How come it’s still so hard to find a parking spot in San Francisco?

It’s by now an almost tired refrain. A new transportation startup launches, promising to fundamentally disrupt transportation and “replace cars”. The company raises millions of dollars and sees exponential growth. People love the service and become hooked on using it.

And yet despite being the global epicenter of new mobility services, it still takes 12 minutes to find a parking spot in San Francisco (3rd worst in the US) and our traffic congestion is the 5th worst in the world. What gives?

We’re clearly not post-car yet.

To most people who are used to driving a car for every trip, you might think that whatever is coming after the car will be one service, especially with all the hype and perceived imminence of autonomous vehicles. Any day now, you’ll be able to just tap a button and your roboUber will drive you anywhere you want to go.

But that’s not exactly how it works. The future will be an integrated network of differentiated transportation solutions, each optimized for a different type of trip, and you will subscribe to a bundle of services that meets your needs, much like a cell phone plan. Because most of your needs will be met by rideshare, bikeshare, scootershare, transit, etc. you will only need a car occasionally for “longest mile” trips.

Let’s unpack these one at a time.

Optimized Options

You don’t use a hammer for a screw. So why drive a car to go 1 mile? What we are beginning to see is the unbundling of different types of trips and development of new services optimized for specific trips. Here are some core transportation situations that are being reinvented:

transportation options being reinvented

Many services are developing better solutions for all types of trips, with most of the focus on commuting, 1 way, and last mile trips, understandably as these are frequent trips that haven’t been well solved to date.

But we build ovens to roast a turkey on Thanksgiving day and our mall parking lots for Black Friday crowds. People buy SUVs for the occasional trip to the mountains. You have to build for every use case, particularly the largest use case, or no one will buy your solution.

For example, we built electric cars to solve 80% of your needs- commuting and shorter trips around town. What happened? Hardly 1% of cars sold in the US are electric. We left people to fend for themselves (typically with a second gas powered car) for the “longest mile” trips. We are only now getting electric cars with 250+ miles of battery range, which people hope will be game changing- if we can get the vehicle prices competitive and ubiquitous quick charging infrastructure. Of course, a better solution would be a smaller electric vehicle, optimized for short distance trips in town- and you just get access to a full sized hybrid vehicle when you need to go a longer distance.

We need to rethink the transportation network holistically, not just put new technology or batteries into cars.

Currently, we are building a new transportation paradigm using the same logic - and getting the same outcome. We are focusing on the frequent shorter trips, neglecting the long distance trips, and not thinking holistically. Hardly 1% of trips in America are made by any shared transportation solution and even in San Francisco only 9% of trips are made by rideshare.

Clearly rideshare alone is only part of the solution, and on its own rideshare is primarily replacing transit, taxis, walking and biking. Actually, more people are making trips they wouldn’t have otherwise made than replacing car trips! Bikeshare and scootershare data are also only solving shorter trips- direct competitors with rideshare on trips under 4 miles, which is about 70% of rideshare trips. There is even compelling data that rideshare is actually adding to congestion. Drivers are actually incentivized to drive at rush hour downtown with surge pricing, which is the best time to make money if don’t want to give up your weekend nights or have drunk people puking in your car.

Rideshare will eventually be an efficient part of the network- once the vehicles are custom built, electrified, and automated. Robomicrocars won’t ‘deadhead’ as many miles, take breaks, or drive in from Sacramento. We can optimize vehicles to share rides efficiently at low cost. It will be great. Someday.

I’ve spoken with 1,000s of people who still own cars in cities despite all the new options and asked them why they still own a car. Sure, there are a handful who just love their cars. Some people still have horses too. But invariably, the answer is “to get out of the city”. And, currently, there is no reliable, affordable, convenient way to do that.

People own cars in cities primarily because there is no reliable, affordable, convenient way to get out of the city.

So what happens? People can’t get rid of their cars because they still have a need for them. And cars are a mountain of sunk, fixed costs even if you don’t drive often: car payment, insurance, parking, maintenance, depreciation, registration, cleaning… AAA estimates owning a car costs $706/month, and that’s assuming you live in middle America. It doesn’t consider extra costs in cities like San Francisco of $350/month for parking, our now $4/gallon gas prices, or $150/month average city insurance costs. It can easily cost you over $1,000/month for a car you rarely use. Once you have a car though, especially if you have a parking space, you feel more inclined to use it even if there is a better alternative solution for the trip you are making- because you've already paid for it. Using an alternative is just added cost. 

The whole network is only as strong as its weakest link. And if we don't have a complete network, it's not much better than having no network at all.

And we all subsidize this madness - even the 31% of people in cities who don’t even own a car. Anyone who doesn’t want to pay for a garage can park on the street for a highly subsidized $128/year. That’s the cost of two cups of coffee each month! And that’s in a city where the average 1 bedroom apartment right above that parking spot is $3,700/month! The fair market value of street parking in San Francisco is (at least) $117/month - about 10x the current cost. That’s just 50¢/hour between Monday through Saturday from 9am - 6pm, the cheapest parking meter rate in San Francisco. If we started to charge a fair market rate for street parking, most of the cars parked on the street would be sold in a month.

By my estimate, there are about 233,117 underutilized cars - around 36% of the city population - parked on the streets of San Francisco that can be replaced by charging market rate for parking (the stick) and by providing a more compelling “longest mile” solution (the carrot) integrated into a holistic network of solutions that meets all your needs. There are about 275,000 on-street parking spaces in San Francisco. That means we have an opportunity to effectively eliminate street parking and replace it with bike/scooter lanes or parking, rideshare pickup zones, bus lanes, parklets to enjoy a coffee... Virtually anything would be better than a car that just sits there idle 99% of the time.

We have an opportunity today to eliminate street parking by building a better "longest mile" solution and integrating it into a network of transportation services.

“Longest Mile” Solution

Upshift focuses on the trips that can’t easily be done without having a full size car all to yourself: longer trips like going for a hike, surfing, golfing, meeting people outside of the city, carrying lots of stuff, shopping, skiing, kids, dogs…

So what is wrong with existing solutions to address the longest mile trips? When it comes to transportation, people want to move like water. Whatever the path of least resistance, they will take. Assuming people think the solution is safe to use and socially acceptable, people will take the option they perceive to be the most convenient, reliable, and affordable.

We provide a service that feels like owning a car. You always get an identical car whenever and wherever you need one. It's just like having your own car in your garage but more flexible, affordable, and convenient.

People want to move like water, taking the path of least resistance.

The problem with existing “longest mile” solutions is they are inconvenient, unreliable, and expensive. Rental car is far away from neighborhoods where people live, is closed evenings and weekends, and is too much of a hassle for day trips. Carshare has taken away much of the transaction cost with vehicles available 24/7 within walking distance- but if the car nearest you isn’t available you will need to get across town to another car. Cars may also be dirty, low on fuel and, in peer-to-peer carshare, poorly maintained. Carshare is pretty pricey too, especially for day trips, so most people won’t use it for more than a few hours- and then they still have to go to rental car for multi-day trips. Sometimes they even go out to the airport because the rental shops downtown are all closed weekends. That’s crazy.

Carshare has been shown to eliminate 9-13 cars for every car shared and eliminate 0.84 tons of annual greenhouse gas emissions per household. That’s amazing data. What replaces owning a car isn't going to be a better taxi or transit service, or easy access to a bicycle or scooter. What replaces owning a car is accessing a car.

And what's even better is that shared fleet cars are newer and get better fuel economy. They get used at vastly higher rates. When costs are variable (not fixed), people tend to ‘trip chain’ (add trips together- like a weekend outing + shopping) and use a more convenient and climate friendly solution for shorter trips in town that don't require a car.

Every shared car replaces up to 13 cars and 0.84 tons of annual greenhouse gas emissions per household. We need to eliminate friction to scale up this impact.

But there is so much friction with carshare today that only the 31% of people who never drive actually use it. It’s effectively "car insurance". As in, insurance for that one time a year when you actually need a car for something. 82% of people using carshare drive <2,000 miles/year. These are people who essentially never drive that probably wouldn’t have ever bought a car. 36% of people in cities drive more than 2,000 miles but less than a 10,000 mile lease. Most people driving over 2,000 miles/year just buy a car and rarely drive it. These are the people who sign up for Metromile's mileage based insurance - or put their car onto a peer-to-peer carshare platform. It's possible that these options actually incentivize car ownership by providing new ways to reduce the cost of owning a car you rarely drive. Some people, for instance, are now buying cars to put onto carshare platforms like Getaround and Turo, including some people who are becoming micro fleet operators!

We built Upshift to provide the freedom of owning a car, but better. Never worry about parking- just get a car delivered whenever you need one and it goes away when you are done. Every car is an identical Prius. After driving it a few times, you know how to synch your phone and how it accelerates and handles. You don’t drive a different car every day when you own a car- that would be maddening. That's why frequent users on carshare platforms always try to get the same car. If you need a car regularly you want consistency, not variety.  

In the future, we expect a car will arrive at your door, take you to your destination outside of the city for the day, and then disappear when you return. It will reliably be available, maintained, clean, and sufficiently fueled or charged to get you to your destination and back. In short, it will be far more convenient, affordable, and flexible than owning a car.

This may sound like rideshare, but we expect the vehicle - and service - you want for a <25 MPH, 15 minute, 2 mile trip across town to be very different from the one you want to spend the whole day with your family or friends driving mostly on the highway - maybe even with a surfboard or a dog. The city rideshare trip might be in a city car (eg, a smart car like Daimler Car2Go) or even a custom built microcar like Google Waymo. The longer trip will need to be a larger format vehicle with very different design requirements. A service like Upshift could even replace the need for existing carshare and rental car models once it is autonomous. Once the cost of delivery is eliminated, even shorter hourly trips are possible (eg, picking up that new dresser at IKEA).

Integrated Network

What comes after car ownership will need to replace all of people’s trips and it will integrate them all into a single holistic solution. I actually started building this in 2011 before realizing we didn’t have all the pieces to integrate. It would be like having a puzzle with only half the pieces. Not much fun to assemble. But today we have nearly all the pieces. Now we need to start putting them together.

You will access all these services from one central place, which will help you choose which option is most convenient. That place could be Uber, Lyft, Google Maps or new services like Migo, Whim, or Upshift. You might take a shared scooter to the train to get to work. Then maybe take a rideshare home from happy hour in a custom autonomous microvehicle. And on the weekend, you can take an Upshift to do a hike in nature with your partner and swing by the grocery store on the way home.

We’ll also need a new business model. I don’t pay every time I drive my car. Driving is free. Most of the costs are spread all over so it’s impossible to tell what it costs, certainly not on a per trip basis. I have a monthly car payment, insurance payment, and parking garage cost. I pay for gas whenever it’s low. Maintenance, damages, and tickets virtually fall out of the sky unpredictably. Depreciation is like a vampire that virtually steals money from me suddenly when I try to resell my asset.

Owning a car is expensive, with costs spread out over a variety of things that are hard to calculate. But driving any given trip is free.

Just as there are a variety of cell phone plans depending on how much talk, text, and data you need, you will be able to bundle all the transportation services into a monthly “mobility plan” that meets your personal needs. It will be one transparent, flexible, and all-inclusive price with no negotiations, and no long term contracts.

There will be some people who rarely use these services that can opt for pay-per-use. Just like some people get a pay-as-you-go cell phone and leave it in the car glove compartment just for emergencies. Do you still know anyone like that? I don't. Most people will just subscribe to the mobility plan that best meets their needs. This will make any given trip feel ‘free’, just as driving today feels free.

Your next car will fit in your pocket.