Uber, DoorDash, and Lyft did not have a demand problem in Q3 2025. I care more about what they had to spend to keep that demand moving. The earnings headlines made that harder to see.

I wouldn’t put these three stocks in a tidy “growth versus profit” chart. Uber generated $2.23 billion of free cash flow in one quarter. DoorDash was already profitable a year earlier. Lyft crossed $1 billion in trailing twelve-month free cash flow, not in Q3 alone. Each number is real. They answer different questions.

Uber’s miss doesn’t prove robotaxis ate the margin

Uber’s Q3 release shows $49.7 billion in gross bookings, up 21% from a year earlier. Trips rose 22%. Revenue was $13.47 billion, up 20%, while income from operations reached $1.11 billion, up only 5%. Adjusted EBITDA rose 33% to $2.26 billion.

That spread between revenue growth and operating-income growth deserves attention. It doesn’t, by itself, tell us autonomous vehicle partnerships caused an earnings miss. The filing doesn’t assign the gap to the Lucid–Nuro deal, driver EV grants, or AI data-labeling work. General and administrative expense, for example, rose from $630 million to $1.18 billion year over year. I’d rather start with the expense lines we can see than give every dollar of weaker GAAP leverage to a robotaxi story.

Uber can afford to place several bets at once. That’s the advantage. The risk is that “we’re investing in the future” becomes an all-purpose answer to a present-day cost question. My test is simple: do those bets eventually add trips or lower the cost of serving each trip? A partnership announcement doesn’t pass that test.

DoorDash’s profit wasn’t new; the spending plan was

DoorDash reported $244 million in Q3 net income attributable to common stockholders. It earned $162 million in the same quarter of 2024. Calling Q3 2025 its “first profit as a public company” was wrong.

Orders rose 21% to 776 million, and marketplace gross order value rose 25% to $25.0 billion. Revenue grew 27% to $3.45 billion. Net revenue margin moved from 13.5% to 13.8%. DoorDash said advertising, fewer credits and refunds, and lower Dasher costs as a share of order value helped that margin. That’s more useful than saying people like delivery: it identifies where the extra money came from.

Management also said it expected to invest several hundred million dollars more in 2026 than in 2025 on new initiatives and a global technology platform. That is an increase over an existing spending base, not a total 2026 budget. The platform is meant to bring DoorDash, Wolt, and Deliveroo onto shared technology. Autonomous delivery is one experiment within a much wider plan.

I think building common infrastructure across acquired businesses is a better use of money than running three separate stacks forever. But management should earn that argument with evidence. If engineering and integration costs climb while order economics stop improving, “long term investment” will just be a nicer name for overhead. The 13.8% net revenue margin is the number I’d keep next to the spending promises.

Lyft bought itself room to choose

Lyft’s Q3 results put free cash flow at $277.8 million for the quarter and $1.03 billion for the trailing twelve months. Rides reached 248.8 million, up 15%, and active riders reached 28.7 million, up 18%. Net income was $46.1 million.

That’s a stronger story than an earnings-per-share surprise in either direction. The cash gives Lyft choices: improve service, fund partnerships, or absorb a bad quarter without reaching for the emergency financing drawer. Its TBR acquisition and Waymo partnership are examples of where it chose to go. The cash-flow milestone didn’t “enable” either deal on its own.

I still wouldn’t call Lyft’s model settled. More rides can also mean more insurance exposure and more support work. Its 10-Q discusses higher commercial auto insurance rates per mile alongside increased ride volume. A cash milestone is worth celebrating. It isn’t permission to stop reading the cost lines.

The comparison I actually want next quarter

These platforms don’t report one clean, comparable “cost per trip.” Their businesses and accounting differ. So I watch the closest operating evidence each gives us: Uber’s operating expenses against trips and bookings; DoorDash’s Dasher cost and net revenue margin against order value; Lyft’s insurance and free cash flow against rides.

I’m also skeptical of the easy labor-to-robot story. California’s AB 1340 created a bargaining path for covered rideshare drivers. It did not set a new wage overnight, and it did not cover every DoorDash courier. A robotaxi or delivery robot still needs a vehicle, maintenance, charging, insurance, and someone to handle the physical problems the software can’t. The broader gig-platform picture will turn on those details.

Demand gave all three companies room to work in Q3 2025. What they do with that room is the story. I want receipts for lower service costs, not another slide about the future of mobility.