Senior leaders are frequently the last to learn about problems that have been visible on the floor for months. The delay is structural, produced by how information is filtered at each level it passes through.
Every layer rounds the report slightly
A manager reporting upward describes their own area, and describing it as failing invites intervention and reflects on their competence.
The natural response is to report the problem alongside the steps being taken, which softens it by a small and defensible amount.
Repeat that through three or four layers and a serious operational failure arrives at the top as a manageable issue under active review.
The incentive to withhold is strongest where the problem is
The people closest to a failure are also the people most likely to be held responsible for it, which is a poor position from which to raise an alarm.
They frequently believe, sometimes correctly, that they can resolve it before anyone above needs to know.
Disclosure therefore tends to happen only once the problem has grown beyond local capacity, which is the latest possible moment and the least useful.
Asking does not solve it
A leader who asks whether there are problems receives an answer shaped by what the person believes the leader wants to hear.
Open-door statements do not change the calculation, because the risk of being the person who raised something is unchanged by an invitation.
What alters the behaviour is observable evidence about what happened to the last few people who brought bad news, which is information the organisation shares readily among itself.
Independent channels reduce the filtering
Information that reaches leadership without passing through the management chain is less degraded, which is why customer complaints, exit interviews and support data are so valuable.
Leaders who spend time with people several levels down hear a different account, though only where those conversations are routine rather than exceptional.
An unusual visit is treated as an inspection and produces prepared answers, whereas a regular one eventually produces ordinary ones.
Speed of escalation is the measurable variable
The useful question is not whether problems exist but how long they take to travel, and that interval can be observed after each incident.
Organisations that examine the delay rather than the failure learn something about their reporting structure that the failure itself does not reveal.
Those that focus only on who caused the problem generally find the next one takes even longer to surface.