How Uber makes money: Building the super app of the driverless future

Every time you tap "Request" on Uber, a lot more money changes hands than you might think. The driver collects the fare. Uber takes a cut. The total value of all those fares, meals, and grocery orders worldwide added up to $193.5 billion in 2025, roughly the size of Morocco’s entire economy.

But here's the thing: Uber only reported $52 billion of that as its own revenue. The rest flows through to drivers, couriers, and restaurants. Uber is the middleman, and it gets paid like one.

That middleman position is the whole game. Uber doesn't need to own a single vehicle to profit from every ride. It doesn't need a kitchen to profit from every meal. It just needs to sit in the middle of the transaction, taking a percentage, across 70 countries and 15,000+ cities, with close to 200 million people using it every month.

For most of its existence, that model lost money at scale. Then in 2025, Uber generated nearly $10 billion in free cash flow. A company that was once the poster child for Silicon Valley cash burn quietly became one of the most profitable technology businesses.

In this breakdown, we'll unpack exactly how Uber makes money: where the revenue comes from, what it costs to run the business, who it's up against, and whether the profit machine can last.

Uber was founded in 2010 to fix a simple piece of urban friction: getting a ride at the tap of a button. Fifteen years later, its mission, "to create opportunity through movement," covers far more ground.

By early 2026, Uber operated in more than 70 countries, 15,000+ cities, and reached roughly 200 million monthly active platform consumers (MAPCs). Structurally, it is a three-sided technology platform that connects riders with mobility drivers, eaters with merchants and couriers, and shippers with freight carriers. Consumers get reliable logistics, earners get flexible income, and merchants get demand.

The product lineup reflects that scope:

The value proposition is asset-light orchestration. Uber owns no vehicles, kitchens, or warehouses; it provides dispatch logic, dynamic routing, payment rails, and identity verification, the digital connective tissue that clears the marketplace in real time.

The most distinctive feature is its cross-platform network effect. Roughly 40% of Uber's customers use more than one service, and in Q4 2025, 58% of first-time Delivery consumers were new to the platform.

In markets where both products exist, consumers using Mobility and Delivery generate more than 3x the Gross Bookings of single-product users. The 50 million-member Uber One program deepens that flywheel further.

Uber is also broadening from "move people and meals" into wider local commerce and travel. It has integrated 700,000+ hotel listings via Expedia, OpenTable reservations, a "Shop for Me" retail concierge, and Uber Direct white-label logistics. And it has plugged its Eats catalog into ChatGPT while partnering with Toast for merchant-side advertising.

Uber makes money in three main ways: rides, food delivery, and freight. Rides is the biggest, delivery is catching up fast, and freight is the odd one out.

For years, the company grew fast and lost money doing it. That changed in 2023, when Uber turned a profit for the first time. Since then, profit has grown faster than revenue.

In 2025, rides brought in $29.7 billion (57% of total), delivery brought in $17.3 billion (33%), and freight added $5.1 billion (10%). On top of those three, Uber also makes money from advertising and its subscription program Uber One, both growing fast and explored below.

Rides is how Uber started, and it's still where most of the money comes from. In 2025, it brought in $29.7 billion, more than half of total revenue, and it's also where Uber makes the most profit per dollar earned.

The basic mechanic is simple: every time someone takes a ride, Uber keeps a cut of the fare. That cut is roughly 20–30 cents of every dollar the rider pays, with the rest going to the driver. Uber doesn't publish an exact figure, and the actual percentage varies by city and trip type, but it's consistently the company's most profitable slice of the business.

What's changed recently is the product range. Uber Black (premium rides) and Uber Moto (motorbike taxis, big in Asia and Latin America) each grew 40% in 2025. Cheaper and premium options together now make up about 35% of all trips. That matters because Uber earns different margins from a $7 moto ride in Lagos than from a $40 Black car in Manhattan.

Uber Eats started as a side project. In 2025, it generated $17.3 billion in revenue and turned a genuine profit for the first time. It now accounts for a third of Uber's total business.

The money flows from three directions at once: restaurants pay Uber a commission on every order (typically a percentage of the basket), customers pay a delivery fee, and Uber takes a margin on the gap between what the customer pays the courier and what the courier actually earns. It's a cut taken from every side of the transaction.

Grocery and retail are the fast-growing parts of this. One in five Uber Eats deliveries in 2025 was groceries or household items. Uber has struck exclusive deals with major supermarket chains in Australia and Canada, pushing into bigger, more frequent orders that restaurants alone couldn't drive.

Advertising inside the Eats app has quietly become a significant revenue stream on its own. Restaurants and consumer brands pay to appear higher in search results and get promoted placement. In 2025, advertising added $568 million to Delivery's revenue, and unlike delivering food, it costs Uber almost nothing to run.

Freight is the quiet third leg of the business, and the weakest one. Uber connects companies that need to ship goods with truck drivers who can move them, taking a fee in the middle.

It brought in $5.1 billion in 2025, about 10% of Uber's total, but it's the only segment that still loses money. The trucking industry had a rough couple of years: too much capacity, not enough demand. Uber's freight business felt it.

Strategically, it keeps Uber relevant in a broader slice of the transportation economy. If the trucking market recovers, the margins should follow.

Uber didn't think of itself as an advertising company until recently. Now advertising is one of its fastest-growing businesses, crossing $2 billion in annualised revenue by the end of 2025, up more than 50% in a year.

It runs two types of ads. The first is straightforward: restaurants and brands pay to appear at the top of Uber Eats search results or get featured in the app. The second is more unusual: brands pay to show ads to passengers during active rides, through in-app video or digital screens on the roof of the car.

What makes Uber's ad business valuable is what it knows about its users. It can see exactly where people go, what they eat, and when they order. That data lets brands target ads with precision that most platforms can't match.

Uber One is a monthly membership priced at $9.99 a month or $99.99 a year in the U.S. It gives subscribers free delivery, discounts on rides, and priority service. By the end of 2025, 46 million people had signed up across 47 countries, a number that's grown 55% in a single year.

The interesting thing isn't the subscription fee itself. It's what subscribers do differently. Members spend roughly three times more on Uber than non-members. By early 2026, they accounted for half of all the money flowing through Uber's rides and delivery combined.

Uber is essentially trading a small discount today for a customer who barely thinks about opening a competitor's app.

Running a platform that moves people and food across 70 countries every day is extraordinarily expensive. Uber spent roughly $46 billion to make $52 billion in 2025.

The biggest costs aren't salaries or office space. They're the direct costs of every trip and delivery: paying drivers and couriers, covering insurance, and processing billions of card transactions. Everything else comes after.

The single largest cost Uber has is the money that flows to the people doing the actual work. In markets where Uber sets the fare and pays the driver directly, those driver and courier payments are booked as a cost. In 2025, those payments grew by $1.6 billion on the rides side and another $1.6 billion on the delivery side, reflecting how fast both businesses grew.

Insurance is the cost that surprises most people. Every active trip on the platform needs commercial auto coverage, which is far more expensive than personal car insurance. For years, this was a growing headache; in 2025 insurance costs rose by $851 million. Uber has been working to bring that down by improving its safety data and stabilising how it prices trips.

Payments are the other hidden tax on every transaction. Every time someone pays by card, Uber pays the card network a fee, typically between 1.5% and 3.5% of the transaction. Across nearly $200 billion worth of trips and orders, that adds up to billions a year before Uber earns a cent.

Uber spent $4.9 billion on sales and marketing in 2025. A big chunk of that, $1.6 billion, went directly back to customers in the form of discounts, free deliveries, and promotional credits. That's how Uber keeps riders and eaters coming back in competitive markets.

The company's advantage here is that its own products do a lot of the marketing work. Rides and food delivery cross-promote each other, and Uber One members barely need a reason to open a competitor's app. That means Uber spends less acquiring each customer than it would if it were running a single-product business.

Uber spent $3.4 billion on research and development in 2025. That's the cost of keeping the platform running: the algorithms that match a driver to your location in seconds, the surge pricing that balances supply and demand, the fraud detection that runs on every transaction, and the maps that route couriers through city traffic.

One reason this is efficient: rides, food delivery, and freight all run on the same underlying technology. When Uber added hotel bookings or integrated with ChatGPT, it didn't need to build a new platform from scratch. The existing infrastructure stretches across new products without a proportional increase in cost.

Uber employs about 34,000 people in corporate roles. Running the actual day-to-day operations, from customer service to the physical hubs where drivers get onboarded, cost $2.85 billion in 2025. Corporate overhead, legal, and administration added another $3.24 billion.

The legal line is worth noting. Uber has been fighting court battles in multiple countries over whether its drivers should be classified as employees or independent contractors. Those cases generate ongoing legal costs and, occasionally, large settlements. It's a cost that isn't going away any time soon.

Uber has locked itself into $2.1 billion worth of cloud computing contracts running through 2029, plus $10.6 billion in debt from earlier in its history. These are fixed obligations the company has to service regardless of how the business performs in any given year.

On top of that, Uber is placing bets on where the industry is heading. It invested $300 million in Lucid, the electric vehicle maker, and is funding Nuro to put 20,000 self-driving vehicles on its platform over the next six years. These aren't costs that show up dramatically today, but they signal where Uber thinks its future competition will come from.

Uber doesn't have one main rival. In rides it competes with Lyft. In food delivery it competes with DoorDash. In Asia and Europe, entirely different companies dominate. Each fight is different, and Uber's edge is that profits from one business can fund battles in another.

Lyft is Uber's only real competitor in the U.S. rides market, and the gap between them is wide. In 2025, Lyft completed 945 million rides. Uber completed 13.6 billion trips globally. In the U.S. specifically, Lyft holds about 24% of the market to Uber's 76%.

Lyft's strength is that it focuses on one thing: rides. It turned a genuine profit in 2025, generating $1.12 billion in free cash flow. The trade-off is that it has no food delivery business, no freight operation, and no subscription flywheel anywhere near the scale of Uber One.

In 2025, Lyft bought FREENOW, a ride-hailing app with a presence in 180 cities across Europe. It's a meaningful expansion, but Lyft is still a single-product company trying to compete with one that earns money from rides, food, groceries, freight, advertising, and subscriptions all at once.

In U.S. food delivery, DoorDash is the clear leader. It holds about 61% of the market to Uber Eats' 26%. In 2025, it processed 3.2 billion orders, generated $13.7 billion in revenue, and turned its first real profit of $935 million.

The bigger move came at the end of 2025, when DoorDash paid $3.7 billion to acquire Deliveroo, the dominant food delivery app in the UK and several other European markets. That acquisition turned DoorDash from a U.S.-centric business into a genuine global player overnight.

DoorDash also has its own subscription program, DashPass, with 35 million members. That's still well short of Uber One's 50 million, and crucially, DashPass only covers food delivery. Uber One covers both rides and food, which makes it stickier and harder to replicate.

Grab is what Uber could have become if it had stayed in Southeast Asia and gone much deeper. It operates rides, food delivery, and grocery delivery across more than 900 cities in the region, but it also runs its own digital banks in Singapore and Malaysia.

In 2025, Grab generated $3.3 billion in revenue and turned a profit for the first time, earning $200 million. The banking side alone brought in $347 million, with over $1.6 billion in customer deposits sitting in its digital banks.

Uber sold its Southeast Asia business to Grab in 2018. The two are no longer direct competitors, but Grab shows how deep a platform built around transportation can go when it stays in one region long enough.

Bolt is the underdog trying to take Uber's European business by being cheaper, especially for drivers. Where Uber takes a cut of roughly 20–30% from drivers, Bolt charges under 20%. That lower fee attracts drivers, which means shorter wait times, which attracts passengers.

The company operates in more than 600 cities across Europe and Africa and brought in around $2.15 billion in revenue in 2024. It hasn't turned a profit yet, reporting an operating loss of €87.7 million, but it raised new financing in 2024 and has been signalling a potential stock market listing.

In cities where it competes directly with Uber, Bolt often wins on price. The question is whether that's enough to build a sustainable business at scale.

DiDi is the dominant ride-hailing company in China, with about 72–76% of the market and 580 million users who took 18 billion trips on the platform in 2025. At roughly $31 billion in revenue, it is one of the only companies in the world operating at a comparable scale to Uber.

The more interesting story is what DiDi is building next. It's developing its own self-driving car technology, partnering with Chinese car manufacturer GAC to build purpose-built robotaxis, and rolling out one of the world's largest electric vehicle charging networks.

If DiDi succeeds in building a self-driving fleet at scale inside China, it has a template that other regional players could replicate. The long-term question for Uber is whether the middleman model holds up when the cars start driving themselves.

Uber is making two bets on the next decade. First: self-driving cars are coming, and Uber wants to be the app people use to hail them. Second: people will eventually use Uber for hotels, restaurants, groceries, and shopping, not just rides and food.

Uber isn't building its own self-driving cars. It's partnering with the companies that are: Waymo already offers robotaxi rides in Austin and Atlanta through the Uber app. Uber's logic is that it doesn't need to own the cars, just the app people open when they want one.

In early 2026, Uber launched a unit to sign up more self-driving companies, targeting robotaxi rides in 15 cities by year-end. If that works, Uber becomes the distribution layer for an industry it didn't build.

The expansion beyond rides is already happening. You can now book hotels through Uber via Expedia, make restaurant reservations through OpenTable, and order groceries through the same app. Uber's food catalog is also searchable inside ChatGPT. These don't add much revenue today, but they build the habit of reaching for Uber beyond getting somewhere.

The biggest risk is a court ruling. Across Europe, governments are pushing to reclassify Uber's drivers as employees, which would require Uber to pay wages, benefits, and pension contributions to millions of people it currently treats as self-employed. Courts have split: wins in France and the Netherlands, a loss in New Zealand. The EU deadline is December 2026.

Analysts expect $58 billion in revenue in 2026 and close to $90 billion by 2030. Uber is returning half its free cash flow to shareholders through buybacks, which signals confidence of a mature company. 

Whether that confidence is justified depends on two things Uber can't fully control: how European courts rule on driver classification, and whether self-driving technology reshapes the industry before Uber has embedded itself as the platform that sits on top of it. Both will be clearer by 2027. Until then, the business is profitable and expanding in ways that would have seemed implausible five years ago.