Since day one, Uber has been committed to changing people’s lives by revolutionizing urban transportation. When Uber was first launched in San Francisco, it was designed really just for the enjoyment of a few friends in our home city. But word spread quickly and people wanted to try this new idea for themselves. Today we are in over 300 cities in 56 countries across 6 continents.
But our vision is to help solve some of the most pressing problems cities around the globe are facing. Congestion is projected to get much worse over the next decade, and it’s barely tolerable now in many cities. The only real solution is to reduce the number of cars on the road.
Our carpooling service, uberPOOL, is helping to make that vision of fewer cars a reality. With uberPOOL you share the ride – and the cost – with another person who happens to be requesting a ride along a similar route. Riders can save up to 50% while adding only a few minutes of time per trip. With the lower prices, people can move past car ownership, as taking Uber becomes less expensive than using and maintaining a personal vehicle. And that impact on congestion can be powerful.
We first launched uberPOOL in San Francisco in August 2014. Since then, we’ve added POOL in Paris, New York and Los Angeles and have heard some great stories from our riders about their POOL experience (including an engagement!). And after making its debut at South By Southwest, uberPOOL is now permanently available during nighttime hours Thur. – Sat. in Austin.
We’ve seen millions of uberPOOL trips, and with thousands of users taking POOL trips during commute hours more than five times in a week, POOL is becoming a true commuting option for many people. In fact, in San Francisco, match rates from the Marina to the Financial District are over 90% during commuting hours.
As we passed the six month mark of uberPOOL, we took a deeper dive over a one month period to see what other patterns have emerged in how people POOL in the first four cities where it’s available. Here’s the poolside view of how people are incorporating uberPOOL into their daily lives.
The choropleth (heat) maps below show the most popular pickup and drop-off neighborhoods within each city by numeric count¹. The color scale runs from light tan to dark red. Darker red colors for each neighborhood designate that there are comparatively more matched uberPOOL trips originating there, or dropped off there². A bit of technical context: these maps show matched uberPOOL trips across all times of day between February 20 (the launch of uberPOOL LA) and March 20 (unmatched uberPOOL trips are effectively uberX trips). To make the maps, we associated anonymized pickup and dropoff GPS coordinates to neighborhood shapefiles using QGIS and OpenStreetMap.
The big picture takeaway here is that across all of our POOL cities and matched trips, pickups occur in a more concentrated set of neighborhoods than drop-offs do. That’s likely because many of these areas are where a high density of people pass through for work or play: cultural or business hubs such as the Financial District and SoMa (San Francisco), Midtown Manhattan (New York City), Bastille (Paris), and Santa Monica, USC and UCLA (Los Angeles).
Driving Saved: (Not) Going the Extra Mile
It’s not hard to see why over several hundred thousand different people have tried uberPOOL. Riders save money and, collectively, fewer miles are driven in cars on the road. We looked at how many miles — and metric tons of CO2 emissions — uberPOOL has saved in this one month time period in San Francisco alone.
For each matched trip, we took the anonymized pickup and drop-off coordinate pair for each rider and ran them through the Open Source Routing Machine (OSRM) to generate typical routes between these points. This gives an estimate of the distance that each rider’s individual trip would have taken, had they not used uberPOOL. Then, we took the pickup and drop-off coordinates in the order they occurred for the uberPOOL trip and ran them through OSRM, which gives us the approximate distance of the uberPOOL trip using the same benchmark.
The miles savings estimate for San Francisco is the distance difference between the sum of the individual rider routes and the uberPOOL route, about 674,000 miles in aggregate from February 20th to March 20th. This method accounts for any small amount of additional distance an uberPOOL trip went to accommodate the two rider routes.
Conservatively assuming that every SF uberX vehicle is a Toyota Prius — thus getting its gas mileage of 50 mpg — uberPOOL trips saved around 13,500 gasoline gallons. Accounting for a savings of 8.91 kg of atmospheric CO2 emissions per gallon, San Francisco’s uberPOOL efforts prevented about 120 metric tons of CO2 emissions from Feb-Mar 20th, equivalent to the output of over 128,000 pounds of coal. And every week as we keep growing, so do CO2 savings.
We’re excited at what we’ve seen from uberPOOL so far. As we continue into what’s been called “The Urban Century”, it’s clear that congestion problems will only get worse. Many cities may not keep pace with physical fixes — new roads, traffic rerouting, additions to public transportation infrastructure — without the technology that’s enabled an urban progression to occur in the first place.
A digital grid connecting carpooling, on a mass scale, could be one of the bigger effects that can bend the curve in favor of increased efficiency. Early returns suggest uberPOOL could scale and be part of the solution, and we’ll keep analyzing this product over time as we work towards a world with more efficient transportation options around the city.
Appendix
¹ Each trip can have multiple neighborhoods associated with it — for example, two drop-offs in two neighborhoods are counted as one apiece for each neighborhood. A matched pickup is a pickup location from a matched trip. A matched drop-off is a drop-off location from a matched trip. Since our goal is visualizing where most uberPOOL trips are, we don’t normalize by the number of people in a particular neighborhood.
² These maps don’t show match rates; a high match rate neighborhood will still display as light tan if the cumulative number of matched uberPOOL trips is lower than busier areas.