Your first reaction to that headline might be that it is not a problem.
Fewer things are reaching you. Your people are handling more. You hired capable adults and gave them room, and the room is being used. That is what delegation is supposed to look like, and if you have spent any part of the last two years trying to get out of the weeds, a quieter inbox is the evidence that it worked.
That reading is available and sometimes it is correct.
Here is the difficulty. Delegation working and information dying on the way up produce the same experience from where you sit — fewer problems arriving, less firefighting, a calendar with more room in it. One of those is the outcome you were after. The other one costs you about eighteen months and usually announces itself as something else entirely.
You cannot tell them apart by how it feels. You can tell them apart by what stopped arriving.
Why has my team stopped raising issues with me?
Because at some point, raising something cost more than it returned, and people adjusted.
That adjustment is not disloyalty and it is rarely conscious. It is what any competent person does with repeated evidence: they update. Somebody brought you something inconvenient, and what followed was worse than what would have followed if they had absorbed it. Maybe the response was sharp. Maybe it was fine but nothing changed. Maybe it did change, and the person who raised it ended up owning the fix on top of their existing work.
None of those requires you to have done anything unreasonable. The third one is what happens to good organizations run by responsive leaders — you raised it, you understand it, you take it — and it teaches faster than the other two, because it is fair.
Whatever happened, it happened in front of people. The person who raised it learned something, and so did everyone who watched, and so did everyone they told. One event of that kind is a data point. Three of them is a policy.
Nobody decided to stop telling you things. They ran the numbers, and the numbers said don’t.
What does it mean when problems arrive already solved?
It means somebody decided which solution you would get to consider, and it was not you.
This is the tell most leaders miss, because a problem arriving with a solution attached looks like initiative. Frequently it is. But there is a version that looks identical and works differently: two people work it out in a hallway, arrive at something defensible, and bring it to you as a finished thing. The third person — the one who would have caught the flaw in it — was not in the hallway, and by the time the item reaches you it has enough momentum that raising an objection would mean sending competent people back to redo work they have already finished.
So you approve it. It was reasonable. And the flaw shows up two quarters later attached to a different department’s number.
The distinguishing question is not whether solutions arrive. It is whether alternatives do. Healthy escalation brings you a problem, a recommendation, and the thing they considered and rejected. Degraded escalation brings you a recommendation. The missing piece is not the work — it is the visible evidence that the decision was contested before it reached you.
How do I tell delegation from concealment?
Ask questions — none of them require asking anybody except yourself.
Are you getting fewer problems, or fewer surprises?
Real delegation reduces the volume of things reaching you and does almost nothing to the variance. Capable people handling more of the routine still hit things they did not expect, and those still travel upward, because that is what delegation is for. If the routine noise dropped and the surprises dropped with it, the second one is not delegation. Nothing about giving people authority makes the world more predictable.
What kind of problem stopped arriving?
Sort the last six months of escalations by category rather than by count. Delegation thins the operational ones — scheduling, resourcing, the ordinary friction of getting work out. It does not thin the ones with a person’s name attached, or the ones where two departments want incompatible things, or the ones where somebody has to say that a commitment already made is not going to hold.
Those are the expensive ones. If they have gone quiet, they are not being handled. They are being carried.
When something goes wrong, are you finding out early or finding out late?
This is the cleanest test and the most uncomfortable, because it can only be run in retrospect. Take the last three things that went wrong. Not the ones that were caught — the ones that landed. Work backward and establish when the first person in the building knew something was off.
If the gap between first knew and you knew is measured in weeks, you have your answer, and the length of the gap is roughly proportional to how long the signal has been degrading.
What is it costing you?
More than you realize, because they are cumulative and cascading in nature. The order matters, because each one enables the next.
You are deciding on filtered information. Every decision you make is being made from a version of the situation that has already passed through people calculating what you want to hear. Not lying — filtering, which is a different activity with the same effect. You are still smart, still experienced, still good at this. You are working from a sample that has been cleaned.
Your best people are absorbing things that are not theirs. The person who used to raise the hard thing is still noticing it. They have concluded that raising it costs more than it returns, so now they are managing around it, which costs them time, discretionary energy, and eventually the belief that being here is worth what it takes. This is the mechanism behind the exit that surprises you — the one where a good person leaves a job they said they liked and the exit interview produces nothing useful.
And the problem gets more expensive the entire time. A thing raised early is a conversation. The same thing raised at the point where it can no longer be absorbed is a crisis with a cost attached, and the cost is not the problem. It is the interval.
How do I get honest feedback from my team again?
Not by asking for it, and this is where most repairs fail.
The standard responses are all announcements: an open door, a skip-level, an anonymous survey, a stated commitment to candor at the next all-hands. Every one of them says it is safe to bring me things. None of them changes what happens after somebody does.
You are asking people to pay a cost you have already demonstrated. They are not confused about your invitation. They understood it. They are weighing it against the evidence, and the evidence is what happened the last several times, which is more recent and more reliable than anything you can say about the future.
The announcement is not neutral either. An invitation that produces the same outcome as before does not leave the situation where it was — it confirms the original read and adds a data point about invitations.
So the repair is not upstream of the behavior. It is downstream of it, in what reliably follows.
The calculation is changed predictably, and it is structural rather than attitudinal.
What happens to the person who raised it. If raising a problem reliably means owning it, the price of speaking is a second job and everyone can see it. Separating who identified it from who owns it is a decision-rights change, and it is the single highest-leverage one available here.
Whether it comes back. Track escalations and close them out loud. Not because tracking is virtuous, but because an item that returns with an answer is a demonstration and an item that vanishes is a different demonstration. The organization is watching which one this place does.
What happens when somebody was right. The rarest of the three and the most powerful. When somebody raised a risk that materialized, going back to them afterward — in front of others — and saying so is worth more than a year of open-door language. It is the only move that pays the cost of speaking rather than promising to.
None of this requires you to be more available. It requires the consequence of speaking to change, and consequence is a design question.
Where this leads you
What you are looking at is not a communication problem and not a people problem.
Your organization learned something from watching what happened, the way organizations always learn — from consequence rather than from intention. In Leadership by Design, that is the Signal phase, and it is the first of eight, which means whatever distortion it carries travels into everything downstream. The quiet inbox is not the failure. It is the receipt.
And the fracture is probably not where it hurts. The place you feel this — decisions made on thin information, a good person leaving, a problem that arrived far too late — is downstream of the place it started, sometimes by several phases and sometimes by a couple of years.
The useful move is not to work harder on being approachable. It is to find out what your organization learned, and when.
How to trace a failure back to its source → The diagnostic descent, worked end to end.
