I joined Uber as a researcher in August 2018. Within my first few days at Uber, I drove my car to a Greenlight Hub*, had it inspected, took the Seattle knowledge test, and became an Uber driver.
I sometimes joke to friends that I drive with Uber because I didn’t make enough money working for Uber in the office job. In fact, the idea of becoming an Uber driver arose even before I started to work at Uber.
At one point, Uber was a share-economy platform where people would offer spare resources (namely, their car seats) to others. However, now that the product has become more widely adopted, there are more and more “professional” drivers who drive Uber full-time. Over time, people, including myself, started to see and treat the driver as a service provider, instead of just another person who happens to have additional space in his/her car.
As an experiment, I wanted to know if people would treat me differently as a driver. I also wanted to know what expectations people have about their Uber drivers, and how they would react when things are a bit different from what they expected.
Before I had time to put this thought into action, I joined Uber as a User Researcher. This made driving for Uber an even more obvious thing to do. As a researcher, I am a big believer in “doing as much as you can to be the user.” Earlier in my career, as a consultant on a healthcare project, I took the client’s product (a health supplement) on a daily basis in order to immerse myself in the user’s experience. I wasn’t alone in this strategy: Some of my colleagues back then even went as far as buying a car when working on a car loan project. One male colleague used a female hair care product on himself for a shampoo project. This might sound crazy, but it did give us first-hand user insights that we otherwise would have been hard-pressed to learn.
This was how I became an Uber driver. On my first day, I drove 41.2 miles (including looking for rides), completed 5 trips, and made $51 including tips. It was an exciting and unique experience, and I learned a lot about the product and the mentality of Uber’s driver partners.
*Greenlight Hub is Uber’s service and inspection center for driver partners.
Learning 1: More hurdles than expected in the signup process
I submitted my driver application using the Uber rider app. The first thing that almost stopped me from becoming a driver was the requirement to go to a Greenlight Hub for a car inspection. Uber has two Greenlight Hubs in the Seattle area. Most of them are pretty far away from the city center and are open only during business hours. This is not very convenient for people with full-time jobs.
Between sign-up and inspection, I got several text messages from Uber to follow up with the application and ask if I needed help. Those constant check-ins definitely nudged me to complete the process.
When I finally had my car inspected at the Greenlight Hub, I was told on the spot that I also needed to take a “Seattle Knowledge Test.” Without any prior preparation, I was reluctant to take it. Fortunately, Uber provides a cheat sheet to help drivers pass the test, I passed it on the first try.
Afterward, a friendly Greenlight Hub expert sat down with me to collect my insurance documents. I had to search through my inbox for quite a while to find the required documents. I didn’t feel that there was enough prior communication on which documents needed to be brought to the inspection. By the time I found them all, the Greenlight Hub was already closed. The inspection center workers were nice enough to stay late for me, but I felt guilty about keeping them there.
Around 30 minutes after the documents were uploaded, I was declared “good to go” and officially became an Uber driver.
Or, am I? From Uber’s perspective, the story does not end here. A significant number of people who sign up to be drivers do not complete their first trip.
Thought: How might we reduce the friction of joining the platform and actually starting to drive, without reducing quality and standards?
Learning 2: More guidance could be provided during the first ride
There are many reasons why driver partners do not complete their first trip. In my case, I didn’t know how to drive strategically to make the experience rewarding.
It was 8 pm when I first got online and started to drive. I live in Capitol Hill, a lively area with lots of bars and restaurants. I thought it was a good time to start since people were going back home from dinner, and drunk riders were not out yet. I was looking forward to getting a request immediately.
However, I ended up looping around the neighborhood for almost 15 minutes with an empty car. I started to feel like I was wasting gas and it was not worth it, so I drove back to my apartment building. I didn’t go offline when I got out of my car, as part of me was still hoping I’d get a trip. Just as I entered my apartment, a request came in. It showed that I was three minutes away from the rider. In reality, it took me more than five minutes to get back to my car and get out of the garage. My first passenger ended up waiting for ten minutes to get his ride. Fortunately, he didn’t cancel and wasn’t upset.
Over time, I gradually learned from experience when and where to drive. It took me several days of driving to realize that there aren’t many requests around my neighborhood until after 10:30 pm.
This experience made me think: what if we provide more in-the-moment guidance and tips to new drivers? What if we prompted them with suggestions of a better time to drive, instead of just pushing them to complete the first trip?
In fact, Uber does provide some of this advice in the app’s educational content and email newsletters. However, like most driver partners, I went on the road without reading any of those.
Thought: How might we provide educational content in the moment to help driver partners make better decisions?
Learning 3: Driving is like gambling which can be addictive
Driving for Uber is like gambling: you never know when you will get a request, or where that request will take you.
Every time a request pops up, it feels like getting a winning line on the slot machine, especially for people like me who drive mostly during non-rush hours. Sometimes a request would finally come in just when I’d decided to quit driving for the night. It’s hard to resist the temptation to make more money, so I usually told myself, “just one more ride and then I’ll go home.” However, a lot of the time, the ride took me far out of my way, and I didn’t pick up any passengers on the ride home. As a result, I often ended up driving longer than I wanted for not much extra money.
Driver partners who chase quest promotions might experience this even more strongly. Some of them said that they were already tired and wanted to go home, but they ended up driving more in order to meet the target.
Being a driver myself also shed some light on why drivers selectively choose trip requests (known as cherry-picking) even during surge time. Cherry-picking is a known issue for us, but it wasn’t clear why drivers still cherry-pick during rush hours when every trip is surge pricing.
It became obvious to me one night when the entire Seattle area was surging. I got a normal (non-surge) request on my way driving toward downtown, where the surge price was the highest. Without thinking twice, I rejected the request. As I got closer to downtown, I got another request with 1.2x surge. I accepted it but immediately canceled it, thinking a better one would come soon. Then a 1.3x surge request showed up, but I did the same thing. Before realizing I was cherry-picking, I accepted and immediately canceled four requests in a row, until I got a 1.4x surge ride.
Did I make more money out of that trip? Actually, no. The 1.4x surge ride took me to a far and uneventful suburb. I came back to the city with an empty car and ended up making less than what I expected per hour.
Driving for Uber feels like gambling. You never know what you will get.
Thought: How might we provide more certainties in trip requests while preventing cherry-picking behaviors from happening?
Learning 4: Driving is a lonely job
It didn’t take me long to realize why many Uber drivers try to strike up a conversation during the ride. Driving is indeed a lonely job. Passengers commonly ignore drivers for the whole ride. This is especially true for pairs or groups of riders, who have each other to talk to.
One of my initial plans was to talk to users (my riders) and gather some anecdotal feedback from them. However, things didn’t work out as planned. I could tell that most of my riders just wanted some quiet time alone with their phones right after the initial greeting. I even ran small experiments by changing the clothes I wore in an effort to draw riders’ attention. Nonetheless, the results were still the same. I don’t blame the riders, though.When I ride in an Uber, I also want to stay quiet and utilize the ride time to work on my own stuff.
However, this also means that whenever you meet a genuinely nice rider, it can make your day. I still remember a rider asked me to stop driving and go home because “it can be dangerous for a woman to drive alone in the middle of the night.” I also like when riders said “drive safe!” when they got out of the car. Small gestures like these made me feel cared for.
Thought: In August 2017, Uber launched driver profile to help riders get to know more about their drivers. How might we better utilize features like this to lubricate the interaction between riders and drivers, without creating extra burdens on both ends?
As I am writing this, it has been a while since my first trip. I gradually decreased driving as I realized how tough a job it is. However, this experience helped me tap into the mindset of driver partners and allowed me to build a little bit of empathy with them. I’ve been encouraging many of my colleagues at Uber to start driving. The product manager I work with is also a driver. She and I set up a regular time for colleagues to drive along with us. The engineering manager on the team not only drives but also delivers food. He carefully documented the journey and hosted a presentation to share the experience and thoughts with the entire team.
Doing as much as you can to build empathy with users is key to designing good products. And this shouldn’t be difficult most of the time. We just need to find time and ways to do it.
As a side note, I like to drive with those neon lights to entertain my customers who are usually going home after a few drinks. Meanwhile, I play the“peaceful piano” playlist from Spotify to ensure that stay calm and we all have a peaceful ride home. :)
