“Crowd Source”, the app: The Next Frontier for the MTA

Every superhero goes to where they’re needed when they heed the call. This time, our team aimed to be NYC’s superhero in solving the problem with subway travel:

I would get to my train station expecting to get on the train and get to my destination on time, but when I get to the station, there would be so many people already on the train that I and the others waiting wouldn’t be able to get on. If I had known this I could have gone to get a coffee first, taken an alternate train, or taken a cab.

With NYC’s population at 8.5 million and growing, it’s no surprise that the subway is super crowded throughout the day and night. But, as we surveyed and interviewed users, it became clear that the only thing worse than not being able to get on a train was the expectation that they would be able to.

User Research

Of the 43 users who took our initial survey, 95% ride the subway and 70% of those riders use it everyday. The busiest times of subway riding are what we deem “rush” hours; 8am and 6pm. Most of these users use Google Maps to chart their route and check on travel time. Additionally, 44% of our riders walk as an alternative. Whew. So, from this information, we can assume that:

  • People use apps for their commute
  • They need the affordability of the subway
  • Most of our users are riding the subway at the same time

Post interview, we set out to organize the common themes. The top 3 complaints of being a commuter were:

  1. Overcrowding
  2. Communication/Managing Expectations
  3. Delays

We also found that most people use an app even though 100% of the users live or at one point had lived in NYC. From our user interviews and survey, we were able to pull out 3 personas who represented the majority of our users.

What’s Already Out There

So remember how our users mostly use Google Maps for transit? Well, they also use NYC Subway, TransitApp, CityMapper, Hopstop and a few others. All of these apps have essentially the same goal: to give users real time updates on the trains and help them map routes. The apps use MTA’s API via Google for time tracking called GTFS (OMG!). But, HOW MANY APPS DO WE NEED BEFORE THE PROBLEM IS REALLY SOLVED?

This is where great UX comes in. It’s a world of over-saturation and when a team can come together, truly listen to the problem at hand and create a simple solution that stands out from the others, that’s amore.

The super fancy feature that will set Crowd Source apart from the other transit apps is one that addresses the crowd issue. We had the idea to notify users how crowded the next couple of trains are so that they could plan their trip accordingly, whether it be taking an alternate route or relaxing with a coffee until a ridable train rolls in.

You Get Out What You Put In: Crowdsourcing Location

Initially we planned on incorporating the D6T Infrared Sensors as our “population reporting” tool. These sensors detect body heat to determine capacity and the population within a space. Another solution was utilizing the weight sensors that are already installed in the train cars. After much brainstorming and iterating, we came to the conclusion that “Crowd Source” should utilize the simplest technology solution, one that is already in place; the users. This app will function on a location-based API similar to how Waze obtains information. The incentive to having the most useful version of our app is actually using it. This only requires that the app is turned on.

According to an article published by the MTA in January 2016:

More than 140 underground subway stations already have cellphone, data and Wi-Fi service, and the deployment of this enormously popular amenity will now be accelerated. All 277 underground subway stations will have Wi-Fi service by the end of 2016, and cellphone service will be available in all of them early in the following year.”

Governor Cuomo is underway in bringing the MTA into the digital age in full which means that data and useful information should be available to all users who choose to be informed.

Apple’s Core Location API locates the current position of the device and allows us to use that information in the app. More information can be found here.

We spoke to some developers about our potential solutions and decided to go with the crowdsourcing option. Why not just use what’s already there? It’s incentive for users to not only download the app but to actually use it. The more they report, the more they know.

WatchOS Companion App

Though wearables haven’t completely caught on yet, we wanted to be ahead of the curve. Many of the transit apps we looked into have a wearable companion and we thought it would be a great way to allow the user to receive notifications. We also wanted it super simple and straightforward. Our MVP for the Apple Watch app was primarily notifications of crowd, trip time and alternate routes. We dabbled with the idea of a map but didn’t think the small platform warranted the squint.

Sketching, Prototyping and Testing

We put our heads together, combined our sketches and set out for the MVP.

Iteration #1 (iOS)

1st Iteration Testing

1st Iteration (iOS) Results, 3 Users

  • “Crowd” icon should be on alt. route lines
  • Users are used to Google Maps
  • “Directions” button was confusing, users wanted to swipe alt. routes because the page looked like “Apple Wallet.”
  • User knows where they are so they tap their destination
  • Users confused by delay and time frame context (what are the delays? how much time do I have?)

Iteration #1 (WatchOS)

Iteration #1 (WatchOS) Results, 2 Users

  • Users confused by delay and time frame context (what are the delays? how much time do I have?)
  • Users are used to Google Maps
  • Users would respond to alert colors instead of words, particularly on the smaller platform
  • Wearables don’t require a “dismiss” button
  • Train logo should replace the name of train

Iteration #2 (iOS)

Iteration #2 (iOS) Results

  • Learnability was easy
  • Stop sign alert icon was distracting/more noticeable than crowd count and users wanted to tap it
  • Users want to tap route line on map to get directions
  • Be very clear with descriptive wording or take it out all together
  • Nav icons must have clear destinations-we don’t need a “home” button or home page

Iteration #2 (WatchOS)

Iteration #2 (WatchOS) Results

  • Learnability was easy
  • Colors make sense


To wrap it up, I should mention that we couldn’t ignore Apple’s HIG (Human Interaction Guidelines) and made affordances based on those.

Feel free to delight yourselves with our prototypes (aka 3rd iteration of iOS and 2nd of WatchOS) and get to know the first MTA app that gives you the option to avoid throwing elbows. And then get sad that it doesn’t actually exist yet…

One clap, two clap, three clap, forty?

By clapping more or less, you can signal to us which stories really stand out.