Was your ride home a little cheaper than usual this weekend? If you're an Uber user, you're not imagining things.
Uber lowered prices in more than 100 U.S. and Canadian cities starting on Saturday, according to a statement on the company website.
The price cuts are an effort to boost rides in the "winter slump," when business slows down, the ride-sharing company stated.
For some cities, this is the third consecutive year that Uber has lowered prices in January.
"Seasonality affects every business, and Uber is no exception because when people hunker down at home, demand for rides drops," the company stated. "Fewer trips are tough on drivers, many of whom want to save money and pay off their holiday credit card bills now that January is here."
The size of the discount depends on the city.
"While pricing is a science, every city is different: different economic circumstances; different regulations; different competition," Uber stated, noting that in the past, price cuts haven't always worked.
"Last year, for example, earnings fell in some cities and we changed back," Uber stated. "In Charlotte, for example, we pulled a 40 percent price cut back to 29 percent, and earnings for drivers grew by nearly 20 percent in 2015. And in two cities, including Seattle, we ended up reversing the price cuts entirely when it became obvious that prices were already low enough."
In a pinch and need a lift? There's nothing more frustrating than getting slapped with an Uber fare four times the ride's usual cost. Well, what if waiting 5 minutes or walking a couple blocks could make the surge-charge disappear?
A study from Northeastern University researchers published Thursday, billed as "the first in-depth investigation of Uber," offers some tangible tips — and other fascinating insights — about the mysterious algorithm that guides the country's most popular ride-share service.
The bottom line: Uber's surge-pricing algorithm, which is based on supply of drivers versus demand of rides needed, resets about every five minutes, and changes based on zones that are often close together. That's just the knowledge you need to save a couple bucks (the study claims 10% to 20%) on your ride after the big game, impromptu thunderstorm, or any other instance where it feels like you and the rest of the world need a ride, stat.
"We see that around 40% of surges only last five minutes, while about 70% of surges last 10 minutes or less," Christo Wilson, an assistant professor at Northeastern and one of the writers of the study, tells the USA TODAY Network in an email. He is in Tokyo presenting the findings at the Internet Measurement Conference. "This essentially means that if you observe surge prices on Uber, your best bet is to just wait it out, because they typically don't last long."
"The other recommendation is to look at the prices being offered in adjacent surge areas. One of our key findings is that Uber divides cities up into areas, and that each area has a different surge multiplier," he says.
The study, "Peeking Beneath the Hood of Uber," gathered four weeks of data from Uber by emulating 43 copies of the Uber app on Android phones and distributing them throughout densely-populated downtown Manhattan and San Francisco. Each of Uber's ride-sharing options, which include everything from luxury black cars to shared carpools, were tested. (Wilson says the study did not test competitor Lyft because there's no way to predict fares without actually hailing a ride.)
"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," said a spokesperson for Uber, who clarified that all the data in the report, including information that surge pricing only sometimes lured a supply of more drivers, came from "limited," publicly available information and could not monitor certain nuances, like drivers who come back online when they notice an occasion that could present surge pricing in certain areas.
"Normally, when you open the Uber app, it downloads the eight closest cars to you, the estimated wait time for a ride, and the surge multiplier. The app then refreshes this data every five seconds," Wilson said. That, plus Uber's open API, was how he and his co-authors, Le Chen and Alan Mislove, conducted the research.
"We've seen this work in practice day in day out, in cities all around the world," Uber's statement continued.
Wilson says the study's findings are more important to users than helping them save on their next ride. Like Google altering search results based on location, or Facebook selecting what you see in your newsfeed, Uber's algorithms are a "black box" — not transparent to the user, or completely to the driver, for that matter.
Wilson says he hopes the results "help the public better understand and evaluate the role of these algorithmic systems in everyday life" — especially after criticism Uber has faced for charging surge rates during Hurricane Sandy and the Sydney, Australia, hostage crisis, the study notes.
Other insights from the study:
- Surge pricing does make demand for Ubers go down. People see a higher rate, and often choose to either wait or make alternate plans.
- The team found a six-month long bug in the system that was causing some people to randomly get lower fares than others requesting rides in the same area. (The team contacted Uber, which has since fixed the bug.)
- Plus, in the battle of East-West expensive cities, Wilson says New York City is the better market for riders. Because its prices may be consistently higher, there are fewer surge situations than in San Francisco.
