Mostly everyone who has used Uber has experienced that unfortunate day where prices spike and you just had to go with it (mine was Halloween, last year, stranded in Chapel Hill, North Carolina).
In a lot of ways, it’s easy to figure out when prices will surge. Being wary of your Uber reliance on holidays where a lot of people are out (read: New Years Eve) or really rainy days will insulate you from scarily large bills at the end of your trip.
But surges can occur for seemingly no reason sometimes, and there’s no public data from Uber on what will trigger them.
A Northeastern University study titled “Peeking Beneath the Hood of Uber” looked at Uber price surges to figure out how users can avoid them. It reconstructed Uber’s pricing system by mimicking 43 customers using the app and looking at Uber’s public interface for software developers.
The study found that there are areas in Manhattan and San Francisco that always have surges.
The good news: these areas are so small that, many times, crossing the street could be the difference between paying a surge or normal price.
The bad news: we can’t exactly breakdown what specific streets to avoid since the study doesn’t delve into that much detail.
But the study did cluster these locations to show what neighbourhoods are hot zones for surges. It found that the east and west villages in Manhattan always have a surge, and downtown San Francisco (specifically SoMa and Marina) always have surges.
People living in San Francisco have worse luck with surges than those in Manhattan. The study found that 86% of the time there is no surge in Manhattan, compared to 43% in San Francisco.
For those looking to steer clear of pricey Uber rides, avoiding these areas all together might be the best call. Another trick to consider would be calling an Uber for pick-up in a nearby area.
In more than 50% of cases, people who walked to a different location that was seven to nine minutes away to meet their car were able to reduce their savings by 50%. Because surge areas in San Francisco are larger, users had to walk a minimum of seven minutes for this to work (Manhattan users can walk as little as 5 minutes).
The study also noted that 90% of the time, surges had durations that were multiples of 5 minutes. In cases where there was a surge, 40% only lasted five minutes. So it is worth waiting it out.
But overall, the study has a dim view of Uber’s surge pricing methods.
“We argue that Uber’s reliance on discrete surge areas introduces unfairness into their system,” the study reads. “Two users standing a few meters apart may unknowingly receive dramatically different surge multipliers.”
Business Insider Emails & Alerts
Site highlights each day to your inbox.