Google Maps and Apple Maps are making parking harder than it has to be

It’s a familiar frustration for car owners: Before heading to a meeting downtown, you open a navigation app to ensure you’ll get there on time. Driving takes about as long as predicted, but you hadn’t planned for the hassle of parking. The closest lot turns out to be full, as are two others nearby. Anxiety rising, you finally find a spot further away and race several blocks to your appointment. When you arrive, you’re embarrassingly late.

Popular navigation apps like Google Maps and Apple Maps have given little guidance about parking, leaving users to fend for themselves as they decide where to hunt for a spot and how much time to budget for the search.

New research from MIT suggests that these services could take some of the guesswork out of parking by giving better advice. Doing so wouldn’t just lower stress levels; it could also help travelers waste fewer minutes cruising for a spot, thickening traffic and spewing pollution as they circle.

The new paper, entitled “Probability-Aware Parking Selection,” is written by Cathy Wu, a professor of civil and environmental engineering at MIT, along with Cameron Hickert, Sirui Li, and Zhengbing He. Wu says that she grew curious about navigation apps’ handling of parking when she noticed that they predict time spent walking to and from transit stops but not to and from parking spaces.

“What my colleagues and I have been calling ‘time to arrive’ is a metric that is not uniformly presented across modes,” Wu says. If drivers had more useful information, she and her coauthors wondered, could they make better travel decisions?

The answer they reached was a resounding “yes.” Accessing data from the city of Seattle about occupancy of paid parking lots, Wu and her colleagues constructed a model to guide hypothetical drivers toward the lot that best balances proximity to a chosen destination and the likelihood of finding a spot available. (Due to data limitations, their model assumes all parking costs the same.)

The researchers concluded that the potential time savings are substantial: as much as 35 minutes per trip when compared with driving toward the closest lot and waiting for a spot to become free. In the most congested environments, travelers’ total time to arrive drops by as much as two-thirds.

Taking the stress out of finding a parking spot

Adam Millard-Ball, a professor of urban planning at UCLA and codirector of the university’s Center for Parking Policy, was not involved in the study but agrees that navigation apps can skew travel decisions by presenting incomplete parking information. He notes that mapping tools ignore some common annoyances, such as the time drivers need to navigate a crowded, multi-storied garage to reach the level where remaining spaces are located (and exit such garages on foot, which can also take several minutes).

But in the United States, a country with an estimated two billion places to park—more than five for every man, woman, and child—scarcity is not always a problem. Millard-Ball notes that many car trips conclude without any parking headaches at all, including visits to shopping centers with a big lot and commutes to an office with reserved spots. According to a 2023 federal analysis, 7.3 percent and 6.8 percent of car trips in Seattle and Chicago, respectively, involve cruising for parking, although the share can be significantly higher in particular neighborhoods and at busy times. In bustling cities like New York, a third of all street traffic can consist of drivers looking for a spot.

Millard-Ball says that people taking unfamiliar trips could disproportionately benefit from upgraded guidance. “If you’re a visitor and you don’t know the area, you’re not sure where to park,” he says. “Finding a spot is going to take much longer.”

For instance, improved information would allow drivers to avoid leaving too little time for their trip, and it would prevent fruitless journeys to lots that are conveniently situated but already full. And if travelers were given true overall time to arrive at their destination by car—including minutes spent searching for a parking place and then walking from it—they might instead opt to take public transit or bike. Doing so would reduce urban congestion and particulate pollution.

Additional benefits down the road

Ball believes that reducing parking pain will be more than its own reward. Improved parking guidance could also bolster efforts to build safer and more enjoyable communities by repurposing street space away from automobiles. “Most public anxiety about parking—whether it’s about removing spaces on the street to make a bike lane or pressure to build more street parking—is really driven by this perception that parking is scarce,” he says. If it’s easier to find a spot, car owners may feel less determined to retain every last one.

But historically, navigation apps from Google and Apple have largely treated parking as an afterthought. Neither company currently attempts to quantify the expected time required to locate an available spot. Google Maps simply offers a parking difficulty icon (“easy” or “hard”) derived from the circling behavior of vehicles that have completed their journey.

Last week, Google Maps announced a redesign that incorporates Gemini, Google’s AI product, into vehicle navigation, including a feature that draws drivers’ attention to specific parking garages. Amanda Leicht Moore, director of Google Maps, says that the revamp allows users to “ask Gemini for hands-free tips while navigating and see nearby lots highlighted as you approach your destination.” Both Google Maps and Apple Maps also offer integrations with SpotHero that allows users to reserve and pay in advance for a spot in select lots. (Apple Maps did not respond to a request for comment.)

Wu says that she and her colleagues have shared their findings with navigation apps and received a positive response. In future research, she hopes to explore how improved travel information could reduce some of transportation’s ugliest externalities, like gridlock and emissions. She also plans to examine ways in which other kinds of travel information might be systematically flawed, such as ride-hail apps that misrepresent the time users will ultimately wait for a pickup.

It’s intuitive that better information would beget better travel choices. It might lead to better cities, too.