Beyond our mobile application, we wanted to allow casual or one-time users to order rides in our car in Arlington, Texas, especially since we anticipated a large percentage of users to be visitors attending a sports event and wanting to avoid stadium parking hassles. Our goal therefore was to make something very clear and understandable for casual use.
We also need to check that the person approaching our car at the pickup is the one who reserved the ride, and the one whose turn is next in case there is a queue. We decided to handle both by making use of our side communication panels. We allow people to show an abbreviation of their name on the car so they know this is their ride (vs. other vans or that van at a later time). That is also where we show a random 4-digit PIN, which the intended passenger can enter on their phone to show it is them at the car.
However, unlike for those using our mobile application, there is no app to enter a PIN into which presents an extra challenge. We therefore created a few text messages which we send to the phone the passenger enters at the kiosk, first to update them of any time estimates and tell them when the car has arrived, and then to be a message to reply to with the PIN on the car to show they are the one outside.
The process for authenticating a passenger without a phone app or human driver became US Patent US010853629
I led this project from beginning to end, participating in defining requirements, creating the wireframes myself, overseeing the visual design (done by my talented visual designer Scott Rossi), leading usability testing of the kiosk, adjusting the design based on user feedback, and testing again repeatedly. I also worked throughout with development, prioritizing the features and issues with our users in mind.
Examples from Throughout Project:
Wireframes, SMS Message Design, and Flow Chart by Location:
Visual Design, Which the Implementation Matches:
We achieved 95% of people who said it was easy to book a ride in their surveys, despite the unusual interaction due to the current limitations, and we strive to get it higher still over time.