Lessons from the mixed success of microtransit

By Eric Sundquist

Partly responding to the current downturn in transit ridership—some driven by transportation network companies’ competition—transit systems have begun to experiment with TNC-like services. Generically called “microtransit,” such systems have a lot in common with traditional pooled taxis and other shared and jitney services and, in theory, use new technology to provide efficient demand-responsive routing and/or scheduling. One example is the Washington area’s venerable sluglines. Such services, it is hoped, can provide mobility in underserved areas, while reducing costs, congestion, and emissions, as well as syncing with rather than cannibalizing fixed-route transit.

An International Transportation Forum study of Lisbon (pp 24-26 here), for example, suggests that combining shared-use taxis and minibuses with the existing rail and subway network would increase accessibility while reducing carbon emissions by a third and parking demand by 95 percent.

In the real world, however, microtransit is still a work in progress; in the United States the mode’s most prominent event was the start and then collapse of Bridj, though other experiments are ongoing.

A new report from the Eno Center for Transportation examines three U.S. microtransit pilot projects, including one that involves Bridj, for lessons learned. Two of the three projects, in Kansas City, MO, and Santa Clara, CA, died at the pilot phase, while a third, in Alameda-Contra Costa, CA, may become permanent.

A key takeaway seems obvious at first glance: “When considering the development and deployment of publicly operated microtransit services, [an] agency should prioritize customers’ needs over the novelty of new technology and think critically about how to design, develop, and deliver [a system] that puts the customer first.”

While such a statement sounds like motherhood and apple pie, at least two of the three pilots fell short of the standard. The Kansas City experiment was limited by the Bridj algorithm, which focused only on morning-inbound/evening-outbound commuting, and the Santa Clara pilot served a too-small geographic area and was only able to adjust that footprint as the program was coming to a close. In both cases, the services attracted miniscule ridership—a problem that might have been avoided had travel patterns and customer needs been better assessed upfront.

Other recommendations include:

  • Transportation agencies should consider defining performance more broadly than ridership and farebox recovery metrics, and should structure the service contract around continuously improving these metrics.
  • Transportation agencies should utilize a contracting process that empowers those most familiar with the project to make decisions outside of the standard bureaucratic processes in order to be able to fail fast and iterate quickly.
  • Transportation agencies should establish their goals or define hypotheses upfront and work with potential technology vendors to design a microtransit project within those parameters.
  • Transportation agencies should prioritize local, on-the-ground marketing and outreach upon launch of a new microtransit service.

Eric Sundquist is Director of SSTI.