Scientists Claim to Have Cracked How to Beat Uber's Surge Pricing Algorithm
Northeastern University computer scientists studied Uber's surge pricing.
— -- Researchers say they've found a way to avoid higher "surge" pricing from car service company Uber.
The computer scientists at Northeastern University said they collected four weeks of data from Uber by analyzing its surge price algorithm. The researchers tried to re-create 43 Uber smartphone app users, with each requesting rides in downtown San Francisco and midtown Manhattan. They said they wanted to figure out how Uber calculates its surge pricing and understand how the system is "effective at mitigating" supply and demand imbalances.
What they discovered was that Uber app prices are updated every five minutes and the company sets prices according to fixed areas. In places like Manhattan, where the areas are small, the researchers suggest users simply avoid surge pricing by walking to a different area where surge pricing isn't in effect. This is more effective in Manhattan than in San Francisco, where Uber's fixed areas are larger, according to the findings, made public on Thursday.
When the researchers' Uber users simply walked to another area, they were able to pay about half of the non-surge pricing in more than 50 percent of cases.
The researchers' findings also included the following:
1. San Francisco has three times more surges than Manhattan even though the former "has a much greater supply of cars."
2. The majority of surges last less than 10 minutes.
3. Surge prices "have a strong, negative impact on passenger demand, and a weak, positive impact on car supply. However, it is possible that different effects may occur at the macro-scale (i.e., a whole city)."
Uber said it disagrees with the report's findings, and that it finds that surge pricing does in fact help supply issues and gives incentives to drivers to meet high demand.
Last month, Uber released a case study of surge pricing on New Year's Eve last year, when a technical glitch caused surge pricing in New York City to fail for 26 minutes. Without surge pricing that night, "ride requests skyrocketed and only 25 percent of these requests were completed," Uber said.
On the other hand, on the night of a sold-out concert at Madison Square Garden, surge pricing worked as intended. And "even though the number of people opening the Uber app experienced a 4x increase, the number of actual ride requests only rose slightly. In other words, people decided not to request a ride. Meanwhile, 100 percent of ride requests were completed and ETAs were virtually unaffected."
A spokeswoman for Uber said in a statement about the new analysis, “People love the ability to push a button and get a ride quickly and reliably -- wherever they are in a city. And dynamic or surge pricing helps make that possible because it encourages drivers to go to the neighborhoods with the highest demand -- ensuring there’s always a ride available within minutes. Contrary to the findings in this report, which are based on extremely limited, public data, we’ve seen this work in practice day in day out, in cities all around the world."