Judgment is built by deciding and finding out. The finding out is not optional, and a substantial share of what somebody learns comes from the decisions that went badly.
Which places a leader in an uncomfortable position: developing people requires permitting outcomes you could have prevented, and every instinct in the role runs the other way.
What has to be true first
Not every failure is permissible, and the distinction is not about size.
Nobody gets hurt. Safety is never a development exercise. Where physical risk is involved, intervene immediately and without regard for what it costs the learning.
Nothing irreversible. A lost margin is recoverable. A lost licence, a lost major relationship, or a legal exposure is not, and the person cannot consent to those on the organization’s behalf.
Nobody outside absorbs it. A client, a patient, or a member of the public should not be paying for somebody’s development. Where they would, the exercise is not legitimate whatever it teaches.
The person’s standing survives. If failing this would end their prospects in the organization, they have not been given an assignment. They have been given a test nobody told them about.
Within those bounds, a considerable amount can be permitted, and most organizations permit far less than the bounds allow.
Rescuing teaches the wrong lesson twice
The intervention feels like support and it transmits two things nobody intended.
To the person: their decisions do not fully belong to them, the safety net is always present, and their reasoning was never actually tested. They will remain uncertain of their own judgment for exactly as long as it has not been exposed.
To everybody watching: authority here is nominal. Anything difficult reverts to the leader, which means accepting responsibility is not accepting anything real, and the sensible strategy is to defer rather than decide.
The second effect is the more expensive one and it is invisible, because it manifests as a general reluctance to own things that nobody attributes to the rescue.
Watching without intervening
Three things make it possible to observe an avoidable loss deliberately.
Cost it in advance. Decide beforehand what the worst realistic outcome is and whether you would accept it. Made in advance, the decision is calm; made while watching, it will be made by discomfort.
Name the intervention threshold out loud, to them. I am not going to step in unless it reaches this point. That statement binds you as much as it informs them, and it converts your restraint from apparent indifference into a stated arrangement.
Distinguish a different approach from a wrong one. Most of the urge to intervene arrives when somebody is doing it in a way you would not, which is not evidence of anything except that they are a different person.
When it goes wrong publicly
The moment that determines whether anybody in the organization will accept a stretch assignment again.
Take the responsibility externally. To the board, the client, or the community, this was your decision to delegate and the outcome is yours. That is also literally true, and it is the correct account.
Hold the conversation about their reasoning privately, afterward, and let it be a real conversation rather than a soft one.
What must not happen is the reverse: an external account that locates the failure with the individual. It is accurate in a narrow sense, it is understood by everybody as an abandonment, and it ends the organization’s ability to develop anybody for several years.
The conversation afterward
Four questions, and the order matters because it separates reasoning from outcome before either party has settled on a story.
What did you know when you decided? What did you expect to happen? What actually happened, and where did the two diverge? What would you look for next time that you did not look for this time?
Notice that none of them asks what went wrong, which invites a defence, or how they feel about it, which invites reassurance.
The genuine possibility that the decision was sound and the outcome was not has to remain available, and saying so where it is true is one of the more valuable things a senior person can do. A person who concludes that any bad outcome indicates a bad decision has learned something false that will make them cautious for a career.
The second assignment
The single act that determines what the first one taught.
Somebody who fails and is given nothing further has learned that failure is disqualifying, and so has everybody who watched. The stated commitment to development is now disconfirmed by the clearest available evidence.
Somebody who fails and receives another assignment of comparable weight within a few months has learned the intended lesson, which is that the organization was serious. That second assignment costs almost nothing and it is the part most organizations omit.
Where the honest assessment is that the person is not suited to that kind of responsibility, say so directly rather than communicating it by never offering anything again. Silence is read accurately and it is the least respectful available option.
Edited by Patrick J. Wolf, PhD