You did not remove the problem. You removed your ability to see it.

A CEO fired his HR team for "creating problems." Here is what that sentence actually tells you.

Last month the CEO of the fintech company Bolt explained, on a stage, why he had fired his entire HR team.

"We had an HR team, and that HR team was creating problems that didn't exist. Those problems disappeared when I let them go."

It was part of a larger reset: a 30% layoff, the end of the four-day week and unlimited time off, nearly the whole leadership team gone, a company that had been worth $11 billion in 2022 and was down to roughly $300 million by 2024, a drop of about 97%, now running on about a hundred people and, in his framing, AI at the core. He called it a return to startup mode.

I am not going to argue about the layoff. Companies cut, and from the outside I do not have the numbers to tell anyone they cut wrong. I want to stay on the sentence, because the sentence is a diagnosis, and the diagnosis is the part a leader can get badly wrong while sounding completely certain.

Here is the first thing worth noticing. He did not eliminate the function. He replaced the HR team with a smaller "people operations" team to handle training and support employees. In his own words on LinkedIn, "HR is the wrong energy, format, and approach," while people ops "empowers managers, streamlines decision making, and keeps the company moving at lightning speed." So the work did not disappear. The people who were doing it did, and the label changed. That detail matters, because it means the story is not "we no longer need this." It is "we no longer want the people who kept raising things."

Which brings us to the actual claim. Problems that vanish the moment a team leaves were not solved. They were silenced.

The gauge is not the malfunction.

A function whose job is to surface issues will, predictably, surface issues. The complaints, the flags, the friction reports: that is the gauge on the dashboard, not the engine. When a leader decides the gauge is the problem and removes it, the warning light goes off. Nothing under the hood changed. You have simply cut the wire to your own instrumentation and called the quiet an improvement.

For a quarter, maybe two, it works. The escalations stop landing on your desk. The room feels lighter. This is exactly why it is dangerous. The most expensive mistakes a leader makes are the ones that feel like relief.

"Complaining" is usually friction the system built.

The other phrase from that stage was a culture of "not getting things done and complaining a lot." That pattern is real and genuinely maddening to lead through. But it is almost never a character flaw that infected an entire company at the same time. It is structural.

Work stalls when nobody can say who decides, who delivers, and who only advises. It stalls when the decision that needed to close on Tuesday had no Tuesday to close in. It stalls when everyone has an opinion and no one has the pen. People inside that model are not entitled. They are stuck, and "complaining" is the noise stuck people make. Fire them and the friction stays, because it was the architecture's, not theirs.

Silence is not alignment.

The reset reportedly gave people 60 days to adapt, and by the CEO's own count, 99% did not make it. You can frame that as a culture being rebuilt. You can also frame it as most of the company's institutional memory leaving inside two months.

Whichever it is, the building is quieter now. And here is the part I would push on hardest: a room with no dissent left in it is not a room that agrees. Alignment is what you get when you surface disagreement and resolve it. Silence is what you get when you remove everyone willing to disagree. From the stage they look the same. They are not the same, and the difference does not stay hidden. It shows up in the next set of numbers, and the numbers at this particular company have not been kind.

What this is really about

When execution breaks, a leader can look at the roster or look at the architecture. The roster is easier. It gives you a name, a clean cut, a visible before-and-after. The architecture is harder, because the honest answer often points back at how the work was designed, and how the work was designed points back at the person looking for someone to blame.

Blaming a function is the tell. It is what an execution problem looks like from the inside, before the leader is willing to see it as their own. You can fire your way to a quiet building. You cannot fire your way to execution, because the thing that was broken was never a person. It was the structure they were standing in.

So the question is not whether this CEO was right or wrong to cut. The question is the one any leader should ask after a problem suddenly goes quiet: did I solve it, or did I just get rid of the people who kept telling me about it?

Those two outcomes feel identical for about a quarter. Then they stop feeling identical at all.

More to come and I'm watching what happens.

*CS