Uber Layoffs: 3,300 Jobs Cut in a Management Reset
Uber is cutting about 3,300 jobs while business is flying. This is not a downturn story. It is a management reset after the org chart became a tax on speed.
Uber is cutting about 3,300 jobs while the business is growing hard. If your company needs a downturn before it fixes management bloat, you are not running a business. You are waiting for someone else to force you to do your job.
Dara Khosrowshahi has done something plenty of CEOs know they need to do but lack the stomach for: he has publicly admitted that Uber got slower because it got more complicated.
On September 2, Uber said it would reduce its team by about 10%. The company is cutting layers, shrinking tiny management teams, combining overlapping operations and telling nearly all remote employees to return to an office. It is a proper organisational reset, not a polite little cost-saving exercise.
The numbers are brutal enough to make every founder and executive sit up. Uber is reducing employees who sit seven or more layers below Khosrowshahi by 20%. It is cutting the number of “micro-teams” — managers with just one or two direct reports — by nearly 50%. And it is moving to a model in which only roughly 1% of staff can remain remote.
That will upset people. Good. Serious management decisions usually do.
This is not a layoff story. It is a speed story.
The lazy reading is that Uber is slashing jobs because something is broken financially. The facts say otherwise.
In its second-quarter results released on August 5, Uber reported $58.0 billion in gross bookings, up 22% on a constant-currency basis. Revenue rose 12% to $14.2 billion. Trips climbed 18% to 3.9 billion. Income from operations rose 30% to $1.9 billion, while adjusted EBITDA reached $2.8 billion.
That is not a company with its back against the wall.
Khosrowshahi’s own explanation is more useful: Uber’s top line has nearly tripled over the past five-plus years, but growth has brought too much coordination, fragmented ownership and too many people involved in decisions. Translation: successful companies can become bureaucratic long before they become unsuccessful.
I have seen this firsthand. A business adds people to solve immediate problems. A new product gets a lead. The lead gets a manager. The manager needs a planning process. Someone is hired to make sure two teams are “aligned.” Then three people are in a meeting that should have been a Slack message, and ten people are in a meeting that should not exist at all.
Nobody wakes up intending to build a slow company. They build one meeting, one layer and one exception at a time.
Uber’s response is not subtle. It is consolidating its separate delivery-operations teams across restaurants, retail and direct delivery into single-threaded teams at global, regional and country levels. It is also combining Core Services Engineering and Science. More importantly, it is putting profit-and-loss responsibility under clearer single owners.
That last bit matters most.
When everyone owns something, nobody owns it. When three functional leaders need to agree before a customer problem gets fixed, the customer has already gone elsewhere.
The real target is coordination theatre
Uber has not said all managers are the problem, and they are not. A good manager creates leverage: sharper priorities, better people, faster decisions and fewer dumb mistakes.
A bad layer of management does the opposite. It becomes a human routing system for work that should be obvious. It schedules alignment. It produces slides describing the problem. It creates a feeling of motion while the people closest to the customer wait for permission.
Uber’s internal staff feedback apparently made this plain: too much work required coordination across teams, debates took too long, and decision rights were unclear. That is the corporate disease right there.
Most executives diagnose it incorrectly. They say the business has a “communication problem.” Then they add more meetings, more dashboards and more people whose job is to communicate.
Rubbish.
Most communication problems are ownership problems. If the team cannot answer three questions — who owns the number, who makes the call, and by when — the organisation does not need another workshop. It needs a decision.
Khosrowshahi’s move against micro-teams is particularly telling. A manager with one or two direct reports is not automatically useless; some specialist, early-stage or high-risk work requires that setup. But at scale, it should be the exception, not a career architecture.
A title is not leverage. A team is not a business. And being copied on every decision is not leadership.
Why the robotaxi push makes this urgent
Uber has plenty of reasons to want a faster organisation, but its autonomous-vehicle ambition is the big one.
In August, Khosrowshahi said Uber was investing from a position of strength as it worked to build the world’s largest platform for autonomous vehicles. The company’s current restructure is explicitly meant to free capacity for growth, innovation and the capabilities it believes matter over the next few years.
That means this is not merely about taking dollars out of the payroll line. It is about redirecting attention and capital toward the next version of Uber.
That distinction is vital. Cost cutting without a clear destination is panic wearing a suit. You can cut your way into a smaller, tidier failure.
But if the money and talent released by simplification get pointed at the right opportunity, a restructure becomes an investment decision. Uber is trying to protect its core ride-sharing and delivery machine while getting itself ready for a future in which autonomous vehicles are a central part of the platform.
Balaji Krishnamurthy, Uber’s CFO, said trailing 12-month free cash flow exceeded $10 billion for the first time in the company’s history. That gives management genuine room to make choices. The test now is whether Uber spends that room intelligently.
It is easier to announce a new strategy than to build one. It is even easier to call something an “AI” or “autonomous” strategy and keep the old org chart intact. Uber has at least recognised the inconvenient truth: a company cannot move at a new speed with an operating model designed for the last phase of growth.
The overlooked part: remote work is not the main event
The predictable argument will be about Uber’s remote-work rule. Only around 1% of staff will remain remote going forward, and the company is reinforcing its existing three-days-a-week office policy.
Some people will call it backward. Others will cheer it as common sense. Both camps will miss the bigger issue.
The office is not a management strategy. Plenty of terrible companies have everyone sitting together while decisions still crawl through six committees. And plenty of distributed teams deliver exceptional work because ownership is crystal clear.
Uber’s sharper point is co-location between managers and their teams, particularly for earlier-career employees, alongside a narrower hub strategy. Whether that works will depend on execution, not a memo.
If returning to the office just creates more visible bureaucracy, it will achieve bugger all. If it shortens feedback loops, speeds up decisions and helps junior people learn faster, it may be useful.
The contrarian lesson is this: do not copy Uber’s return-to-office policy. Copy its willingness to identify where work is actually slowing down, then remove the cause. For some businesses, that may mean more in-person time. For others, it may mean fewer managers, clearer written decisions and less calendar clutter.
Copying policy is amateur hour. Understanding the operating principle is where the money is.
What this means for you
If you run a company, do this before Friday.
First, count decision layers, not just reporting lines. Pick your five most important commercial decisions — pricing, hiring, product roadmap, customer escalation and capital allocation. How many people can delay each one? If the answer is more than two or three, you have found your speed tax.
Second, audit every manager with fewer than four meaningful direct reports. Do not start with a predetermined firing target. Ask whether that role creates leverage or merely coordinates people who could speak directly. Then make the call.
Third, put one owner beside every meaningful P&L, customer outcome and product metric. Not a committee. One name. That person can consult widely, but everybody must know who decides.
Fourth, kill duplicate teams. If separate teams are serving restaurants, retail and direct delivery — or your equivalent — ask whether they are genuinely creating differentiated value or simply defending their own patch.
Finally, stop treating growth as proof that your organisation is healthy. Uber’s business was growing when Khosrowshahi made this move. That is precisely why he had the chance to fix it from strength rather than desperation.
The best time to simplify your business is when you can afford to be thoughtful. The second-best time is now.
Because the longer you tolerate complexity, the more expensive it becomes to remove — and the more likely your sharper competitor will do it first.