Maeg Keane

Maeg Keane

Facebook Twitter Google Print Friendly and PDF

Maeg Keane is a User Experience Designer and Researcher at the Customer Technology Department of the MBTA. We spoke to her about her work on Skate, a new tool being developed to support the MBTA’s bus inspectors as they do their jobs.

BARI: Can you start by just telling us a little bit about your background and what your role is at the T?

Maeg Keane: Sure! My role at the MBTA is as a user experience designer and researcher. I’ve been at the T for about two years. Before that, I was working more in web design at a media company and before that, nonprofits. When I started at the T, I hadn’t considered myself a user researcher but knew that I was interested in that type of work. I feel lucky that I’ve gotten to do a lot of research at the T — it’s been fantastic, and it’s confirmed that research is the part of design that I really care about.

BARI: Can you tell us about where at the T you work, and any current projects that you’re excited about?

MK: I work in a department called the Customer Technology Department (CTD). We’re a team of designers, engineers, content strategists, and product managers focused on using modern technology, design, and agile processes to improve the rider experience on the T.

We manage the website and the countdown clocks that you see in stations, for example, as well as the data feeds and APIs that power those products. The thing that I’m most excited about right now is work we’re doing on the bus system.I feel really passionately about this because I think there’s a ton of opportunity to improve the service we provide, and it serves a lot of people in Greater Boston. Right now, we have a team that’s making a product called Skate. There’s actually a Medium blog post about it. We are trying to give the people who manage our buses, especially inspectors who work at stations and on the road but also dispatchers at our control center,  the modern tools they need to make sure that the system runs in a reliable fashion. There’s a very elaborate dance that inspectors need to negotiate so that there’s always a driver and there’s always a bus when a bus is scheduled to leave.

Previously, inspectors were using technology that’s kind of clunky, and they’re also using a lot of paper. So we’re trying to give them a tool that streamlines all of that. It’s my job right now to be out interviewing inspectors in the field to figure out exactly what they need. So it’s not just a bunch of people in an office deciding what the people in the field need without actually addressing the problems that they face.

BARI: So would you say that the end user for Skate is dispatchers and bus drivers, not necessarily riders?

MK: The end users are bus inspectors and eventually, dispatchers. They’re the ones who keep things flowing and they address problems as they arise, like a disabled bus or if a driver has to leave early. However, at CTD, our ultimate goal is always to improve the transit system for the people who use it. Although MBTA riders aren’t going to ever see Skate, our goal is to build tools that result in better service for the rider, and we think this will. If it doesn’t, then we haven’t done our job.

We want to take the best and most current approach to technology — the tools and philosophies that tech firms and other private technology companies use —  to bear on the problems that frustrate our riders. We want to equip and empower our operational staff in the field. They genuinely care about riders, and have a lot of challenges in their day-to-day work managing bus operations. Running a more reliable bus system for riders is the goal, and helping inspectors with their job by providing a better, more thoughtful application of technology is the way we do it.

An application like Skate can also help us give better information to riders. We spend a lot of time and effort tweaking and improving our algorithms for live bus and rail arrival predictions. But running transit is a human exercise — we can’t just “math” our way to better predictions. If a bus official can use Skate to, for example, flag that they are holding a bus to even out spacing between vehicles, that’s information that we can pass on to riders in predictions.

BARI: How much have you seen the technology or the processes that the bus inspectors are using change in the time you’ve been working on this project? Where did you start versus where you are now?

MK: When we started, some people out in the field had pretty minimal technology. Some of them had handheld tablets that they would need to operate using a little pen, which were state of the art when we bought them years ago. They broke a lot and were hard to fix, so many inspectors had no technology at all to support them. So we have been giving them Android tablets that have the Skate software pre-installed. We’re moving towards equivalency right now–the folks who did have something in the field still have that, and the folks who didn’t have anything are at least brought up to that level. We’ve been able to prove that we can do this in-house. Plus, the design of Skate is more user-friendly that what inspectors had previously. It takes fewer clicks to get the most commonly required information and we’ve removed some features that no one uses.

Now we’re able to say, “OK, we’re giving you what you had before, and now we’re going to add features on top of that that you never had”. So the technology hasn’t changed much other than the fact that we have distributed these new tablets that are more like what you use in your own home versus something that’s been issued by a bureaucracy.

BARI: Looking at the photos in the Medium post about Skate, the difference is pretty striking between the photo of the paper schedules on a clipboard versus those screenshots and animations you have of how the software you’re building for the tablets works. It seems like a pretty transformative tool that you’re building.

MK: We hope so. One of the things we’re really excited about is the most recent feature to hit that equivalency benchmark: “ghost buses.” On the old system, there was a little white triangle, which represented a bus. It showed when a bus ought to be there according to the schedule. But sometimes, it’s not. This is a really frustrating experience as a rider, because you see the predictions on your app — it might say that a bus is coming in three minutes, coming into two minutes, coming in one minute, it’s coming in 0 minutes and then no bus arrives, and then it goes back to 20, or 10. So the bus sort of “ghosts” you. This happens because we had to not run that trip. This can happen for a variety of reasons (e.g., an emergency rail replacment shuttle needs buses) but sometimes the inspector doesn’t know the trip’s been dropped. So the new ghost bus feature acts as an alert that there’s a gap in service and we need to fill it or try to accommodate that gap by adjusting other bus trips. That is so critical to their job as inspectors, and it addresses one of our biggest complaints from riders.

There was a time when Skate had first launched that we didn’t have that yet, because from a data perspective, it was just a bit more of a lift than some of the other things we had added. Now that it’s in Skate, we have gotten a ton of great feedback. In some of my favorite interviews with inspectors I’ve done lately, inspectors have talked about how excited they were when that first launched without me even bringing it up. They were calling each other and saying, “We have ghosts! We have ghosts!”

BARI: That’s wonderful, I really like that!

Can you tell us a little bit about your process when you start a project like this? Obviously, as a user experience designer, as a researcher, you’re talking to users from the beginning. Can you tell me a little bit about what the user research part of it looks like as you as you continue to develop it?

MK: So I would say that first, we talked to as many internal people to our team, and many of the stakeholders at higher levels, so folks who aren’t necessarily in the field. People like the MBTA’s head of bus operations, and other people at that level, to get a sense of sort of what their goals are and what they see at the high level are the greatest challenge. And then it’s all about going out into the field and seeing how the inspecting and dispatching process actually works. I usually start with supervisors who work with inspectors or dispatchers every day. I ask them who they think I should talk to and why. Usually they give me their very best inspectors, people who have a ton of experience, because they want me to hear how things are supposed to be. But I also have to make sure I talk to folks who are newer and less experienced, because I want the whole picture. So I get in touch with the inspector, I set up an interview, and I go out while they’re on the job and I do a combination of shadowing and interviewing them. Those interviews take anywhere from 30 minutes to two hours, depending on how often I need to stop so that they can do their job, because the last thing I want to do is get in the way of that. But I also want to see all the different interactions they have throughout the day, and how they’re using the tools they have now. Once Skate launched, I also want to know how they’re using or not using Skate.

Usually I go out with a set of questions, but it always changes in the field. You adapt to what’s happening and to the specific inspector. Then I go back to the office and I write up the interview with as many direct quotes as possible, make recommendations, and I circulate it to the rest of the Skate team. And then those recommendations turn into tasks, that turns into software that we design and produce. And then we go back out and verify it with testing. Whether it’s a clickable prototype or a live tool we’ve actually build I head into the field and talk to more people about the thing that we just built. Normally there are many interviews in between those periods as well, because we don’t ever put something into production having only talked to one person. We try to get a good spread.

BARI: It’s interesting that you’re able to do so much fieldwork, like the ride-alongs, it seems like it’s a pretty unique way to get the full picture of what the actual experience is for the people who are going to be using the tool that you that you’re building.

MK: It is. Just even being in their environment makes a big difference. For example, earlier this week, I went out to see an inspector who I’ve never met before. And he was in a completely different kind of environment than I’ve ever seen. Usually the inspectors are at key stations which often a combination of a subway and a busway, like Ruggles or Kenmore. They’re major connection points for the entire system. There’s normally some kind of booth that they work in. Often, they’re situated such that the public can see them, and they can come up and tap on the window and ask a question. So they’ll be fielding customer questions, and at the same time they’re also checking bus operators as they come in for fitness for duty. They’re talking to operators about what bus they should take to do their trip, and making sure that everybody is where they’re supposed to be.

But when I went out this week, the inspector I met was just out there. It wasn’t a subway station, it was just a street. There was no building. He’s just out there under the awning of an office building with a cup of coffee and a bunch of paper. And that’s his office for the day. So where they work can tell us, for example, that we should probably get a strap for the tablet, because the inspector can’t hold everything at once. There is nothing shielding him from the elements or anything, and he has no surface to work on. And on the other side, when I’m in a place where there is a booth, I can see all the papers lying around, see what they’re actually referencing, and all the notes that they’re taking, and all the work arounds they have, and things they have posted on the wall because they don’t want to forget. That also tells me so much–I could just go to their locations and not even talk to them, and come away with so much information. But obviously, it’s even better when to talk to them.

I’m so grateful to the inspectors and dispatchers who have let me bug them for as long as two hours at a stretch while they’re on the job and juggling a million things. It’s only with them that Skate’s success is possible.

BARI: How many bus inspectors does the MBTA have?

MK: We have about 160.

BARI: I know you’re in the very earliest stages of this, but can you tell us about any parts that are coming up in the future that you’re particularly excited about?

MK: I’m really excited about everything that’s coming up, because so far we’ve been laying the foundation for new stuff, and now we get to start building all the new stuff we haven’t had a chance to build yet.

For example, we’re making it possible for inspectors to get a sense of the overall bus schedules inside Skate. Right now, all they can see is what is happening at the present moment. They can’t see what’s going to happen in an hour. That makes it hard to be truly proactive. Currently, they rely on this enormous book that’s called literally, “The Book”, and it is full of all the schedules for any given garage. Let’s say I’m a driver and my shift ends at noon, and I’m leaving my bus behind. And you’re getting on my bus to start your shift. These exchanges of resources have to happen in a timely fashion for service to run smoothly. It’s not easy, with The Book or with inspectors’ current technology, to link up two different drivers’ schedules and how they intersect. We believe we can offer inspectors a simple, quick way to manage these moments.

BARI: Thank you for talking to us, and for your hard work on making the MBTA work better for everyone! Is there a way people can learn more about your work, or get in touch with you?

MK: Thank you so much for talking with me. People can learn more about CTD by reading our Medium posts. Our team is growing so if people are interested in our work and want to be part of it, they can find our job postings here. To connect with me, email or find me on LinkedIn.

Published On: December 5, 2019