Last weekend I was visiting family in Baltimore, and I decided to take the train back. Normally I take MARC (Maryland Rail Commuter) service, but on nights and weekends, AMTRAC service between DC is all that is available.
Knowing that I would likely miss the train for the time closest to when I arrived at the station, I had already decided to wait out the hour or more gap until the next arrival. Upon reaching the ticket terminal, I noticed that I was still able to buy a ticket for the train that I was previously certain I would miss. Seeing that, I bought a ticket and rushed down the stairs. Train in sight, I went down.
Whatever train I was supposed to board was leaving at any second, and I knew that the number looked right, so I rushed as fast as I could to the train. The one other person on the other side of the platform, a conductor, yelled “This is not your train!” It was the only train in sight and I was pretty certain that it was mine, so I did the natural thing. I hopped on board, and decided to troubleshoot after.
I asked one or two people onboard if the train in fact went to Washington, and the consensus was yes. I plopped down in the dining lounge, and got back to work on my laptop. Eventually one of the conductors came by to apologize, and noted that because I had just bought a ticket, they could not see that I was supposed to board. Hearing this, I realized that their system must not be streaming in nature, but rather processing one batch driven report prior to boarding. In other words, they figure out who needs to be on the train, then keep selling tickets.
Last-minute ticket sales are luckily not the norm. Most people do not buy a $47 dollar ticket for 39 miles of transit with four minutes till departure. However, systems should be designed for intuition, rather than having engineered barriers.
When I got home, I thought of the alternative. Imagine if my last minute purchase buzzed the conductor with my information, that way he could direct me onboard. These kinds of small, intuitive changes are totally possible with the technology that we have today, but it requires openness to change. What we have today mostly works, but fixes like what I suggest are the kinds of small changes that accumulate into a first-class customer experience.