Ride-hailing drivers can spend months or years learning the best ways to make money. Now, there’s artificial intelligence for that.
Lyft has been rolling out over the past year Earnings Assistant, a AI tool that gives drivers tips on where and when to drive. It shows how the company hopes to use AI to help drivers earn more in an industry where they’re paid per trip.
The tool has two features. One, called plan guidance, gives drivers advice on where to go to increase their chances of claiming rides, especially if they’re planning to work on the app for a limited time. The option, available to drivers in the US, is meant to help new drivers find productive areas and times to work, a Lyft senior software engineer, Xiaoyi Duan, told Business Insider.
Another feature, called real-time guidance, pinpoints where Lyft customers need rides. Lyft is still testing this feature, which is live in a few cities.
“Drivers want to earn more, and they see various signals in the app, but those signals are not tailored to drivers’ personalized needs,” Duan said.
Lyft is using AI to synthesize the information — such as where ride requests are coming in or when they tend to spike in certain areas — and make personalized recommendations. That could mean pointing a driver toward a stadium when a concert is about to end or suggesting they pick up their last ride of the day at an airport because it’s the busiest time of day for arrivals.
Lyft and rival Uber each have millions of drivers globally, many of whom accept rides on both services, comparing the pay and features. Uber is beta-testing a similar AI feature to help drivers earn more on its app.
Ride-hailing drivers often learn how to maximize earnings on apps like Lyft and Uber as they gain experience. With millions of other people on the apps, though, claiming trips and making money can be competitive. And with no clear bosses or colleagues, figuring out how to make money on the apps is challenging.
Lyft created Earnings Assistant within the past two years, the company said. The company showed early versions of the tool to drivers at events in Dallas, Las Vegas, and Miami last fall. It also tested Earnings Assistant with drivers who offered rides near Santa Clara, California, for February’s Super Bowl.
The testing showed that experienced drivers often know to look for patterns that could help them earn more money, so they want more tips about where to go in the moment, Yuko Yamazaki, vice president and head of driver at Lyft, told Business Insider.
Many new Lyft drivers, meanwhile, are initially overwhelmed by the options when they start using the app, Yamazaki told Business Insider. Yamazaki said the app’s plan guidance focuses on these drivers and offers suggestions on where to work, such as, “Here is what the next two hours could look like.”
“The real-time signals are becoming more of an interest for drivers as they become more mature on our platform,” she said.
Duan said Lyft wants to expand the tool, which currently requires drivers to prompt it for suggestions. Future iterations could tell drivers they’re near a busy event as they pass it, or indicate how the weather is affecting opportunities to make money, she said.
“We are not trying to build an AI product because we want to use AI,” Duan said. “We’re trying to find what are the actual driver needs. That’s the fundamental thing.”
Do you have a story to share about Lyft, Uber, or another ride-hailing service? Contact this reporter at abitter@businessinsider.com or via encrypted messaging app Signal at 808-854-4501. Use a personal email address, a nonwork WiFi network, and a nonwork device; here’s our guide to sharing information securely.
Read the full article here
