Charter operations run on promises measured in minutes: a wheels-up time, a crew duty window, a catering truck at the right ramp. When something unplanned breaks, whether that is a booking platform glitch, a crew-scheduling tool failing or a passenger-communication system going dark, the damage is not only technical. It is a trust problem, and it arrives fast. Technology teams have developed a body of habits for these moments under the informal heading of IncidentOps, and several popular beliefs about it are simply wrong. Here are six, with the corrections and what to do about them in a flight-operations setting.
The technical claims here come from the Ops Atlas, a learning-oriented catalogue of operations disciplines, whose IncidentOps entry has a section on misconceptions. We paraphrase and apply them to flight operations by analogy. Procedures approved by your operator and regulator always come first; nothing here replaces them.
Myth 1: “Rank decides who takes charge”
The chief pilot or operations director seems the obvious choice. The reality in well-regarded incident frameworks is more nuanced. One widely published process makes the coordinator whoever has the highest rank on the call, regardless of their daily role, and then insists that the role be rotated so that many people are trained. The Atlas treats seniority-by-default as a myth because the job requires skills (keeping a calm picture, delegating, speaking clearly) that are learned by practice, not conferred by a title.
In a charter dispatch room, that means your best coordinator on a Saturday night may be a duty controller with four years of experience, and the director’s job is to support, not take over. If the director does arrive, the polite and effective question is, “Do you wish to take command?”, which forces a clear yes or no.
Myth 2: “Raising the alarm is an admission of failure”
Many crews feel an instinct to wait and see, hoping the issue fades. The Atlas flags the opposite: punishing teams for the number of incidents they declare simply teaches them to declare fewer, which hides problems until they are bigger. A written trigger helps: another team is required, customers feel the effect, or an hour passes with no answer. Calling it early and finding it small costs minutes. Waiting costs hours.
Pilots already understand this. A go-around is not a confession; it is the normal safe response to an unstable approach. Declaring is the go-around of operations.
Myth 3: “Nobody should touch anything until the cause is known”
When a crew scheduling feed breaks, there is a pull to understand everything first. The Atlas argues that plenty of effective responses depend only on knowing where the fault sits; nobody needs a theory of how it got there. Undo the last change, steer users to a backup, switch a feature off, or reduce load. A real example in the Atlas concerns a cluster-creation failure that ran for more than six hours; the point of the account is that once the faulty component was located, rolling back could have restored service hours sooner than the team’s investigation did.
On the ramp you do the same. If the primary fuel-ordering portal is down, you place the order by phone and continue; the post-mortem on why the portal failed can wait until the aircraft has left.
Myth 4: “A person made a mistake, so that is the cause”
This one is stubborn because it sounds like an answer. The Atlas quotes a familiar safety-science remark: once a bad outcome is known, hindsight distorts how we judge the people involved: looking backwards, the mistake looks obvious, but at the time it did not. A famous 2017 outage at a large cloud provider began when an operator mistyped an input to a routine playbook and removed more capacity than intended. The valuable lessons were about the system: why the tool allowed such a large removal, and why restart took so long. Stopping at “typo” would have taught nothing.
Aviation reached this conclusion decades ago. Investigators ask what made the slip easy, which checks failed to catch it, and which design could prevent it next time. Apply the same instinct to your own incidents: when your review says “dispatcher error”, ask what the screen, the form or the handover made likely.
Myth 5: “A long task list proves a thorough debrief”
A debrief finishing with twenty-five tasks looks impressive. In practice it creates a to-do list that nobody finishes. The Atlas describes problem patterns such as vague actions, actions that are only “mostly mitigative”, equal priority for everything, and actions closed by quietly downgrading them. It recommends few actions, one owner and one date each, and an explicit deadline target. One large software firm sets its own deadline for priority actions, either four or eight weeks, chosen by how important the service is.
A charter operator can mirror this: choose three actions, give each a name and a date, and track whether they close on time. Count completed actions as a measure of learning, not the length of the list.
Myth 6: “Everyone should join the call”
When a sizeable disruption happens, everyone wants to help, and in a modern workplace they can all join the chat. The Atlas reports one case in which forty-one people were in the chat for a single cloud incident, and cites field research on cases with hundreds of participants, where the core team retreated into private channels to get work done. A crowd does not accelerate a fix; it fragments the record.
The remedy is two spaces. A small working channel holds the coordinator, the fixers, the voice and the note-taker. A second channel carries short updates so that everyone else, such as sales, ground handlers and customer service, can stay informed without interrupting. The same design suits a busy operations centre during irregular operations: a quiet floor for the people solving the problem and a visible board for everyone else.
Looking for the technical side of aircraft systems? Our explainer on how the communication equipment aboard a jet works and our overview of what a flight management system does on board are good companions, since on-board systems and ground operations both rely on clear, disciplined communication.
Why these six beliefs survive
Each of them protects people’s feelings or sense of order at the expense of speed and learning. Seniority feels safe, silence feels modest, root-cause hunting feels responsible, blaming a person feels final, long lists feel diligent, and crowds feel supportive. The better habits are less comfortable at first: rotate the role, declare early, act on the location of the fault, blame the system conditions, keep the list short, and keep the room small.
A short scenario to test yourself
Try this on your team. It is early evening and the tool that sends itinerary updates to charter customers has stopped sending. Nobody knows why. Ask: who takes the coordinator hat, and does it matter that she is a junior controller? What written trigger lets her declare the incident now instead of in an hour? Which generic move gets itineraries flowing again without any understanding of the cause, perhaps a manual send from a backup mailbox? What goes in the first customer message, and who posts it? If the director walks in, what is the sentence that clarifies who is in charge? And at the review, what two or three actions with owners will actually be finished by the next quarter?
If your team can answer all six in a couple of minutes each, the myths have little purchase on you. If the answers stall, you have found the best place to start practising, and a thirty-minute table-top session will close most of the gaps.
Where to read more about IncidentOps
The Atlas entry is a long document, but the sections on roles, communication rhythm and review practice are easy to read on their own. You can find it as a reference page that catalogues IncidentOps myths and good practice. Skim the myths section first, then come back to your own operation with a notebook. Write down which of the six your team currently believes, pick one, and plan a small experiment this month. The first time a duty controller calmly declares an incident, assigns hats and posts a first update inside five minutes, you will see the difference at once.

