I was visiting New York a couple of weeks ago, transferring from Amtrak to the PATH trains at Newark.  PATH takes you to Wall Street – the #1 financial center in the world – and yet the process of paying for my $2.75 PATH ticket was excruciating.  When I arrived at Newark, my colleague, who had arrived 30 minutes earlier on a different train, was still in line at a PATH farecard machine. Credit card transactions were taking minutes before being mysteriously denied.  The most efficient means of buying my ticket to the world’s financial center? Cash.  

With this story we inaugurate a “small ball” feature in which, as a counterweight to the hype surrounding radically transformative AI initiatives like self-driving cars, we highlight ideas for incremental improvements in life brought by analytics. (“Small ball” is a baseball strategy for scoring via a collection of small, controllable goals like stolen bases & sacrifices to advance runners, rather than waiting for less probable but bigger impacts from home runs.) Here’s one idea:

Better reporting for on-time airline performance  

One of the most stressful things in air travel is  wondering whether you will make your connection. A 1-hour delay on a flight to your ultimate destination is a nuisance.  A delay enroute to your connection is a huge stressor – “Are there any more flights to my destination? Is there space on them?  Will they hold my connection? Will I get stranded someplace for days because all the flights are booked?” If you arrive on time, inconveniences suffered enroute may be forgiven.  If you are stranded or suffer a major delay, snack vouchers and polite gate agents do little to lighten your mood.

Airlines have sophisticated systems for handling passengers who have missed their connections or are on canceled flights, but their systems for minimizing missed connections in the first place are not nearly so sophisticated.  Consider this flight that you can book today on United:

depart Washington Dulles 5:20 pm

connect in Newark  (1 hour 5 minutes layover)

arrive Madrid 9:20 am next day

What you don’t know is that the first flight is late 58% of the time, delays lasting 73 minutes on average.  United knows, from prior data, that passengers will miss their connection half the time – United is selling a service that it knows, beforehand, is only 50% likely to be delivered.  So why offer it?

The existing on-time reporting system for the passenger, focusing on individual flight segments, is not helpful for the hub-and-spoke system that relies on connecting flights.  The solution is to make the passenger experience of on-time arrival at their ultimate destination a centerpiece of measurement and analysis. Airlines collect, disseminate, model and analyze vast quantities of data, including the on-time performance of specific flight segments, yet this key metric – predicting whether the customer will be delivered on time to their destination – seems to be missing.  

Calling all thinkers:  Send me your “small ball” idea here