Human-centred Change at LaGuardia Airport

I was furious.

United Airlines had managed to get me to New York 20 minutes early and LaGuardia Airport (LGA, for the frequent flying cognoscente) helped me blow all my logistical winnings on the enormous queue for taxis. First world problems, I know, but they were the only ones on my plate that morning.

LGA is in the midst of spending $4 billion to will itself out of “third world” status and serve as a fitting gateway to America’s largest city. However, this massive construction effort has severely disrupted traffic flow for some time, routinely putting a dent in the passenger experience and leaving travelers with a bad taste of the Big Apple before their visit even begins.

The queue I reluctantly joined formed a tightly-coiled snake shape nearest the taxi dispatcher and extended its tail down the sidewalk for about 40-odd metres before doubling back on itself. Taxis were notable by their absence.

There were plenty of drivers offering their services at special, surge-time only prices for the most desperate among my fellow travelers. The rule of thumb appeared to be that if someone is going to ask for a lift to the city the driver might was well ask for the moon in return. The hustlers cast about for takers among the frustrated masses on the arrivals level offering rates at twice what a Yellow Cab would charge, all with a backing track over the loudspeakers advising passengers not to accept such rides. This, I have learned, is something only for the most hardened of New Yorkers who know what to include in the negotiated price (hint: Everything!).

As a Canadian, I am accustomed to queues. It is as much a part of the national character as ice hockey, apologizing, and a sense of beer superiority. I have been told that I look like someone who is waiting to go places and one of the points I had reached in this case was an equilibrium where people looked at me, saw that progress was impossible, and opted for alternative travel arrangements.

Still, I held firm.


Queue theory suggests I was experiencing a confluence of two phenomena:

  • Balking — People saw the backlog of passengers, mouthed some expletives and turned away to find an alternative solution.
  • Reneging — Travellers had plenty of time to download Uber or Lyft, learn the app, arrange a ride and release themselves before the costs of waiting sunk in.

Both of these were occurring behind my position. Nothing happened forward of it. Nonetheless, I was committed to getting a taxi. I was deeply invested in this particular hole and I was going to dig my way out in true Homer-ic (Simpson) fashion.

You know that both passengers and drivers are in trouble when the taxi dispatcher, clearly feeling the pressure, starts looking every few seconds at his wrist to see how much time has passed since the last cab came along. He is fully aware of the scores of irate passengers tutting and rolling their eyes at the situation. He is also not wearing a wristwatch. But I understand his behaviour. He isn’t looking at the elapsed time between taxis. He is trying to display empathy for the people in the queue. He wants to show them that he is impatient too!
However, none of this changed the prospects for the taxi hopefuls.

A t some point, a queue has got to go somewhere and it was announced that we had to shift our position down the walkway. In fact, our intrepid taxi marshal moved us three times. The first time was to give us hope. The second, suspicion. The third served as farce. People lost their places each time we migrated along the platform — something that prods at my very Canadian sense of justice when it comes to queues.

“We need to move you just one more stop down to meet the taxis”, he announced. 
 “We don’t believe you!”, we shouted back. “You promised us taxis before!”

Two weeks later, I braced myself for a similar experience.

United Airlines over-delivered once again and we were early in arriving at our gate.

As I walked to find the taxi rank I ran into Kevin, my colleague. He is one of those rarest of creatures — an American Airlines passenger that arrived in NYC on schedule.

This time we found a smaller line of people waiting at the usual location and were quickly escorted by a uniformed sherpa along the walkway. She indicated that we should wander over towards what looked like an improvised “shack” rather than a taxi stand. We found a makeshift shelter with numbered bays into which travelers were steered as clutches of Yellow Cabs moved into line.

We were intrigued!

Our dispatcher called the taxis out from their waiting area in batches of five or six, each one lining up in front of a numbered position to pick up their allotted passenger(s). They departed in the order in which they arrived (i.e., no leapfrogging in the line), smoothing and harmonising the flow of traffic — this at an airport where most other experiences tend to be rough and dissonant for passengers.

As each quintet of cabs departed the taxi stand for their respective missions around the city, more passengers were moved forward into the bays and another set of cars was summoned.

Within 5–10 minutes of meeting outside of Terminal B, my colleague and I were called over to bay #5 and, shortly after, were on our way into Manhattan.

I’ve tried to take a human-centred perspective in writing about this particular experience — adopting the vantage point of a person in medias res with respect to a process (or network of processes) in which they find themselves and in relation to other human beings each of whom have a role in making those processes work. In other words, the focus is on people interacting with other people to achieve some collective outcome — in this case, getting passengers from the terminal and into a taxi to move them on their journey.

Think of it as an exercise in the ethnography of operations.

As a life-long student of organizational change, this human-centred perspective illuminated a few insights, including:

⚀ People waiting in long lines for taxis are the quintessential conversion points for new ride sharing customers.

Rather than letting this particular pain point serve as an opportunity for unscrupulous drivers to take advantage of desperate passengers, how might ride-hailing services (e.g., Uber, Lyft or Arro) use this situation to promote their services (via a discount for first-time users) or help prospective users download the requisite app and coach them through the process of arranging their first ride? It’s very much a hands-on, labour-intensive approach, but it is also a much more effective and, in my opinion, ethical way for drivers to engage passengers looking for a ride and to potentially build a book of repeat customers.

⚁ When customer experience is bad, you need to intervene … and that requires the activation of change agents.

In the old days of top-down transformation programs you would hear managers say things like “This situation is not optimal … but stakeholders will just have to adjust through this transition period while we execute our plan.” This would result in, at best, some friendly posters seeking to educate people on the complex and vastly important work under way and that it is “merely a temporary inconvenience”. At worst, you would get total indifference.

However, on that hot day in August when airport construction resulted in maximum disruption for passengers, someone at LGA must have said: “This situation is unacceptable! How can we make a positive difference for airport passengers and improve the flow of taxis along the concourse?” Framing the challenge this way (“How can we …?”, “How might we …?” ) opened up space for change agents to devise and deploy a stop-gap measure that would take the edge off the poor passenger experience.

⚂ Going live with a good-enough improvement is better than waiting for the perfect solution.

Every morning, LGA has an opportunity to stage a passenger experience better than the day before. As Terminal B’s taxi queue reached the point of maximum frustration that August morning the airport could only improve — the questions were how and how quickly? The “taxi shack” wasn’t a polished solution. It wasn’t even painted! At the time there were no visuals to lead passengers to the new pick-up location. The numbers indicating where passengers had to line up looked as though they were bought at a local office supply store. It was a solution constructed with materials and skills not far from my own preferred methods (n.b. These involve, almost exclusively, plastic ties and duct tape).

But it was sturdy, functional and designed to work and generate feedback, not to look pretty or finalized. It is a good example of what has been called pretotyping — designing a solution that is good enough (and cheap enought) to test whether you have a valuable concept that can be developed further and potentially deployed more widely. The concept to be tested, in this case, was not the physical structure of the taxi station but a different process for matching passengers with taxis (see ⚃).

⚃ Solutions to problems that exist in one part of an organization can often be solutions to problems in other parts of the same organization.

My colleague has written elsewhere noting the practice of simultaneous (or parallel) divesting of personal belongings in the context of TSA security screening. This is a method being tested at airports such as Hartsfield-Jackson Atlanta International Airport. What LGA has managed to do with the taxi stand is install a solution analogous to a problem shared with the TSA — queue management. Passengers are moved into positions and taxis are called out in batches that match the number of waiting stations. Once the taxis are lined up, passengers can be loaded simultaneously (as opposed to sequentially) resulting in a higher average throughput for the queue.

This presents several possibilities for airports, at least in North America. If TSA and airport authorities require further proof that passenger processing solutions like this will work in their context, they can pretotype them at their respective taxi rank locations, observe behaviours and measure the results. If they want to engage in subtle, but effective, behavioural modification for passengers in advance of a shift in TSA pre-screening methods, the taxi queue offers an excellent opportunity for learning about what works (and what doesn’t).

In the end … it worked

In these early days of the LGA transformation effort it is difficult to visualize how it will all turn out. As I brace myself for my next trip to New York, I hold a sincere hope that my experience will at least be incrementally better than last time. Regardless, I remain impressed by how the change agents at LGA adapted to the traffic congestion challenge caused by the construction.

William Morris once said:

“Nothing useless can be truly beautiful.”

Or maybe it was Tony Wilson. Either way, I believe this useful interim solution offers an initial step towards the beautification of LGA’s passenger experience. It can only get better.

P.S. If anyone can think of a collective noun for a group of taxis that is better than “fleet”, “scrum” or “clutch”, please feel free to make a contribution at: