The $1 Million Salary Band Just Made Headcount Obsolete

TL;DR

ClickUp cut 22% of its workforce while announcing $1 million cash salary bands for the remaining staff who can produce 100x output by managing AI systems. CEO Zeb Evans framed the cuts as a compensation redesign, not a cost reduction. His line that the people who automate their jobs with AI will always have a job captures the new compensation logic. This signals the end of headcount-based pay and the start of agent-leverage pay. Three roles get the salary uplift in this model: agent designers, orchestration engineers, and trust-and-safety operators.

---

ClickUp's CEO posted a thread on X this week that is going to be quoted in every compensation committee meeting for the next year.

Zeb Evans announced that ClickUp had cut 22% of its workforce. Then he announced that the people who remained would have access to a $1 million cash salary band if they could produce what he called "100x impact" by creating or managing AI systems.

The line that mattered most: "The people that automate their jobs with AI will always have a job. They become owners of the AI systems. Agent managers."

That sentence is doing a lot of work. It is signaling the end of headcount-based compensation logic and the beginning of agent-leverage compensation logic. And ClickUp is not the first company saying it. They are just the first company posting the salary number publicly.

The Cuts Are the Less Interesting Half of the Story

Most coverage of the ClickUp restructuring led with the layoffs. 22% is a big number. It hurts. And it is being framed as part of the broader 2026 tech layoff wave.

But the part that should make every executive look up is not the cuts. It is what Evans said about the survivors. The savings from the restructuring are being redirected into the compensation of the people who can manage AI systems. The savings are not being banked. They are being repaid to the workforce as performance compensation tied to agent leverage.

In effect, ClickUp turned its layoffs into a recruiting campaign for the next tier of employee. The message to the labor market is: come here, manage agents, and you can earn the equivalent of two or three traditional senior roles. The implicit message is that ClickUp is no longer hiring traditional senior roles at all.

Why This Will Spread Faster Than the Layoffs Did

Tech layoffs spread through the industry on a roughly 18-month cycle when a precedent gets set. AI compensation models are going to spread faster than that. The reason is simple. Boards look at competitor compensation closely, but they look at competitor talent leverage even more closely.

If ClickUp produces 100x output per employee with $1 million salary bands, and a competitor produces 10x output with $250,000 salary bands, the competitor's board is going to ask why. The answer will be the comp structure. And then the competitor's board will copy it.

BusinessToday noted that Evans is positioning the change as a generational reset of how engineering teams are sized and paid. He is not wrong about that. He is just the first to say it publicly with a specific number attached.

The Hive Structure Has a Salary Sheet Now

Two roles only. Bees handle execution at machine speed. The beekeeper directs the bees and judges the output.

For 18 months, leaders have asked me what the compensation structure looks like in a Hive Structure organization. The honest answer was: I don't know yet because no one had set the precedent publicly. ClickUp just set it.

The bee work is performed by agents. Agents have a marginal cost denominated in tokens, not salaries. The beekeeper work is performed by humans, and those humans are paid as if they are running a small business inside the company. Because they are.

This is why the middle of the org chart cannot survive. The middle existed to coordinate between many bees and a few beekeepers. When the bees are machines, the coordination becomes a software problem, not a management problem. The middle compresses into the prompt.

What This Does to Performance Reviews

If you are an HR leader, the implications of the ClickUp model are bigger than the salary number. Performance review systems built around individual contribution metrics start to break when most of an individual's output is mediated through AI systems.

How do you evaluate someone whose 100x impact came from designing an agent that ran for six months without supervision? You cannot use traditional rubrics. You have to measure the output of the system the human is responsible for, not the inputs the human spent on it.

This is going to surface a brutal reality at a lot of companies. The people who were strongest under the old performance system, the ones who were rewarded for hours worked and tasks completed, are not necessarily the strongest under the agent system. The strongest under the agent system are the people who can clearly specify outcomes, design feedback loops, and trust the machine to run.

That is a different person from the one your HR team has been promoting for the last decade.

The Three Roles That Win Under This Model

Three roles get the salary uplift in any company that adopts a version of the ClickUp model.

The first is the agent designer. This person writes the specifications, prompts, and feedback systems that turn a general-purpose model into a domain-specific agent. The work is part product design, part engineering, part operations. A great agent designer is worth a team of average ones, and the salary band reflects that.

The second is the orchestration engineer. This is the person who wires multiple agents together into a workflow. They own the handoffs, the failure modes, and the human escalation logic. As more business processes get automated end-to-end, the value of this role compounds.

The third is the trust and safety operator. This person evaluates what agents produce, catches errors before they reach customers, and ensures that the company's AI systems do not create regulatory or reputational risk. Boring on paper, indispensable in practice. As agent autonomy increases, the cost of a failure increases too, and this role exists to keep that cost manageable.

If you are a leader reading this, the work for this week is to identify whether each of these roles exists inside your company today. If they do not, you are still operating in the assistant era. If they do, you have started the transition to the agent era, but you almost certainly do not pay them what their leverage is worth.

The Replacement Exercise, At Industry Scale

I have written before about what I call the replacement exercise. The idea is to constantly automate yourself out of the tasks that can be automated, so that the tasks you remain irreplaceable on are the ones the agent cannot do.

ClickUp just turned the replacement exercise into a compensation philosophy. The people who replaced themselves on rote work and moved up the stack to managing agents got the salary increase. The people who held onto the rote work got the offboarding.

This is going to feel cruel to people who valued the rote work. And there is a real conversation to have about transition support, retraining, and the social cost of running this transition badly. The version of this story Evans will tell in 12 months will be defined by how ClickUp treats the people it cut. If they were treated well, the model becomes a template. If they were treated badly, the model becomes a cautionary tale.

Either way, the structural conclusion is the same. Headcount-based compensation is being replaced by agent-leverage compensation. The boards making that change at $50 million companies this year will be doing it at $5 billion companies next year.

So Where Are You on the Line?

Look at your most recent quarter of work. Ask yourself two questions.

What percentage of my output came from work I personally did, hour by hour, task by task. And what percentage came from systems I designed, agents I directed, or workflows I orchestrated that ran without me.

If the first number is high, you are a bee. That is not a judgment. It just means your compensation curve is going to flatten as the work you do becomes cheaper.

If the second number is high, you are a beekeeper. That means your compensation curve is going to steepen, because the leverage you produce will only get more valuable as agent capability improves.

The question Evans posed implicitly with his thread is one you can answer for yourself right now. Which side of the line is your role on, and what would it take to move?

Sharon Gai is an AI transformation strategist, keynote speaker, and author of How to Do More with Less Using AI. She advises Fortune 500 companies on AI adoption and organizational redesign.