Rental bikes

Better Bike Sharing

By Tom Vanderbilt

May 29, 2018

Editor's Note: Get outside! You have no excuse: The weather is nice and bike sharing programs are everywhere.

Which makes me wonder, are these programs as efficient, effective and profitable as they could be? Turns out the answer is "not always."

If we add a little behavioral science, however, they can be.

How far out of your way would you go to help another person?  What would make it worth your while to do so?

Last September, Motivate, the company that operates New York’s CitiBike system, as well as systems in San Francisco, Chicago, and other cities, launched “Bike Angels,” an incentive program that aims to get individual users to help improve an overall inefficient system. 

The program shows how a variety of tools can be applied to achieve results in a sphere, transportation, where behavior change is not easy.

For students and practitioners of behavioral change, transportation is a particularly tough nut to crack.  Transportation choices, tied to a complex nexus of availability, time and money, among other personal factors, are strongly habitual. In fact, there’s an old saw among planners that the most likely trip a person will take today is the trip they took yesterday.

Behavior is not always optimal. Research has shown drivers do not often take the most efficient routes. And bikes, by the way, are also not the only mode of transportation that get poorly distributed. According to one analysis, nearly one-third of all trucks in the U.S. are carrying nothing.

Getting people to change modes — say, from cars to public transit — is particularly challenging. There may simply be no feasible alternative, but even when there is, people may not be aware of it, or they find ways to rationalize their personal status quo. Studies have shown, for instance, that car drivers routinely underestimate car travel times, and overestimate the travel times of transit. Interestingly, when people move residences they are more amenable to other life changes too, including changing their mode of transportation. Changing context and environment is a strong spur to changing behavior. Think of how many people will ride a bike or public transit while on a holiday (or at Disneyworld) but not at home.    

For students and practitioners of behavioral change, transportation is a particularly tough nut to crack

Bikes get unevenly distributed for all kinds of reasons, most having to do with human behavior. It might rain in the morning, but clear up in the afternoon. Or, there are plenty of bikes at a downhill station but none at an uphill. Stations near parks are almost always empty (they get more tourist use), and stations that are more visible to more people get used more. But mostly it’s just the patterns of people’s daily lives.   

Here is a typical example from my life: I open the CitiBike app, checking on the availability of bikes at the station one block from my Brooklyn apartment. The icon is red: zero available bikes. No surprise, really, as people grab bikes early and ride them to work in Manhattan. The stations they ride them to typically suffer the opposite problem — rather than not enough bikes, there are not enough open bike docks.

To tackle these asymmetries in supply and demand, Motivate, like other bike-share companies, typically sends trucks around the city, physically moving bikes from one to station to another. “Rebalancing,” as it is called, is not only expensive and time-consuming (particularly in periods of peak congestion), it seems a bit off-message for a service designed in part to take cars off the road — for example, CitiBike measures impact in terms of pounds of CO2 reduced.

So Bike Angels, an opt-in program available to all CitiBike users, looks to get riders to rebalance the system, by offering points for picking up or dropping off bikes from certain stations.  These points, interestingly, largely go to encouraging more CitiBike use — collect 80 points in one month, for example, and you’ve earned four weeks of membership extension. There are also gift card rewards at higher levels and so on.  

Almost a year on, Bike Angels has proved remarkably successful (launched in New York, it’s since been expanded to San Francisco). According to Motivate’s Collin Waldoch, who manages the program, Angels account for an average one-third of rebalancing activity — on some days, as much as half. There’s nothing particularly revolutionary in the approach, behaviorally speaking, but the program shows how a variety of tools can be applied to achieve results in a sphere, in this case transportation, where behavior change is not easy.   

The thing about well-designed behavioral nudges, like optical illusions, is that even when you see right through them, you can still fall for them. 

Here are a few key lessons from my conversation with Waldoch. 

  1. Understand that different people have different motivations. So, incentives need to be tailored to users. For example, “Power Angels,” people who “want to earn a ton of points and get exercise.” Then there are those who will say “I’m slightly willing to adjust my pre-planned route.”   While some Bike Angels might be happy with lower-level rewards, Power Angels won’t just want to stack up numerous free day-passes they won’t be able to use. In addition to gift cards and other tiered incentives, there are more symbolic rewards, like a “Top 10” leaderboard, or a special white key that users only receive after accruing 400 points.
  2. Don’t forget about altruism. Monetary rewards and membership extensions are great, Waldoch says, but it’s clear that most people participating in the Bike Angels program are simply hoping to extend their CitiBike membership. “You can see a shelf of all the people who end at 20 points” — the plateau for a one-week extension — “rather than 19," Waldoch says. “People just want to keep riding more.” To reinforce that positive feeling (about themselves and CitiBike more generally), the e-mail notifications that come when a user has earned a point stress the pro-social impact, letting them know how many people they’ve helped and how many additional rides they’ve enabled.
  3. Harness the power of incremental change. Do you remember making snowmen when you were a kid? Making that first snowball takes a bit of effort, and it sort of rolled along clunkily in the beginning as it took shape. But once you got it rolling, it rapidly picked up snow, virtually by itself.  Waldoch says one general finding from the Angels program is that it's easier to get people to tweak their existing behavior (e.g., to get them to make more of a trip they already make), than to get them to alter their behavior (to make an entirely new trip).
  4. Make it easy. Is there any more fundamental lesson from behavioral economics? Motivate likes to keep Angels rewards as frictionless as possible. Waldoch stresses the importance of auto-rewarding. “People really don’t like having to redeem something; they feel it will fall through the cracks.”  Thanks to troves of data enabled by the bikes and users’ smartphones, Motivate can literally show users how to earn points. “We can say, It looks like on your past commutes, you could have earned points, and helped users, by adjusting your trip a couple of blocks.”
  5. Never stop experimenting. It’s become almost a cliché in the incentive world that “rewards work better when they’re random,” but that approach didn’t seem to work very well for the Bike Angels. With the program still in its infancy, Motivate has been tweaking as it goes. That elusive White Key? It will soon be made available to users who accrue 400 lifetime points, not just annual. (Motivate may have to watch out for “status creep,” with people hitting 400 in a year wanting a special platinum key.) Earlier in the year, Motivate tapped the altruism theme with an option for Angels to donate extra points to an initiative giving memberships to underserved communities. When I suggest a special sound (a round of applause?) to be played upon docking the bike on a points-earning trip, Waldoch nods knowingly, “I’ve got all kinds of pie-in-the-sky ideas.”   

The thing about well-designed behavioral nudges, like optical illusions, is that even when you see right through them, you can still fall for them. I know why Netflix opens the little box in the corner of the screen with the next episode of the series I am binging, even before I am finished with the current, but this doesn’t make me any less likely to keep binging on. So. I readily admit my own desire to earn that 400-point white key; sure, I’m being tempted into habitual behavior, but it’s behavior that’s not only good for me, but good for the planet.

By Tom Vanderbilt


Get the latest behavioral
science insights.