Amtrak’s unfortunate trip planner
Hi, Brendan here. I’m Trillium’s data engineer and I love working on this side of transit, the side that figures out new ways to make using public transit a convenient joy. Being able to use the bus or train even if you primarily use other forms of transportation can be a freeing experience. Which is why I’ve found Amtrak’s trip planner to be so frustrating.
Recently, I tried to plan a trip to Seattle, WA from Portland, OR. I typed in “PDX” as my origin and “Seattle, WA” as my destination and filled in the dates. The screen below was what I got as my result.
Search results for query "PDX" to "Seattle, WA"
At first I thought that there were no trips available that day. This would most likely lose some users who were not committed to taking the train and had other transportation options. After clicking around a bit more and searching through the Amtrak national station list, I realized that there are two stations in the Seattle area and the one that returns when a customer searches for “Seattle, WA” is not served by the Starlight or Cascades routes, two of the most popular in the area. After choosing the other Seattle station, the trip returned fine, but Amtrak just added multiple customer-losing steps between a potential train rider and their ticket.
It seems like a major oversight, and it is, but it is not the only time that this has happened with the Amtrak trip planner. It’s even worse with the San Francisco Bay Area, which has more stations than Seattle. It seems like common sense to show a customer all the possible trips for their query area, as nearly all trip planners, from airlines to Google Maps, do. Amtrak doesn’t do that, though. What Amtrak ends up doing is providing a great example of the problems an organization creates and the revenue it loses when they design a customer interface without the customer in mind.
Plan a trip with conventional transit schedules and maps in just SEVEN (count ‘em) easy steps!
I work with transit schedules and maps on a daily basis, and, of course, my clients are transit agencies, and my colleagues are other transit consultants or vendors. As “transit people,” it’s easy to forget to forget that conventional schedules and maps aren’t clear or easy to use for regular people (more on that here).
I’ve been including a slide like this one in most of my presentations. This shows all seven steps necessary to plan a trip from my apartment in Portland, OR to the 16th Anual Northwest Tribal Transportation Symposium in Jantzen Beach last week. And this example is a comparative piece of cake — there are only two legs of the trip, both on the same agency’s service, TriMet, that offers easy-to-use maps and timetables. (Of course, passengers can also plan trips on TriMet at maps.google.com and maps.trimet.org.)
It’s easy, right!?!
Online information and its effect on ridership
The connection between easy-to-use customer information (schedules, maps, etc.) and transit ridership seems obvious. And online tools provide a powerful way to make transit information easier to use (remember: unfamiliar transit riders find conventional maps and schedules difficult to use; in fact, in one study, they failed to plan trips about half of the time).
I am searching for, and cataloging, studies and accounts of online transit information and its potential, or observed, impacts on ridership.
Today, I turned up “The Effect of ITS on Transit Ridership” (Abdel-Aty, Mohamed A.; P.P. Jovanis. ITS Quarterly. Vol. III. 1995). Unfortunately, the article is old (1995) and the ITS Quarterly is not published online. But the literature review paper (“The Factors Influencing Transit Ridership: A Review and Analysis of the Ridership Literature”) gives a few highlights:
ITS-delivered transit information might encourage shifts to transit: 58.7 percent of respondents were likely to use transit at least once per week given the availability of ITS-delivered transit information, and about half of the non-transit users who might consider transit would be more likely to use it if certain information items were available.
I wish the whole paper was available. Does anyone know of other research on the connection between online information, or customer information in general, and ridership?
Customer information: not a side dish but part of the main course
Since the fundamental purpose of a transit agency seems to be putting transit vehicles on the street, it’s easy to assume that basic service provision is a much more important than providing information about that service.
Inspired by More Riders, I would argue that service provision and providing customer information are equally fundamental to serving a community and providing mobility options. “Customer information” is the term transit agencies use for their public-facing information — route & system maps, timetables, etc.
Even if an agency offers a great system from a planning and operations perspective — frequent, on-time service on a dense network of well-connected routes — if their customer information doesn’t do the job of helping people use and benefit from this system easily, then the system is under performing its potential in crucial ways — in rider satisfaction and the number of passengers transported to jobs and services, for example.
Consider software as an analogy. Even if you offer powerful, feature-rich software, if it’s difficult to use most people won’t benefit much from it. With software and with transit, ease-of-use is crucial to overall usefulness.
