>Uber is a $3.5 billion lesson in building for how the world *should* work instead of optimizing for how the world *does* work. – @levie
I remember seeing that quote from Aaron Levie and thinking it was great. It’s exactly how I think of things. We entered the travel industry with people constantly telling us “this is the way things work and you won’t be able to change it.”
We were firm believers that things should work differently.
A good example is that for many ground transportation bookings you need to select a hotel from a dropdown list. Or you need to figure out the zipcode of where you are going. General informational searches aren’t possible, if you want to figure out the general price to get from SF to SFO you can’t search for that, you need a specific place in SF.
We thought that was wrong, and we have spent a lot of time integrating Google Places, messing with geolocation libraries so no matter where you search from we can give you options.
That was a good choice: we thought there was no reason why this shouldn’t be possible and went after it.
However, I think it’s important to remember that sometimes things work the way they do for a reason.
We tried to redesign the user experience for how we thought it should be. We thought that we could just put in two search boxes and that they would type in their start address and the airport and it would be just that simple. We thought that the dropdowns that our partners/competitors used to select an airport were unnecessary, why make them select an airport, they could just type in LAX and that’d be easy enough, right? Why make them specify “To the Airport” or “From the Airport,” we could figure that out easily based on whether or not the airport was first or second. Why do they ask for Flight Time when some people cut their flight times close and leave 1.5 hours before and others need to be there 3 hours before to check in. We could just ask for what time they wanted to leave, wouldn’t that make more sense?
What we found out AB Testing was that there was a reason all these companies operated the way they did.
Shuttle and Limo companies need to know the flight time to make sure to get them there on time, and suggested pickup times vary depending on the flight time.
When we did usability testing people did not know the airport code or the name of the airport and Google wouldn’t recognize where they were trying to go.
We tried similar design tweaks on our results page. We didn’t think people needed to see what the van or limo car looked like. **Wrong. A lot of people asked.**
We just went through a redesign and I took a look and our homepage looks remarkably like one of our partner’s pages, and our search page has a lot of similarities to a competitor of ours.
After all that fretting we kind of ended up back where we started, with the types of forms and results pages the rest of the industry was using.
The moral of the story: yes, it’s important to imagine how things SHOULD work instead of settling for how they DO work, but at the same time there are often reasons for industry norms.
People in your industry may be entrenched in a certain way of thinking, but they often do know some things you don’t know, so before you try to reinvent an industry try to understand what those might be so you don’t waste time on the things that you will end up doing the normal way anyways.
David Litwak 