Can Uber’s new data tool change how cities plan transportation?

Few companies in the world have gathered as much transportation data over the last decade as Uber, which operates in over 500 cities on six continents. Yet up until now, Uber has been extremely recalcitrant about releasing that data—even to the cities it considers partners. Now, for the first time, Uber is making data for over two billion trips publicly available.

A new tool named Movement that launched over the weekend is Uber’s first public-facing data release (although it’s not officially open to the public yet—you still need to register to access it—but it will be in a few months). As The Verge reports, Movement is the work of a team at Uber named Project Metropolis, which is working specifically on civic relationships.

Uber has long cited passenger privacy as the reason for being so fiercely protective of its data. This has led to some major rifts with cities, which desperately need to understand how and where Uber is changing transportation habits. In New York City, for example, a case that begins hearings this week is asking Uber to provide the same type of detailed trip data that’s collected by taxis.

But Uber is getting around privacy concerns in Movement by anonymizing the data, meaning that it takes the GPS information collected for each trip and divides it into geographic “zones.” So what you see when you look at an Uber Movement map for any particular period is travel times for its trips between those zones. Nothing about where exactly the trip originated, or demographic information about who is taking the trip.

Uber Movement

Why release this information now? “We’ve gotten consistent feedback from cities we partner with that access to our aggregated data will inform decisions about how to adapt existing infrastructure and invest in future solutions to make our cities more efficient,” says the statement on the Uber Movement page, suggesting that so many of its cities have been requesting the same thing that the company might as well make it all public at once.

But the timing might be a bit more strategic: to get out in front of the first major study to gain access to Uber’s voluminous data cache.

After San Francisco’s request for trip data was denied, the city brokered a compromise: Uber would give its data to the University of California at Berkeley’s Transportation Sustainability Research Center and the National Resources Defense Council (NRDC), which would act as a third party to keep the actual information private but share big-picture findings with San Francisco planners. The study, which will look at Uber and Lyft’s impact on car ownership, carpooling, and emissions in the Bay Area, should be out this spring.

Uber Movement

Still, it’s hard to see if Uber Movement, as it’s presented, will have any real effect on transportation planning. Take one of the case studies presented on the site: the shutdown of Washington DC’s Metro after an electrical fire. Anecdotally, this was a huge day for Uber, as people turned to the rideshare company to supplement their commutes. But the trip data is so generalized that it really just shows overall traffic trends, something that DC probably analyzed using its own data, or data from more forthcoming transportation apps like Waze, which has been praised for the way it shares its data with cities.

Uber has a reputation for not always cooperating with city regulations—behavior which has caused it to leave a few cities. But even when it has agreed to work closely with cities, the partnerships haven’t always been fruitful. Boston made headlines in 2015 when it became the first city to gain access to Uber’s trip data, yet city officials have said recently that the data has not been as detailed or useful as the city would have liked.

Will cities really be able to use Uber Movement to plan for a ridesharing future, or is this just a feel-good PR move? It’s a well-designed tool, but let’s hope releasing this type of data is just the first step. Uber Movement needs to offer much, much more for cities to chew on.