After three years and nearly 300 rides, we learned a lot about roadside ridesharing. Here are some of the findings:

The power of hitchhiking

Because modern commutes are so varied, ride matching systems require a large critical mass of participants to get successful matches. Potential carpoolers become frustrated because they can't find an appropriate match. But the hitchhiker can attempt to find a ride with any passing driver who is going in the right direction. Skillful hitchhikers can tap into the stream of passing traffic any time they choose.

astonishingly easy

Nearly all of the riders discovered that getting a ride was not a problem. Wait times were consistently short-- an  average of seven minutes. Drivers were helpful and curious, and most responded positively to the project. Drivers reported that the sign showing the rider's destination influenced their decision to stop. 

more random than we expected

The majority of drivers who stopped were random strangers. We did not see a lot of repeat drivers, which we found surprising. Only about one in four was known to the rider, or had heard about the roadside ridesharing project. That means a full 75% of drivers stopped to help out a completely random stranger on the side of the road. Unfortunately, it was just a bit too random for the comfort of the riders. They wanted more control over who stopped to pick them up.

location matters

Riders had the shortest wait times when they stood at a good location. The best locations had a long line of sight, an optimal amount of traffic, a safe, easy and obvious place to pull over and a place down the road for the driver to turn around and come back for a rider.  Look on the map to see the good locations we identified in the Lawrence area.


Roadside ridesharing is no fun in the rain and snow and cold--and especially in the dark.  The roadside is also no place for small children or those who are not strong walkers.

a big jump off the high dive

In spite of the ease of getting a ride, many of the registered riders were reluctant to take trips on their own. There's no such thing as a free ride, but the costs for roadside ridesharing are not measured in dollars. One cost is the risk that nobody will stop, or that the driver may be unsafe or unpleasant. And hitchhiking takes a toll on a rider's ego because popular culture portrays it as an activity only for the destitute and desperate.  

What about a beacon?

A number of riders chose not to try out roadside ridesharing because they misunderstood how the system worked. They assumed there was a way to signal nearby drivers that they were waiting for a ride. We chose not to use that approach for this experiment for two reasons. First, we did not want the riders to lose the benefit of their compelling physical presence on the roadside --a key motivator for the driver.  Second, an app that allows a rider to give a shout out for a ride assumes there will be a very large number of drivers checking for riders. Otherwise the app will not be even remotely reliable--which gets right back to the critical mass problem.

Trade-offs: closed vs open systems

Ridesharing can take many forms. All versions match people who want a ride with a driver who is willing to give them a ride. How they differ is in the mechanism for finding matches.  

Long distance hitchhiking is the most flexible of ridesharing. A rider can get a ride with any passing driver, so there is no need for a critical mass of participants. These riders nearly always get rides with strangers, and sometimes must wait a long time before someone stops and offers a ride. This version of ridesharing is the least predictable, most risky, but the rider can set out whenever it suits, and a skillful hitchhiker can cover long distances for very little cost.  

Another version is utility hitchhiking. These riders are looking for rides in their local community, usually out of necessity. They still get rides with random passing strangers, but they are more likely to get picked up by someone they know. Wait times are usually shorter, especially if the rider uses a sign.

CarmaHop, Nederland Lift, Go Geronimo and the VAP project in Belgium are all projects that try to combine the flexibility of hitchhiking with community support and/or technology for local ridesharing. Marketing and public awareness are essential to the success of these programs because they give riders a legitimate context for soliciting a ride.

Slugging and Casual Carpool as practiced in Washington DC and San Francisco require a critical mass of users for optimal service. The incentive for the drivers is HOV3 access. These systems use dedicated pick-up and drop-off locations to match drivers with the two riders they need for the HOV3 lanes. The incentive for drivers is a significantly shorter wait time, and the rider gets a free ride. There is no governing body for these systems but a website gives information. Casual carpooling is extremely flexible and reliable, but it is very difficult to replicate.

Web and app based ridesharing and carpooling programs are the most secure because both riders and drivers must register and arrange rides ahead of time. These only work when there is a critical mass of participants and so are difficult to establish in new markets. These ride matching systems are generally very safe, and many have automatic payment systems and ratings for riders and drivers too.