Debriefing Your Own Flight: A Method for When No One Else Will

A pilot in the cockpit at dusk, reviewing flight data in quiet reflection before leaving the aircraft

Four sectors, last rotation, end of the evening. The go-around on the second sector was never discussed, either in the moment or afterward; the usual goodbyes were exchanged and the car park emptied within minutes. In training, that go-around would have filled the debrief on its own. On the line, it will produce nothing — unless you do something with it yourself, within a window of time far shorter than most people assume. What follows describes a self-debriefing framework that fits into a few minutes at the gate and another twenty minutes later, and above all what it has to compensate for if it is to be worth more than a memory.

Three Things Your Instructor Provided Without Saying So

A training debrief is not a conversation: it is a mechanism, and it supplies three things you do not have on your own.

The first is external observation. The instructor saw what you did; you remember what you meant to do. The gap between the two is exactly the raw material of a debrief, and it is invisible from the seat, where action and intention blur together. That is why « I debriefed myself mentally on the descent » produces almost nothing: the film you replay is the one of the plan, not the one of the conduct.

The second is an explicit reference standard. In training, the expected manoeuvre is described, scored, and the deviation is measured. On the line, the reference still exists — standard operating procedures, stabilisation criteria, task-sharing, the operator’s standard — but nobody pulls it out of the drawer at the end of an ordinary day. Self-debriefing largely consists of going to fetch it yourself, which means knowing where it is written down and in what form it is binding.

The third is an obligation to verbalise. Saying out loud, « at that moment, I was waiting for the speed to come back, » turns an impression into a verifiable — and therefore contestable — statement. As long as the reasoning stays in your head, it is immune to scrutiny: it quietly adjusts itself to what you now know. Writing serves the same function as speaking, provided you write full sentences rather than keywords.

A fourth thing is added, more mundane: a protected slot of time. In training, the debrief is on the schedule. On the line, it competes with the trip home, rest, and personal life — and it loses, unless it is short and structured.

What each mechanism supplies, and what self-debriefing must manufacture on its own
Element Training Debrief Line Crew Debrief Self-Debrief
External observation Continuous and scored Partial, limited to what the other pilot saw Absent: must be replaced by time-stamped facts captured in the moment
Reference standard Brought out and cited Implicit, rarely reopened Must be looked up yourself in the documentation
Verbalisation Imposed by the format Optional, often rushed through Must be imposed by writing in complete sentences
Time slot Scheduled and protected A few minutes at the gate Must be reserved, or it disappears
Dominant bias Examiner’s authority Politeness and hierarchical relationship Hindsight: the outcome justifies the decision
Record Training file None, apart from the report Personal logbook, with no particular protection

A Boundary to Set Before Writing Anything

A self-debrief is not an occurrence report and never replaces one. Whatever falls under mandatory reporting goes into the operator’s system, within the required timeframe and in the required form; a note in a logbook discharges no obligation and protects no one. The order is therefore fixed: the report first, if one is due, and only then the personal work.

The boundary also matters in the other direction, and it is the point people discover too late. The protections attached to a report — confidentiality of the source, handling under just-culture principles, information compartmentalisation — apply to the reporting system, not to your logbook. A personal logbook is neither confidential nor destroyed after a set period; it lives in a bag, a drawer, sometimes a synced cloud folder.

This observation has a very clear practical consequence for how you write. What serves progress is the mechanism: the configuration set up late after accepting a shortcut, the decision deferred while waiting for a confirmation that never came, the call made on the descent and never followed up. The mechanism can be written without naming anyone, without a registration, without a flight number, without an exact date, and it loses nothing of its instructional value through that anonymisation. What is useless, on the other hand, is the detailed account with names: it teaches nothing more and turns a tool for progress into an exhibit.

The trap: the logbook that narrates instead of describing. A named, dated, detailed note duplicates a report without any of its protections, and a second version of the facts written down later always ends up diverging from the first on some detail — time, order of calls, wording. Describe the mechanism, never the case.

Your personal notes should focus on the process, not the people. Write about what happened in the flight (like a late configuration) rather than who did what (like “Captain Smith forgot to check”). This keeps your notes useful for learning without creating legal risks.

Capture Hot, Analyse Cold: Two Separate Operations

The most costly confusion is doing both at once. In the heat of the moment, analysis is worthless: you are still inside the action, you are less interested in understanding than in concluding, and you produce a verdict — « I was bad, » « that had plenty of margin » — which closes the file before it has even been opened. Once things have cooled down, on the other hand, the raw material is gone.

What fades fastest is precisely what is useful: the order of events, durations, the exact wording of a call, what was displayed, what the hands were doing. What does not fade, but is instead manufactured, is coherence. The narrative repairs itself, acquires a cause, sheds its hesitations and false leads, and ends up forming a tidy story that no longer contains anything usable. The curve of raw material falls while the curve of the good story rises; a useful debrief sits before the two curves cross.

Decay of factual detail and rise of the reconstructed narrative after the flight Two curves plotted against time elapsed since the end of the flight. The solid line, retrievable factual detail, starts very high and falls rapidly in the first few minutes, then more slowly. The dashed line, reconstructed narrative, starts low and rises steadily. The two cross in the middle of the graph. A light band on the left marks the window during which capturing the facts is still possible. recall quality factual detail (solid) crossover reconstructed narrative (dashed) capture window time elapsed since the flight
Raw material falls while the story rises. Usable facts — order of events, durations, the exact wording of a call — fade within minutes, while the narrative gains coherence and sheds its hesitations. Capture before the crossover, analyse afterward.

Hence a split into two stages with opposite requirements. Capture happens at the gate, or in cruise if the next sector leaves a quiet moment: five factual lines, without judgment — phase, what was perceived, what was done, what was surprising. It must be sparse enough to survive end-of-duty fatigue. Analysis happens later, in a calm setting, with documentation at hand, and lasts only as long as the capture deserves. The ratio between the two durations is heavily skewed, and that is normal: you capture often, you analyse rarely.

The right reflex: note the surprise, not the fault. « I didn’t expect to be given descent there » is a usable fact; « I should have anticipated it » is already a conclusion, and it prevents you from ever recovering the information you actually had at the time. Surprise is the best marker of a gap between your mental model and the situation, and it is the only one you can capture from the inside.

On an Airbus A320 with a 2,000m runway at 30°C, you notice during the approach that the flaps were set to 2 instead of 3. You capture this at the gate with: “Approach, saw flaps 2 instead of 3, no call made, noticed at 1,000ft.” Later, you analyze why you didn’t correct it – was it because you were focused on the unstable approach, or because you assumed the other pilot would catch it? This captures the key facts before memory fades.

Threat, Error, Undesired State: A Grammar, Not a Set of Boxes

The threat and error management framework is taught everywhere and, most of the time, used as a nomenclature: the event gets filed into a box and everyone moves on. Its value lies elsewhere. It is the branch points that carry the information, not the boxes.

A threat gets through or it doesn’t; an error is detected or it isn’t; an undesired state is recognised as such or absorbed without ever being named. At each node, the question is not « what happened » but « what made it branch. » And a distinction is worth drawing every time: was the barrier that worked designed — a cross-check, a standard call, an alarm, a parameter monitored by design — or fortuitous: a glance at the right moment, a question asked by chance, a reminder from ATC?

The two produce the same outcome and have nothing in common. The first will happen again; the second will not, and counting it as a success amounts to crediting a mechanism that does not exist. A debrief that records only outcomes fails to distinguish between them — and that is precisely the work the collective debrief was doing without anyone noticing, by forcing each person to say what they saw and when.

The threat, error, undesired-state chain and its branch points Three boxes aligned left to right labelled threat, error, and undesired state, linked by horizontal arrows annotated unmanaged then undetected. Below each box, a dashed arrow points down to a single band representing the exit from the chain, obtained either through a designed barrier or through luck. A note at the bottom recalls that the debrief is about the branch points. threat error undesired state unmanaged undetected managed detected recovered exit from the chain: designed barrier — or luck the debrief is about the branch points
It is the branch points that carry the information, not the boxes. At each node, the useful question is not what happened but what took the chain off course: a designed mechanism, or a lucky break. Both give the same outcome and do not predict the same future.

A corollary often overlooked: a flight without an error is not a flight without material. Threats managed without error are actually the richest seam of all, because they show which barriers genuinely hold on the line, not just in the simulator. Debriefing only the bad days amounts to studying only the failures of a system whose successes remain a mystery.

Reconstruct the Decision Point, Don’t Replay the Film

A single question does most of the work: at what moment was the information available, and at what moment was the decision made? The gap between these two instants is the object of the debrief. It is rarely zero, it is measured in seconds rather than instants, and it is far more instructive than the decision itself.

Reconstruction requires putting yourself back in the information state of the moment: what was displayed, announced, requested, and what was not yet. The right question is therefore not « should I have gone around sooner, » which reasons from the outcome, but « at that precise point, what did I have in front of me, and what was I waiting for. » The answers are few and remarkably repetitive: you were waiting for a confirmation that never came, you were waiting for the other pilot to speak, you were betting on a trend you had no proof of, or you had filed the situation under a category it did not belong to.

Gap between the moment information becomes available and the moment the decision is made Side view of an approach. A straight flight path descends from the left toward the runway threshold shown at the bottom right. A first point marks the moment the information needed for the decision becomes available; a second point, noticeably further along the path, marks the moment the decision is actually made. Two opposing arrows on the ground represent the gap between the two instants. A vertical dashed line represents the stabilisation gate, located between the two points. approach profile information available decision made stabilisation gate threshold gap to be explained
The gap between information and decision is the object of the debrief. The question is not whether the decision was good — the outcome will always answer that — but what you were waiting for during the seconds that separate the two points.

The same exercise applies to what went well. The moment you decided early deserves to be reconstructed with the same care, because it points to the cue that worked — and an identified cue can be reused, whereas a good day cannot be learned from directly.

The trap: debriefing the outcome. A flight that ends well retroactively validates every decision; a flight that ends badly condemns all of them; it is the same bias in both directions. A simple, unforgiving test: if your conclusion could not have been written before you knew how it ended, it is made of hindsight, not analysis.

Twenty Debriefs Are Worth More Than the Same One Twenty Times

A single debrief is worth almost nothing. Its value appears on rereading, the day the fifth note starts to resemble the second one. That is when the mechanism stops being an exercise in self-awareness and becomes an instrument: it brings out recurrences that no one else can see, because no one else follows you across twenty consecutive flights.

The sorting axes that pay off the most are few: flight phase, point in the duty period, role held, configuration or aircraft type, and the nature of the other party involved. An item that recurs several times within a quarter is no longer a bad day: it is a personal characteristic, and it is the only thing that deserves a work plan. Recurrences shared by many people, on the other hand, are usefully discussed on the FL330 forum, where they stop being an individual problem.

Sorting still has to be done, because not all recurrences are equal, and the temptation is to work on the one that is annoying rather than the one that is costly. Two axes are enough: frequency and potential consequence.

Sorting personal recurrences by frequency and potential consequence A two-axis matrix. Frequency of recurrence on the horizontal axis, potential consequence on the vertical axis. Two dashed lines divide the space into four named zones: address first, top right; keep an eye on, top left; automate, bottom right; let it go, bottom left. Three example points are plotted: a late configuration, frequent and serious; a missed cross-check, rare but serious; a forgotten call, frequent and minor. potential consequence address first keep an eye on automate let it go late configuration missed cross-check forgotten call frequency of recurrence
Work on what costs you, not what annoys you. A frequent, serious recurrence comes before everything else; a frequent, minor recurrence is fixed with a personal routine rather than with extra attentiveness, which never holds up.

The right reflex: schedule a rereading session, not a writing session. Writing is already prompted by the flight itself, which provides the occasion; rereading is prompted by nothing and is the first thing to disappear. A monthly half-hour of rereading is enough, as a rule of thumb, to surface the recurrences — and to discard, without regret, the notes that led nowhere.

Running Someone Else’s Debrief, and Checking Your Own Ground Rules

The same framework carries over to the right-hand seat, with one inversion: the person running the debrief must not supply the chain, they must draw it out. A debrief where the instructor states the threats, errors, and corrections themselves is a lecture; it trains a pilot capable of receiving a debrief, not of running one. The proportion of talk time is the best available indicator, and the easiest to measure on yourself.

Two pilots in a cockpit review flight logs and diagrams, one writing in a notebook.
Self-debriefing improves flight performance and safety.
Distribution of talk time depending on how the debrief is run Two horizontal bars of equal length, compared. The first, labelled lecture-style debrief, is occupied for the greater part by the person running it, leaving only a small portion to the person who flew. The second, labelled guided self-debrief, reverses the split: the person who flew occupies the greater part, and the person running it speaks only for a small portion. A dashed line marks the division within each bar. lecture-style debrief guided self-debrief the person running it the person who flew the person who flew the person running it share of talk time
The proportion of talk time gives away the nature of the debrief. A rule of thumb, not a measurement: as long as the person who flew talks less than the person debriefing them, they are being handed content, not a method.

What remains is what needs to be checked at your own operator, and it cannot be guessed. The operations manual states what must be reported, to whom, and within what timeframe, and what need not be: the boundary between a report and a personal note is written there, not in an article. The operator’s safety policy states what becomes of a report and whether any feedback reaches crews — a flight data monitoring programme that sends nothing back deprives self-debriefing of its only source of external observation, and that is worth asking about. The training department, finally, may have a view on keeping a personal log, particularly for anyone preparing an upgrade to the left seat or an additional type rating: at that stage, a documented quarterly review carries more weight than a speech on situational awareness, and it is built months in advance, like the rest of the file. The grey area that remains concerns the status of these notes in the event of an investigation: it is not settled the same way everywhere, which is one more reason to write anonymous mechanisms rather than dated cases.

This article describes a personal working method and has no regulatory standing. The applicable regulation, the instructions of the competent authority, and the operator’s approved documentation — operations manual, occurrence reporting policy, training manual — prevail in all circumstances over what is written here.

Questions fréquentes

How much time does this need to take to be worthwhile?

Allow a few minutes per flying day for the capture, and roughly twenty minutes a week for the analysis. The capture must stay sparse enough to survive end-of-duty fatigue: five factual lines are enough, and they are more useful than three paragraphs written once. It is the rereading, half an hour a month, that gives the whole process its value, because it brings out recurrences that no single note can show on its own.

What should I do when I no longer remember the exact order of events?

Don’t reconstruct it. An order rebuilt after the fact is a product of the narrative, not of memory, and it leads to false conclusions with great confidence. Note what you are sure of, explicitly flag what is uncertain, and draw a conclusion of correspondingly limited scope. An honestly incomplete note is worth more than an invented chronology.

Should flights where nothing happened be debriefed too?

Yes, and they are often the most instructive. A threat anticipated and managed without error shows which barrier genuinely works on the line, something a difficult flight never tells you. Limiting yourself to the bad days amounts to studying only the failures of a system whose successes remain a mystery, and never learning which reflexes to keep.

Can a personal logbook be used against me?

Confidentiality protections attach to the operator’s reporting system and to the mechanisms provided for under regulation, not to a private logbook, whose status in the event of an investigation is not handled the same way everywhere. The safeguard lies in how you write: describe the mechanism, with no name, no registration, no flight number, and no exact date. An anonymised mechanism keeps its full instructional value and stops being evidence.

How do I debrief alone an event where the other pilot played a decisive role?

By sticking to what you controlled: the information you had, what you announced, when, and what you were waiting for. The other pilot’s conduct is not your object of study and you do not have the material for it — you have access to neither their intentions nor what they perceived. The effectiveness of your own calls, on the other hand, debriefs very well on your own.

Does this replace the crew debrief at the gate?

No, one does not substitute for the other. The crew debrief supplies external observation, which is precisely what self-debriefing cannot manufacture; it remains the most reliable and least costly source. Self-debriefing takes over for everything that was not said at the gate, which is the vast majority of flights.

Commentaires

Leave a Reply

Découvrez l’écosystème FL330

FL330, c’est toute l’aviation professionnelle — explorez nos autres espaces.