Uber’s chief product officer on hotels, robotaxis, and why the company doesn’t want to be “everything to everyone”

Uber has spent the past year quietly pushing beyond the two businesses most people associate with it. There’s pickup, yes, and delivery, but spend some time in the app and you’ll now find Expedia-sponsored hotel reservations, concierge “shop for me” features, and boat rentals in Europe.

Under the hood, so to speak, there is also a lot going on. Think debit cards for drivers, a data-tagging side hustle for those earners looking to make some extra moolah, and a six-month-old business unit called AV Labs, which is building a fleet of sensor-equipped cars with Uber’s regular driver network and designed to collect ever-larger amounts of driving data. Uber is pitching the program as a way to strengthen its relationship with its autonomous vehicle partners, several of which hold equity, but look like hedges, too. Uber competes directly with some of those same partners, with Waymo chief among them, and having the data layer gives Uber both power and choice.

Whether Uber becomes a full-fledged “everything app” like other top Asian apps like Grab, is still an open question. But in this interview, Uber’s Chief Product Officer, Sachin Kansal, walks TechCrunch through the financial services company’s ambitions, its strained relationship with Waymo, its new AV Labs data operation, and how AI is starting to show up in ways that riders and drivers will notice.

This interview is edited for length and clarity.

TC: You unveiled hotels, boat rentals, and other shopping features earlier this year. How was that list made, and what didn’t make it?

SK: Every year our teams obviously create a lot of things, and it’s a piece of that that we decide is worth sharing with the world on the biggest stage. This year our focus theme was travel. 1.5 billion trips on the Uber platform every year actually take place outside of the user’s home city, so we know that travel is the most common activity for Uber users. Our headline announcement this time was actually introducing hotels to Uber as a partnership with Expedia. But traveling is more than that – you need a ride to get from the airport to the hotel, and you need food. We heard from many of our users that many of them stopped using room service and were instead using the Uber Eats app. With “shop for me,” the intent was to be able to shop at any local store even if that store is not available on Uber Eats through the entire catalog. Travel is really, in my opinion, the third leg of the saddle – we were on board, then we added food, and now we add travel.

Is Uber looking to offer its financial services, the way “every app” in Asia is doing?

Financial services for us cut across many different organizations – consumers, but also drivers and messengers, and sellers. We have many products today that are focused on drivers and messengers, where we have what we call the Uber Pro card, which they can use as a debit card and transfer all their income. We’re starting to test some of those vendor products in certain parts of the world right now. In terms of consumers, we’ll see if that makes sense for us over time. Currently there is a currency that consumers can use – we call it Uber credits – and this is related to our membership program. At hotels, for example, members get 10% cash back on $1,000, that’s $100 cash back as a credit that you can use on rides or dining.

Can Uber offer its buy now, pay with the latest product?

I’m not sure, because we want to make sure that experts do what experts do. We’ve already announced partnerships with others in the industry who already provide that service, so when you go out you’ll have the ability to do that. In our general brand strategy, we don’t try to be everything to everyone.

For boat rentals, in Europe, tapping a tab directs users to the partner’s booking method rather than checking within Uber. Is that handoff model a template for what’s to come?

There are certainly some cases, especially when we do something new, to rely on our partners, because combining two things takes a lot of time, and in some cases it is good to try before we integrate deeply. In the case of Expedia, we decided that it made sense to integrate deeply – we built the entire UI ourselves in cooperation with Expedia. But in some cases it may make sense for us to transfer all the experience of experts in that field, and if you get good traction, we can always integrate it deeply.

Your Uber One membership product now has 51 million members and accounts for nearly half of its bookings. Do you have data that shows that cross-selling really works – that the delivery user later starts to ride more?

On the delivery side, it takes you two to three orders to break even on the monthly fee you pay. As members become familiar with the program, it increases their frequency within the line of business they already use. And it leads to more usage on other sides of the business – we’re seeing people who only travel and start using delivery, and people who only deliver also start using travel.

Delivery has been one of the most difficult businesses in technology to make a profit. Does Uber Eats still rely on rides to stay healthy?

During the early years of Uber Eats it was not profitable yet, but in the last few quarters, Uber Eats has been a profitable business for us, and it generates a lot of profit.

A story I wrote this spring featured Uber as an unexpected direct competitor to Airbnb, which now offers a flight partner. Do you see it that way? Who you focused on?

There is no shortage of competitors – Lyft in the US, Didi and 99 in Latin America, Bolt, Ola around the world, and for delivery, DoorDash, Delivery Hero. But I spend very little of my time thinking about that. A large percentage of my time, or what keeps me up at night, is giving our users all the value we can.

You just dropped off a Waymo driver in Phoenix while you were measuring elsewhere. How do you keep the transaction consistent when you partner with – and in some cities compete with – the same provider?

Phoenix was the first city where we introduced Waymo, with about a dozen vehicles, but the introduction of our scales was in Austin and Atlanta, where we also have hundreds of vehicles. When we just watched the pilot of Phoenix, we both decided it didn’t make sense to continue. Waymo is a great partner, but in many cities they are also competitors. We are not in the race to be the independent L4 provider – our focus is to lay down the race tracks to enable multiplayer. We believe in a hybrid network, human drivers and autonomous vehicles in the same city, because it allows us to balance demand and supply.

In terms of AV labs, what can Uber offer their independent partners that they don’t already have?

We will be equipping hundreds of vehicles with sensors, operated by our fleet partners, and thereby collecting millions of miles worth of driving data. That really helps with the long tail problem – you want to see all the cases, not just the P95, P99 level. Beyond the data itself, there is a wealth of information from 10 million beneficiaries about how to take and leave what works. We handle 25 million lost items every year – how do you handle that in the freelance world? That’s the kind of operational technology we can deliver.

Is Uber selling driver and passenger data to Gen AI companies?

I can divide this into two parts. In terms of Gen AI companies, we’re able to label the data using a per-user basis, or through audio collection, and yes, we have a commercial relationship with them and sell it to them — that’s part of the new business, and we’re very passionate about it. AV labs are different, and we are still looking for those models to share that data with our partners. It’s still small.

Are drivers recording conversations with passengers in this data valid?

No, no, no — I want to be clear, there is no conversation being recorded as part of that while they’re on board. When they are not on the trip, the drivers, the delivery people, just talk, or listen to a piece of audio and write it down. They get paid to do that, of course.

Where has AI really manifested itself in ways that a passenger or driver can see?

If you’re a lead person on our platform, we have a lead assistant – the first question on their mind is how do I make more money, and it’s going to say, look, it’s really bright in the South Bay, but you might want to go five miles where there’s a lot of demand. On the Eats side, there’s a grocery cart assistant where you can say “I want milk, eggs, bread” and it creates a cart very quickly. And in boarding, you can use voice to request a ride – say “I want a ride to the airport, I have six pieces of luggage, six people.”

So a fully functional Uber – “plan and book all my trips” – is around the corner?

I can’t put a date on it, and I can’t tell you exactly what the feature set will be, but I think AI will be a big source of that, where I can leave the complexity on the platform and just tell the agent what exactly I want. It’s easier said than done — we want to make sure we’re not just checking a box by sending an agent who might not be performing as well.

As a CPO, how do you prioritize with so many ideas on the plane?

I would say I spend 70% to 80% of my time making sure our existing products, or products we’re about to launch, are as robust as possible. All new ideas are like shiny objects – if you have 100 ideas, maybe five of them are good, and those five require a lot of investment and conviction. So about 20% of the time is on new ideas – including, by the way, I go out and drive and bring them, just to see our product from another side.