We (that is to say, Flywheel, my employer, for which I am not an official spokesperson) launched a feature today allowing advance booking rides to the airport in San Francisco. This was a feature I was initially fairly dubious about, and have come around to. It speaks to the complexity, I think, of the transportation space.
On a pure technology level, it seems inelegant. We have this considerable investment in the whole concept of “get a ride right now,” and in some ways it weakens the premise of “get a ride right now” to say, “but, no seriously, if you really need a ride at a very specific time, you should book it in advance.” I, and other engineers, initially argued that we should just continue to try to improve our core value proposition, rather than go into advance booking.
But air travel is sort of unique. You’ve paid hundreds or thousands of dollars for a plane ticket, and in order to get the value of that purchase, you need to arrive at the airport at a certain time. And the fact of the matter is that even though Flywheel (and Uber, and Lyft) can usually get you a ride in a few minutes, they sometimes don’t. Sometimes it takes 15 minutes. Or more.
Now, sure, you could just build some additional cushion into your schedule. Say, “Okay, I’ll take a ride ‘right now,’ but I’ll make my ‘right now’ be 15 or 20 or 30 minutes early, so that if I happen to be in time when it’s tough to get a ride, I’ll still make my plane.”
But air travel also already involves a lot of hurry-up-and-wait. You already need to arrive early to handle potential TSA security kabuki delays, and to handle the whole deal where airlines lie to you and claim that boarding will happen at 11:30 but actually 90% of the time, it’ll happen at 12:00. Stacking another 15 to 30 minutes onto that because you had to build in a buffer to deal with rides right now is just another hassle in the increasingly hassle-filled modern air travel experience.
Having the ability to say, “There will be a car waiting for me right at the time when I need one” is a welcome respite from hassle. And when you try it, it feels awfully good. Happily for the technologist in me, it actually also has some interesting engineering challenges in terms of making sure that we fulfill our promise, so while we may have sacrificed some conceptual elegance, the result is both a serious value for customers, and a fun project to work on.