Architect’s Essay · The Corporate Training Trap
An essay on why trained people still cannot use what they were taught
You have a door that will not open. You call a locksmith, who does what locksmiths do: cuts you a key from the template used for every door. The teeth are precise, the metal is good, it turns smoothly in the hand. You pay, the locksmith leaves, and the door still does not open. It fails twice. The lock was never the kind the template fits. And behind it, the door was barred from the inside.
Hold onto that door. It is the most expensive misunderstanding in the way organizations develop their people.
The model everyone runs
Something is not working. Decisions are not delegated. Functions do not cooperate. Bad news arrives late. Decisions are made twice. So someone asks the natural question: what are our people missing? And the answer comes back in the form the question invited. They are missing a skill. Delegation. Collaboration. Candor. Decisiveness.
So the organization buys keys. It sends people away to be taught the thing they are said to lack. They rate the program highly. They return.
And the door stays shut. Six months later decisions still travel upward and the functions still do not cooperate. The conclusion is as natural as the question was: the key was not good enough, or the people did not really want to open the door. A better program is commissioned.
The model has two parts — the person lacks a skill, so supply the skill — and it fails in both.
The wrong key
Start with the key itself, because the trap begins in the classroom.
The capabilities organizations most want — delegating, disagreeing with whoever holds the decision, telling the truth early, cooperating across a boundary — depend on judgment and on other people. They are taught as if they were procedures. A model with four steps. A script for the difficult conversation. A role-play with a colleague who plays the part as written. A certificate at the end.
The method is built for the other kind of skill. The instruments measure what people can recall and how they rated the day, not what they do on a Tuesday under pressure. And the timing is wrong: away from the work, in one block, long before or long after the moment the capability is needed. A key cut from a template, for a lock that turns differently every time it is used.
Even so, many people leave the room able to do something they could not do before. That is where the second failure begins.
What a key cannot do
A skill of this kind is not used alone. That is the fact the model has no room for.
To delegate, a person needs someone who is permitted to decide and a role that does not hold them answerable for every outcome they hand over. To cooperate across a boundary, they need a measure that does not punish them for the hours it takes. To tell the person they answer to that the plan will fail, they need someone in that position who has, at least once, been told such a thing and left the speaker standing.
None of that is in the key. It is in the door: in who holds the bar, and from which side.
This is not a refinement of the skills model. It is a different claim. For work that is done the same way every time — operating a machine, following a procedure — what the person learned is nearly all that matters, and teaching it works. For work that depends on judgment and on other people, what the person learned is necessary and nowhere near sufficient. Whether it is used is decided by what the person returns to. The research is consistent on this. The surroundings predict the use of such capabilities many times more strongly than they predict the use of technical ones.
If you take nothing else from this, take that. The capability is cut in the person. The occasion to use it is held by others.
Watch it scale
Once you see the bar on the inside of the door, you see it at every size.
Someone returns from a program ready to ask instead of tell. On Monday the people around them bring the same questions, wanting the same fast answers, because their days are built on getting them. The person they answer to wants the numbers at the usual hour. Within weeks they are working as before. They did not lose the key. The people around them were holding the door.
A whole division is trained in working across functions. Everyone attends. Everyone agrees. Two years on, nothing has changed, because each function is still measured on its own result, still answers to its own head, and those heads still do not work as one. The organization taught cooperation in a room and paid for separation everywhere else. Its people obeyed the payroll.
People are trained to plan better and to stay calm under pressure. They are under pressure because a customer set a date and a price that left no room, and the organization accepted. They are being taught to turn a key in a door that is barred by a contract signed in another building.
Three sizes of the same door. In each, intelligent people are cutting keys with great care while the thing that holds the door is on the other side, in other hands, and has not been touched.
The trap inside the trap
Here is the subtler error, and it catches the careful — the ones who already know that context matters.
They check whether the door can open. They look at the role description, and the authority is there. They look at the values statement, and candor is encouraged. They look at the calendar, and the time for the new way of working has been allowed for. On paper, the door is unbarred. So they go back to cutting keys.
They are still wrong, because permitted and possible this week are not the same thing.
A new way of working is slower at first. It takes longer to ask than to tell, longer to bring someone along than to decide for them. On a day with room in it, that cost is small and worth paying. But the people being trained do not have days with room in them. Their hours were spent before they arrived. And in a week that is already full, a small added cost does not produce a small delay. It breaks the week. The person falls behind at once, and the old way of working is still there, faster, already practiced, asking nothing.
So the old way is taken. Not because the person was unwilling, and not because anything forbade the new way. Because there was no margin in which to do it badly the first few times, and nothing is done well the first few times.
Whoever confirms that the behavior is allowed has confirmed that the door has a handle. They have not asked whether anyone on the other side has a free hand.
What it costs to keep cutting keys
Every program you have seen fade among people who valued it faded in two places: in teaching built for the wrong kind of skill, and in the return.
And because nobody examined the return, the fading was read as something else: low motivation, weak follow-through, resistance. The diagnosis pointed back at the person, because the person was the only thing in the whole affair that had been changed and could therefore be found wanting. The remedy was more of the same — a better program, a firmer reminder — applied to a door that was never the key’s to open.
The cost is not only the price of the programs. It is what the people conclude. Someone who has been taught a better way and then left in a place where it cannot be used does not go back to where they started. They now know that what the organization says it wants and what it is built for are two different things. That knowledge does not fade.
What to do instead
Go to the door first.
Before you commission a single hour of teaching, ask the question the model leaves out. Not what do our people lack? Ask what would meet them if they did it? Who would have to respond differently. What is being measured that pays for the old way. Who is actually allowed to decide. What would have to stop so that there is room to do the new thing slowly.
If the answers show a bar on the inside, lift it. Change what is counted. Move the decision to where the knowledge is. Take work away before adding learning. Bring the people the person depends on, and the people around them, into the same change, so that what was learned has somewhere to arrive.
None of this lets the person off, and none of it moves the blame to the system instead. The person, the people around them, the organization and the world it answers to each hold part of the bar — each in its own way and to its own degree. So each holds part of the way out, the person included: they still have to learn, to practice, and to use what has been made possible.
Then teach — and teach differently. Not a template, but practice on real cases, in the work, over time, with the people who must respond. Teaching is the last step and it is not a small one: a door that has been unbarred still needs someone who knows how to walk through it. In that order, development holds, because it is no longer being asked to do what only a change in the conditions can do.
You are not short of locksmiths. What you are short of is someone who looks at the lock before the key is cut, and walks around to the other side of the door.
The key was cut from a template. It turned smoothly in the hand. The door stayed shut, and everyone looked for the fault in the person who held it — and no one asked what kind of lock it was, or went around to see what was holding it from the other side.