← Back to blog

Managing a Manager Who Is Drowning

They do not send up a flare. They send up a slightly late report, a slightly shorter one-on-one, and a slightly defensive answer to a small question. Here is what to do in the six weeks that follow.

Kiyong (Eric) Lee·Operations Executive·August 2026·10 min read

The short version

Drowning managers do not say so. Watch for late or copied reports, one-on-ones ending early, and defensiveness on small stuff. Three signals in a fortnight is a pattern.
The first conversation is triage, not performance. Name the workload, take specific weight off, tell them they are not in trouble.
For two weeks, take back the weekly report, one draining customer thread, one difficult direct report, and cancel two low-signal meetings. Name each removal in writing.
Cover the optics. One sentence to peers, no narration, no gossip. The story that leaks outlives the triage.
At week six, decide with clear eyes. Most managers come back to full load. Some need the role to split. A few need a graceful move. All three outcomes beat the resignation email.

The first sign I had that Marcos was drowning was not a missed number. It was a one-on-one that ended twelve minutes early with him saying "I think we are good, right?" and me saying yes because I did not want to make him feel worse. He was a general manager in his second year running a Hana Group unit inside a Sam's Club outside Phoenix. On paper his P&L was fine. In person he had the look of someone who had stopped sleeping properly around a month before.

I did not catch it in time. Six weeks later he handed in his notice at 11 pm on a Sunday, in a two-paragraph email that read like a hostage note. He wrote that he had been trying not to let the region down, that he was sorry, and that he needed to leave in two weeks. I called him at seven the next morning. He cried. I felt sick.

That week I sat with a notebook and wrote down every signal I had missed. There were twelve of them. Almost all of them had been visible for a month. This piece is what I do now, so that I do not miss them again, and so that the manager gets a fair shot at the role before it decides for them.

What "drowning" actually looks like on a Tuesday afternoon

A manager who is drowning almost never says so. They cannot. If they say it out loud they think they have failed at the job, which is exactly the fear that is drowning them in the first place. So they hide it, and the hiding is what you learn to spot.

The signs I watch for now, in rough order of appearance:

Reports coming in late, then coming in short, then coming in copied from the previous week with a date changed.
One-on-ones getting shorter. They end early. The agenda is thin. When you ask what is on their mind they say "nothing pressing."
Defensiveness on small stuff. A minor observation about a schedule triggers a paragraph-long defense in a text thread.
Their calendar is packed, but their inbox is stale. Threads from six days ago sit unanswered.
They stop bringing you problems and start bringing you plans. The plans are neat and they never quite mention what is actually broken.
Meetings with their own team get canceled or shortened. The floor visit gets skipped.
The photos they used to send from the floor stop showing up in the group thread.
They start using the phrase "I've got it" in every single message.
They apologize for things you did not notice, and it lands weird.
They get quiet on the group thread where they used to be one of the first to reply.

None of these signals is diagnostic on its own. Any one of them can be a bad week. The pattern is when three or four show up in the same fortnight. That is when you have a drowning manager and a closing window.

Fig. 1 · The drowning pattern rarely shows up all at once.

The first honest conversation

You do not fix drowning by adding coaching hours. You fix it by removing weight. But before you can remove weight, you have to acknowledge, out loud, together, that the person is carrying more than they can lift.

That first conversation is not a performance conversation. It is a triage conversation. The frame I use, more or less verbatim: "I have been watching your reports for three weeks and I think you have been carrying more than the role gives you time for. I want to spend forty five minutes today just listing everything on your plate. We are going to move some of it off you for the next two weeks so you can breathe. You are not in trouble. This is a normal thing that happens, and it is on me for not spotting it sooner."

Two things about that script. First, "you are not in trouble" is not a favor. It is the truth. Second, "it is on me" is also the truth. A drowning manager is almost always a signal that the role has drifted past what one person can do, and the person above them did not catch the drift.

A drowning manager is not a weak manager. They are the early warning system for a role that has quietly outgrown one human. Fix the role first. Then decide about the person.

The two-week triage: what to take off their plate

For the next two weeks, you take specific things off. Not "some things." Specific, named things, with a start date and an end date. Otherwise the removal is theoretical and the manager will keep doing them silently because they think you did not really mean it.

Take back the reports

Whatever they send you weekly, you write for the next two weeks. Pull it from the same data source they were using. Send it under their name if the audience is external. They keep the credit. You do the work. This alone recovers about four hours a week and, more importantly, ends the daily dread of the report they know they are behind on.

Take back the top customer escalation

The one guest, franchise partner, or landlord who has been draining them. You take that relationship for two weeks. You do not ask the manager to loop you in. You just take it. Tell the counterparty by email in one line: "For the next two weeks, please copy me directly on this thread." The manager exhales the moment that email goes out.

Take back one direct report

If they have a difficult direct report, one who takes ninety minutes of their week to manage, you take that one-on-one for two weeks. You do not fire the person. You do not reorg. You just hold the one-on-one and take the emotional load out of the manager's week.

Cancel two recurring meetings

Look at their calendar. Find the two lowest signal recurring meetings and cancel them for two weeks. Tell the attendees you canceled them, not the manager. This buys them two hours a week and, again, ends the low-grade shame of showing up unprepared to a meeting they never had time to prepare for.

Name it out loud, every day

Here is the part most operators get wrong. They do the triage silently. They think they are being kind by not making a thing of it. What actually happens is the manager does not realize the load has come off, keeps working at the same pace, keeps drowning, and the two weeks pass without any recovery.

You have to name it. Every day for the first three days, a two-line message. "Report sent to the group under your name. You do not need to do this Friday's either. Talk tomorrow." "Just took the seven pm call with the franchisee. All good, will loop you Thursday." "Canceled the Wednesday standup for the next two weeks. Use the hour."

Naming it does two things. It makes the removal real to the manager, so that they actually stop doing the thing. And it makes visible to them that the drowning is being met with help, not judgment. Both matter equally.

Protect them from optics

While the triage is running, someone above you is going to notice that the reports are coming from you instead of from the manager. Someone in another department is going to notice you are on the customer thread. Someone on the manager's team is going to notice a meeting got canceled.

Your job is to absorb every one of those signals before it lands on the manager's desk. When a peer asks why you are on the customer thread, the answer is "I am covering some things for Marcos for two weeks while we sort a workload issue. He is doing fine." Full stop. You do not explain. You do not diagnose. You do not gossip. If someone pushes for more, "we are working on it, and I will let you know if anything changes" is the end of the sentence.

This is the part I got wrong for years. I would triage the workload but I would also, without meaning to, let the drowning become visible to peers. That visibility became a story about the manager that outlived the triage. Six months later the manager was fine, and the story about them not being able to handle the role was still doing damage. Cover the manager. Do not narrate the cover.

The manager who is drowning is watching to see whether you protect them from the room. If you do not, the fear that got them here doubles. If you do, they usually swim.

Week six: the honest decision

Two weeks of triage. Two weeks of coaching. Two weeks watching them run the reduced role at their own pace, with work phasing back on. By week six you know something the pre-triage state could not tell you. You know whether the manager can do the role you actually have, or only a role that does not exist.

There are three outcomes.

They come back to full load inside six weeks and stay there. This is the most common outcome, about seven times out of ten in my experience across Hana Group and Zareen's. It is why the triage was worth doing. The manager needed a rest and a witness, not a replacement.
They run the reduced role well, but they cannot come back to the full load. That means the role is too big for one person and the answer is not to fire the manager, it is to split the role. This has happened twice at Zareen's and once at Hana. In each case the split created two jobs that were each honest, and the manager who had been drowning became the strongest person on the bench.
They cannot run even the reduced role. This is the outcome nobody wants and it is the reason the honest six-week timeline matters. If the manager cannot swim in the fixed role, they need a graceful move to something they can do, or they need to leave. But you owe them the six weeks first. Firing a drowning manager without giving them a clean shot at a fair role teaches the rest of the bench to hide problems, and hidden problems are the ones that break the region.

What I got wrong before I learned this

Three mistakes I made repeatedly before I built this playbook, from sixteen years running multi-unit operations across a combined $54M in scope:

I confused stoicism with capacity. I had managers who never complained, and I read that as strength. Some of them were strong. Some of them were drowning and could not say so. The signal is not whether they complain. The signal is the twelve small things listed above.

I moved to termination too fast on a second-time drowner. I had one manager at Hana who had recovered from a drowning episode, then started drowning again eight months later, and I read it as a pattern. I moved to a performance conversation instead of a triage conversation. Turned out her personal life had cracked open and she needed a second triage, not a warning letter. She was one of the best operators I ever worked with. I lost her because I read repetition as verdict.

I underestimated how much the team above the manager needed a clear one-liner. When a peer asked me about a drowning manager, I would give too much context, and the context always got weaponized. Now the line is one sentence and I do not go past it.

The point

Managers drown quietly. They do not send up a flare. They send up a slightly late report, a slightly shorter one-on-one, and a slightly defensive answer to a small question. Your job as the person above them is to spot the pattern, take specific weight off, name what you are doing, cover the optics, and give the person six weeks in a fixed role to show you who they actually are.

Do that, and about seven times out of ten you keep a good manager who almost left. The other three times, you split the role or move the person cleanly. All three outcomes are better than the one where you did not notice until the resignation email arrived at 11 pm on a Sunday and you had to call at seven the next morning and hear them cry.

Cadence spots drowning. Silence hides it. Watch the small things.

Have a manager you think is drowning?

I coach a small number of multi-unit operators each year on bench development, general manager triage, and the tough middle stretch of a leader who is one bad month from leaving. If you are watching the signals and not sure what to do next, I am happy to trade notes.

Get in touch

Questions people ask

How do you spot a manager who is drowning?

Watch the small things, not the P&L. Reports coming in late, then copied from the previous week. One-on-ones ending twelve minutes early. Defensiveness on minor observations. A packed calendar and a stale inbox. They stop bringing you problems and start bringing you plans. Meetings with their own team get canceled. Any one signal is a bad week. Three or four inside a fortnight is a drowning pattern, and the window to help is closing.

Should you tell a drowning manager that you have noticed?

Yes, out loud, in a private conversation. The frame is triage, not performance. Say you have been watching the reports for three weeks, that you think the role has drifted past what one person can lift, that you are going to move specific work off their plate for the next two weeks, and that they are not in trouble. Naming it is what unlocks the recovery. Silent help does not land.

What should you take off a drowning manager's plate first?

Four specific things for two weeks. The weekly report you write it for them under their name. One draining customer or franchisee escalation you take the thread. One difficult direct report you cover the one-on-one. Two low-signal recurring meetings you cancel. Each removal is named to them in writing on the day it happens, otherwise they keep doing the work silently and the triage does nothing.

How long should the workload triage last?

Two weeks of active triage, then two weeks watching them run the reduced role at their own pace, then two weeks phasing the work back on. Six weeks total from the first triage conversation to the honest decision. Shorter and you do not learn what the manager can do at pace. Longer and the triage becomes the new normal, which is worse for both of you.

How do you protect a drowning manager from looking bad to peers and leadership?

You absorb the optics without narrating them. When a peer asks why the report is coming from you, the answer is: I am covering some things for them for two weeks while we sort a workload issue, they are doing fine. Full stop. You do not explain, diagnose, or gossip. The story that leaks becomes the story that outlives the triage. Cover the manager. Do not describe the cover.

When do you decide a drowning manager cannot do the role?

At week six. Three outcomes are possible. Most of the time, about seven times out of ten in my experience, the manager returns to full load and stays there. Sometimes the reduced role runs well but full load never comes back, which means the role is too big for one person and needs to split. Occasionally the manager cannot run even the reduced role, and a graceful move is the honest answer. All three outcomes are cleaner than the resignation email.

Should you fire a manager who is drowning?

Not before six weeks in a fixed role. A drowning manager is almost always the early warning system for a role that has quietly outgrown one human. Firing without giving them a fair shot at the reduced role teaches the rest of the bench to hide problems, and hidden problems are the ones that break the region. Fix the role first. Then decide about the person.