Recently I downloaded an app that was supposed to help me find people to carpool with. I'm not someone who relishes the idea of coordinating with strangers for my commute but, still, I wanted to give it a shot. For two weeks, I asked drivers—many of whom, the app indicated, had signed up but never actually accepted any carpool requests—for rides between my home in San Francisco's Castro neighborhood and my office South of Market. When that failed, I tried to use the app to get to Silicon Valley. Nope. Then within Silicon Valley. Crickets. Finally, one morning I got a notification that someone named Krishna was offering to give me a ride. By the time I navigated into the app, Krishna's request had disappeared.
This unfortunate performance wouldn't have been surprising if this app were the work of an obscure startup. But it's actually an attempt by a navigation giant to find a new way of fulfilling its founding mission. The giant: Waze. Its mission: To vanquish traffic.
Waze chalked up my experience to, among other things, the enormity of the task it had set for itself. “Getting people to carpool is HARD,” Waze spokesperson Terry Wei replied when I emailed her about it. “It's asking people to change their routine and behavior.” It's also, in a funny way, an attempt to recapture some of Waze's initial success.
When Waze launched in 2009, what distinguished the Tel Aviv startup from its peers was that it aimed to crowdsource its digital maps rather than simply buy them. The company made a prediction that CEO Noam Bardin says got his team “laughed out the door by every industry expert”: It wagered that every smartphone would soon have a GPS chip and that every car would have a smartphone-toting person inside. And the people in those cars, Waze believed, would be eager to use those new tools to collaborate and do great things for the world—for free!
In exchange for streams of personal location data provided by those very GPS chips, Waze users would receive up-to-date info on traffic. Ultimately, they would also get access to the company's complex routing algorithms—traffic models that tell you when to leave here to get to there, and which roads to take. But there was a hitch: First Waze had to persuade people to sign up. The product would work only if Waze could collect data on billions and billions of miles of road. Which is why Bardin has joked about something called Tim Cook Day.
On September 28, 2012—a day of infamy in the world of online cartography—the Apple CEO posted a rare public apology for the state of his company's newest venture, Maps. The app was notoriously buggy and riddled with inaccuracies. Cook wrote that Apple was working to make it better. But in the meantime, he suggested, people should go ahead and use some of Apple Maps' rivals, including a three-year-old company called Waze.
More than 20 million people had downloaded Waze's app at that point, but Cook's name-drop gave the company's user numbers a shot in the arm. In 2013, Waze was acquired by Google for a reported $1.3 billion.
Today, Waze has 475 employees, 115 million monthly active users, and some 30,000 volunteer map editors. This “community,” as Waze likes to call it, throws itself into one common goal that's something between an objective and a crusade. “Traffic is a global evil,” Waze writes on its website. “Only we, the People, can get ourselves out of this mess.”
Lately, though, Waze's war on traffic has hit a roadblock. After years of pouring unexpected traffic onto local streets (to the frequent consternation of local governments and residents), the company is running out of empty roads and workarounds for users. In many cities, traffic has gotten worse. And autonomous vehicles—which Waze's sister company, Waymo, is reportedly spending more than a billion dollars to develop—will only make the problem worse. “With self-driving cars, people will drive longer distances and will care less,” Bardin says. Which means more traffic.
It turns out that savvy programming, algorithms, and enthusiastic bands of community mappers can take a crusade only so far. So Waze has committed to a more, well, behavioral fix. It wants to consolidate more people into fewer cars—to get them to, ugh, carpool.
Waze rolled out its carpool app across the US last fall, after more than two years of testing. Today, a rider can use it to get matched with a driver who's headed mostly in the same direction during their commute; in exchange, they'll pay the driver up to 58 cents a mile (the standard IRS mileage deduction for business). Drivers are rated, as on Uber or Lyft, but they're limited to two money-making drives a day, at a maximum payout of $25 per trip. A rainmaker for car owners this is not. Waze, meanwhile, makes nada: “Our goal right now is to get as many people carpooling as possible,” according to Terry Wei.
Susan Shaheen, a transportation researcher at UC Berkeley, is all for it. Her research has shown that packing more people into fewer vehicles can come with long lists of social and environmental benefits; the trick is getting people to do it. “What would incentivize one to carpool in the absence of a wage?” Shaheen asks. “That's the question.” Or, to slightly rephrase the query: What would it take to give carpooling its own shot in the arm, à la Tim Cook Day?
Carpooling has had its moments in American history, mostly when resources were dear—during the world wars, during the oil crisis in the mid-1970s. Today, people in some metros like Washington, DC, and the Bay Area use informal “slug lines,” places where drivers pick up extra passengers in exchange for toll-sharing and access to HOV lanes. Yet these options are driven by a very particular combination of sticks and carrots. Drivers pick up passengers because they'll be penalized—by a dramatically slower trip or higher tolls—if they don't. Take away those incentives, and people drive alone. Overall, the share of Americans who carpool to work has dropped, from 13.4 percent in 1990 to 9 percent in 2016.
Waze seems to believe it can do its part with carrots alone. It's optimizing the carpool experience with algorithms that match riders with drivers traveling along their routes, and pinpointing ideal pickup spots; using customization options that allow riders to, say, restrict their drivers to a certain gender; and by releasing tools that let coworkers and friends coordinate regular rides. And in typically upbeat fashion, Waze is also seeking to harness the fraternal energy of its community to help users make ... friends? “We believe in microcommunities of people that live next to you,” Bardin says. “We're connecting you, you know each other, and then you're responsible for each other. You never leave each other stranded.”
That's a nice idea, right? That we can save the planet and bust traffic together? But my own experience suggests it's not enough. (Waze won't release many metrics on Carpool; it has said it increased its carpools globally by 26 percent from January to February.)
What Waze really needs are sticks to go along with those carrots—and it so happens that cities around the world are lining up to supply them. Places like London, Paris, and São Paulo, Brazil, now ban certain cars from certain roads at certain times of day, or require drivers to pay dearly to use them. Closer to home, LA and New York City are considering congestion pricing. Policies like these might put the squeeze on drivers and send them to to the friendliest, smartest carpooling app. In fact, in the first two months of this year, Waze Carpool use in Brazil grew by 53 percent. With any luck, that kind of pressure might sway someone to pick me up someday soon.
Aarian Marshall (@AarianMarshall) is a staff writer for WIRED's Transportation channel.
This article appears in the May issue. Subscribe now.
- Kitty Hawk, flying cars, and the challenges of ‘Going 3D’
- EV upstart Rivian is helping to drive a new auto tech boom
- EV and self-driving tech let auto designers go wild
- Are we there yet? A reality check on self-driving cars
- LA’s plan to reboot its bus system—using cellphone data
