I was sweaty, smelly, coming out of jiu jitsu practice. Two days after Christmas, and San Francisco’s SoMa neighborhood was a ghost town. Lyft wasn’t coming through. Brand loyalty only goes so far — loads Uber.
Cesar pulled up. In the back seat were Alex and +1, who had to have met in the last 24 hours. They were still in that phase where narrating random photos on each others’ phones apparently was an interesting interaction.
Cesar drove straight up Turk towards NoPa. As we approached my apartment, I noticed something on his navigation…
The pin was at the corner
Cesar’s app was telling him to drop me off at the corner, three houses down from me. Not at my doorstep.
I must be late to the party, but this was so damn smart.
Round to the nearest corner. How much gas and time and life and drivers and EVERYTHING would that save? No, seriously, let’s get nerdy:
- Assume each block in SF has ~8 addresses
- Assume that 6 of those are not on a corner: that’s 75%
- Assume that 33% of the time, a driver turning into a block to “be at your doorstep” is not on the way to the next passenger (a car approaching an intersection has roughly three options).
- Assume that every time a driver has to turn into a block to “be at your doorstep” that’s an extra 0.1 miles they add to a trip
…that’s…
0.025 miles saved per Uber pool passenger. On the urban grid, that’s around 1–2 minutes. Not bad. (Please poke holes in that math; I’m going for an order of magnitude.)
I had always wondered when ride-sharing companies would start doing that. It’s like creating drivers out of thin air. But, there’s a clear downside too — you’re reducing your customer’s standard of living. And as we all know, paying customers will tolerate nothing less than a downgrade in service.
Getting dropped off on the corner, with all of its efficiency, hurt a little inside. It reminded me that I’m using the budget option. And that Uber pool will inevitably, gradually regress as close to public transportation as the demand curve will allow for that seductive $5.00.
However, before the insult could sink in, something even more interesting happened. I noticed on Cesar’s screen that although he was dropping me off on the corner, the app still asked him to turn right_._ He would drive right past my doorstep.
AND YET THE APP TOLD HIM TO DROP ME OFF ON THE CORNER.
Why would it ever do that? Why drop me off farther away, when you’re actually passing my house?
Because someone at Uber really thought this through.
See, whichever product manager (or vigilante engineer) made that brilliant decision, they realized that if sometimes drivers dropped us riders off on the corner, but other times they dropped you off at your doorstep, that would be super confusing to riders. During “corner dropoffs,” riders would assume their driver was being lazy at their expense.
This would make drivers look bad. Really bad. Like, ~50% of the time.
For as long as riders held some hope that they could be dropped off at their doorstep, it would drag out the disappointment. It would make it only harder to adjust to the new, lower standard of service.
And so, with that decision, Uber chose consistency over utility. In a two-sided platform, they prioritized consistent expectations between parties. They prioritized making drivers look good in front of their passengers. And yes, while far from altruistic, Uber did good by their drivers.
Uber chose all of these things over basic, logical user experience. And — I thought to myself as I trudged up the sidewalk— they chose right.
Know someone whose craft would benefit from this post? Click the 💚 below to surface this post to more professionals like yourself. Cue cheesy instructional animation: