Cabin Programmes Rarely Fail All At Once. They Drift.
Ask anyone who has managed a cabin programme through a difficult delivery to point to the moment things went wrong, and you’ll usually get a specific answer: the First Article Inspection that failed, the test campaign that slipped, the six-week schedule compression before Entry Into Service. It’s a natural instinct — problems feel like they arrive at a moment.
In nearly thirty years around cabin and seating programmes, I’ve come to believe that instinct is almost always wrong. The moment where a problem becomes visible is not the moment it began. By the time a delay reaches a status report in red, it has usually been building, quietly, for months.
What drift actually looks like
Drift rarely announces itself. It looks like a series of individually reasonable decisions, each made under real time pressure, each defensible on its own terms.
A tolerance gets relaxed because the originally specified value is proving hard to hold in production, and the engineering team judges the risk acceptable. A test gets rescheduled two weeks because a test rig is occupied by another programme, and two weeks feels absorbable. An action item from a design review gets marked “in progress” for the third consecutive month because the person who owns it is now also covering a colleague’s workload.
None of these, in isolation, sinks a programme. I’ve watched programmes absorb a dozen decisions exactly like this and still deliver successfully. What determines the outcome isn’t any single decision — it’s whether anyone is tracking the pattern across time, and whether that pattern is allowed to compound unchecked.
A pattern I’ve seen repeat
I’ve worked programmes where every formal milestone review told a genuinely accurate, genuinely positive story: on schedule, on budget, open items tracking down. And the programme still arrived at a difficult, compressed final phase before Entry Into Service.
The reviews weren’t wrong. They were measuring the state of the programme at a single point in time, against the plan as it stood at that point in time. What they couldn’t capture — because no single review can — was the accumulation: three small schedule absorptions that individually looked fine, but that together had quietly consumed the contingency the programme was counting on.
By the time that became visible, the only tools left were expensive ones: expedited freight, compressed inspection windows, weekend shifts. All of it worked, in the sense that the aircraft entered service. None of it needed to cost what it cost, if the drift had been visible six months earlier.
Why formal reviews structurally miss drift
This isn’t a criticism of programme reviews, which do exactly what they’re designed to do: confirm status against plan at a scheduled point. The limitation is structural, not a failure of anyone’s diligence.
A review happens perhaps once a quarter, sometimes less often for a specific workstream. Drift, by definition, is a slow accumulation across the interval between reviews — the exact period a formal review cannot see, because it isn’t there to see it. Add to that the natural tendency for status reporting to filter for what’s presentable: nobody walks into a milestone review leading with the three small decisions that felt uncomfortable to make. They get folded into “on track,” because individually, they were defensible calls.
The only way to see drift while it’s still cheap to correct is to be present in the interval — not at another formal gate, but informally, regularly, close enough to the work to notice the pattern before it needs a formal review to force it into visibility.
What this means in practice
I don’t think the answer is more meetings, or heavier reporting requirements on suppliers who are already stretched. More process tends to generate more polished status, not more honest status.
What I’ve seen work is presence with continuity: someone who is at the supplier often enough, informally enough, to understand what “normal” looks like there — and who has enough history with the programme to recognise when a new decision is the fourth in a pattern, not an isolated call. That combination — regular presence plus continuity of context — is what turns a sequence of small, reasonable decisions into a visible trend, early enough that correcting it costs almost nothing.
It also requires a certain willingness to raise something before it’s obviously a problem, which is uncomfortable. Nobody wants to escalate a decision that, viewed alone, looks entirely reasonable. But that discomfort is exactly the price of catching drift while it’s still cheap.
The conclusion I keep coming back to
Every difficult programme I’ve been close to shares the same retrospective shape: not a single failure, but a sequence of small, rational decisions that nobody connected until it was expensive to unwind them. The individual people involved were rarely at fault. The structure of oversight — reviews at fixed points, gaps in between — simply wasn’t built to see the pattern.
I don’t think that’s a solvable problem through better reporting templates. I think it’s solvable through presence: someone watching the interval, not just the milestones, with enough continuity to see the shape of a pattern rather than a single snapshot.
Looking back at a programme that went sideways on you — can you trace the individual, reasonable decisions that, in sequence, defined the outcome? And at what point could someone, watching closely enough, have actually seen it coming?