Eric Goldwyn is a research scholar at New York University’s Marron Institute and an affiliated faculty fellow at NYU Shanghai. His writing on cities has previously appeared in New York magazine and the New Yorker online.
Cab-sharing has arrived in a serious way with UberPool and Lyft Line.
After a decade of false starts, smartphone-enabled cab-sharing has finally arrived. In an ongoing quest to drive down the cost of a ride, attract more customers, increase efficiency, and take over the personal travel market, Lyft and Uber have both rolled out beta services in San Francisco that match users going in the same direction: UberPool and Lyft Line. Building on their popular and easy to use e-hailing platforms and sizable networks of passengers and drivers, both services are betting they have the right formula to move more people at a lower price.
On a recent trip to San Francisco, I took three Lyft Line rides. Here's how it works: You sign into the app, select a line ride, and input your destination. As Lyft builds your route, it simultaneously searches its database for other riders going in the same direction. If the system finds a match, Lyft will send you a note about your vehicle, driver, and fellow passenger. If no match is found, you ride alone, and Lyft picks up half the tab. One of my San Francisco rides matched (Hi, Clara!), one didn't, and one did match but the passenger never showed. In all these cases, my rides cost 50 percent less than a solo ride would have, and got me where I needed to go without adding much delay. Drivers are instructed to wait no more than a minute for the matching rider.
Coordinating moving parts is complicated, and since Lyft Line is a new service, it wasn't entirely seamless. Finding a match took longer than just clicking to request a normal Lyft—about a minute or two for my first ride. In addition to this minor hiccup, there was some geo-location confusion that delayed us by another 10 minutes. Fortunately, the driver called me directly and we sorted out the problem. Despite this delay, the ride was social and fun. Between the three of us, we turned a routine ride from the airport into a conversation about Nairobi's climate, California's drought, chocolate, and alternative corporate structures. The experience was so pleasant that I ended up having breakfast with my co-passenger the morning I left for New York.
For my other rides, Lyft picked up the unmatched (or no-show) difference and only charged me the 50 percent fare. Thanks to hundreds of millions of dollars raised by Lyft, they're able to subsidize rides and pay drivers the full fares. This last bit is critical to sustaining the cab-share idea. "We're in a fortunate position to launch Lyft Line on top of a strong platform and guarantee the fare so that there are no surprises for our passengers or drivers," says Logan Green, Lyft's co-founder and CEO.
Entrepreneurs have been circling around the cab-share idea for years. The opportunity was obvious: cheaper cab rides for customers. The allure of cab-share spawned new start-ups by the week in the late 2000s. Companies like Weeels, CabCorner, Split-A-Ride, Ride Amigos, CabEasy, and Hitchster all launched expecting consumers to gravitate to their apps and inherently understand and adopt this new service.
This group of forward thinking developers failed by skipping directly to sharing before establishing a reliable network of drivers and passengers—they had a service, but no marketplace to connect supply and demand. Lyft and Uber, meanwhile, have spent the last few years building just such a network with their regular e-hailing platforms. "E-hailing was the easier problem to solve in 2009," says David Mahfouda, co-founder and CEO of Bandwagon (formerly Weeels), an early cab-share app. "We were trying to solve a harder problem, and it's taken us a couple of years to figure it out and prove our concept." (Full disclosure: I once did a consulting project for Bandwagon.)
Instead of repeating the same mistakes of the early cab-share pioneers, Lyft and Uber are charting a very different course, relying on their deep pockets, reams of data, and a careful rebranding of cab-share (note that UberPool and Lyft Line avoid using the words "cab" or "share") to make it a reality. Instead of asking riders to adopt an unknown service, Uber and Lyft are nudging their existing passengers to share their rides by offering them a lower price when they split a fare. Even when these rides don't match, Lyft and Uber will pick up the difference so that the rider isn't stuck paying the second passenger's fare.
The lesson is that new transportation services require outreach, education, and branding to attract users and explain how they work. Even e-hailing required hundreds of millions of dollars in free credits for riders, waived commissions for drivers, and clever marketing and design to convince people to try it and stick with it. Uber, Lyft, Sidecar, and HailO didn't just develop an app and wait for people to figure out how to use it. They incentivized them to use the app and then held their hand along the way to make sure they got the most out of the service so that they'd keep coming back for more. "Almost $2 billion has been invested in making this feasible," says Mahfouda. This sea change in behavior didn't just happen over night; it was engineered.
While Lyft was designed with cab-share in mind—Lyft's emphasis on fostering a relaxed social atmosphere really speaks to this—years of data-collection from their e-haling platform has confirmed the intuition that gave rise to Zimride, the initial company started by Lyft's founders, Green and John Zimmer. Green told me "90 percent of Lyft rides in San Francisco could be shared if people are willing to wait an extra 5 minutes." (Other new research suggests that nearly 95 percent of rides in New York City could be shared with just a minute's wait.) But matching rides is only half the battle. As Lyft works through the inevitable kinks that come with introducing a new service, they're also building a robust real-time traffic data center that collects information from its cars on the road. The better they can anticipate traffic patterns, the better their software can route trips, find matches, and provide certainty to drivers and passengers.
In the hands of the current crop of technology companies, the cab-share idea is much more ambitious than people even realize. Ultimately, Lyft and Uber want to corner the personal transportation market rather than the much smaller taxi or carpooling markets. Uber's most recent advertisements—at least the ones targeted to me via Twitter—are about avoiding the hassles of owning a car rather than improving my taxi-riding experience. Green told me that the idea for Zimride (which launched six months before the iPhone did, back in 2007) grew out of his experience on the board of Santa Barbara's Metropolitan Transit District. "Public transit wasn't improving, agencies were bleeding money, subsidy was limited, cost recovery was around 30 percent, and there was no platform to scale up," he said. "I wanted to create an infinitely scalable public transit service." With Lyft Line, it's possible he has.