On the evening of 6 July 1988 a Scottish electrician named Andy Mochan stood on a burning oil platform in the North Sea and decided to jump fifty metres into water that was also on fire, because spilled oil had ignited on the surface, and because staying meant certain death while jumping meant only probable death. He jumped and he lived, an American consultant named Daryl Conner later watched him tell the story on television, and in the early nineties Conner turned that night into the metaphor that has opened roughly every transformation programme in the western world ever since, the burning platform. What the board presentations reaching for that image almost never mention is that 167 people died on Piper Alpha that same night.
The story can therefore be told a second way, and the second way is the accurate one. It is the story of a single escape under the condition of certain death, in which the large majority of those present were killed, and we have made it the emblem of eighteen-month organizational development, of programmes with steering committees, mid-point reviews and a change story that gets rolled out again in module three. Whether a person can afford to change, however, depends less on what you explain to them than on the state they are in while you explain it, and the burning platform is a remarkably reliable way of ruining that state.
Fifteen announcements are not fifteen changes
For its Global Human Capital Trends 2026, Deloitte surveyed more than 9,000 business and HR leaders across 89 countries with Oxford Economics, and at first glance the finding reads like a capability problem: 85 percent say the ability to adapt at the speed the world now demands is critical, 74 percent report some form of progress, and exactly 7 percent believe they are leading in this area. You can read the gap between 85 and 7 as an execution gap, as an invitation to professionalise change management and finally consolidate the learning platform, and that is the reading the report itself invites and the one that will land in most boardrooms.
A few figures further into the same study sits a sentence that quietly dismantles that reading without meaning to contradict anything: one third of workers experienced fifteen major changes in the past year. Not three, not five, fifteen, and the consequences are declining well-being for 68 percent, increased workload for 60 percent, and, this is the number I got stuck on, a sense of being less relevant or left behind for 58 percent. An organization that announces something fifteen times in a year has not asked its people to change fifteen times, it has taught them fifteen times that announcements pass if you keep your head down long enough, and that every single one of them carries the possibility that you are the one about to become surplus.
Announcing change fifteen times does not train agility, it trains the art of ducking until the wave has passed.
The effect organizational research has known since 1981
Adaptability is not a skill but a state, and that distinction decides whether a training budget achieves anything or merely enlarges the catalogue. In 1981, Barry Staw, Lance Sandelands and Jane Dutton assembled the evidence in the Administrative Science Quarterly for an effect they called threat rigidity, and which occurs at three levels simultaneously: under perceived threat the individual restricts their information processing to what they already know, the group narrows its attention to the dominant response, and the organization constricts its decision paths by pulling control upwards. The paper has been cited for forty years, more than three and a half thousand times by now, and it states with complete composure the opposite of what every change story assumes, namely that threat produces rigidity rather than agility, and does so reliably.
Neurologically this is trivial and organizationally it is devastating, because a system whose people fall back on the familiar under pressure while control simultaneously migrates upwards will produce precisely the symptoms that appear in the Deloitte report as an adaptability deficit. The board sees people who are not moving and concludes that urgency is lacking, whereupon it increases the urgency, which deepens the narrowing, which appears to confirm the diagnosis. You only need to watch that loop once in a steering committee to never be able to unsee it, and the uncomfortable part is that every individual step in it looks entirely reasonable.
Companies try to manufacture adaptability through urgency, which means through perceived threat, because threat is the only instrument fast enough to break through indifference.
The price is that threat triggers the very narrowing that makes adaptation impossible, so the same measure that creates attention destroys the agility it was deployed to create.
The speculative neighbour you do not need
There is an older and far more dramatic version of this argument, and because it has crossed my desk repeatedly it belongs here, though with the handbrake engaged. Terror management theory claims that making our own mortality salient turns people more selfish, more materialistic, more deferential to authority and more hostile to outgroups, that under existential fear we cling to everything familiar. It is a wonderfully uncomfortable thesis and an excellent dinner-party contribution, but it is empirically bruised: Many Labs 4 failed to replicate the central mortality salience effect across 17 laboratories and 1,550 participants, with and without the involvement of the original authors.
For what actually happens inside organizations the grand existential narrative is unnecessary anyway, which is good news for anyone dealing with a sceptical CFO. The far more modest and far better evidenced finding is quite enough: a person whose professional relevance is in question contracts their attention onto what they already master. Nobody needs to be reminded of death to be made rigid, it is entirely sufficient to tell them fifteen times that the work they happen to be good at is about to look different.
The moment somebody stops being an author
In the same Deloitte survey sits the figure that explains all the rest, and it sits there so casually that most readers skate past it: workers are twice as likely to feel negatively about an employer-imposed change as about a self-imposed one. Same change, same consequences, same effort, and the only difference is who set it in motion. No communication concept in the world closes that gap, because it is not a gap in understanding but a gap in authorship, and you cannot explain authorship back into a person after you have taken it away from them.
That point, at which a person stops being the author of their own decision and becomes the recipient of somebody else's, is what we at Engaginglab call the Zero Moment of Agency. I originally coined the term for the consumer, for the instant in which they hand an AI agent their preferences as fixed parameters and thereby stop deciding for themselves. The employee who is informed of a transformation stands in the same moment, only from the other side: authorship is not taken from them because they conveniently surrendered it, but because nobody asked. Recipients do not adapt, they comply, and compliance looks exactly like adaptation in every reporting dashboard, right up until the day something unforeseen happens and there is nobody who reacts on their own.
What an organization builds instead
Anyone responsible for change in an organization decides every day at which points people are still authors and at which they have already become recipients, and that is a question of architecture rather than of communication. Deloitte supplies the practical footnote itself: the barrier most frequently cited by leaders and managers, at 44 percent, is that a change is not visibly relevant to the individual, and among workers 30 percent say they do not understand how the change relates to them. Both sound like an explanation problem and neither is one, because both dissolve the moment the affected person is involved in shaping the thing, since a person who helped build something no longer needs its relevance explained to them.
There is a test for this and it costs nothing. Take the last major change in your company and ask at which point an employee could have decided something that would genuinely have altered the outcome, not the format of the kick-off or the colour of the intranet banner, but something that would have been visible in the result. If you cannot find that point, you did not run a change programme, you made an announcement, and the people who are now labelled change-fatigued have done nothing except behave exactly as the system taught them to, fifteen times over.