a made-up portal failure. Add a spoken handover for long incidents: the outgoing coordinator says, “You are now in command, okay?” and waits for a yes. Everything else can wait until that routine feels natural.
That Sunday-night message will arrive eventually. When it does, the best outcome is not a flawless fix but a calm team that knows who is deciding, who is talking to tenants, and when the next update goes out.

