Who: Myself (John Scott Siri)
What: I created an tested a redesign for the NJ Transit mobile app, which is the primary application for public transit throughout New Jersey.
When: September, 2024
(3 weeks)
Where: Northern New Jersey
NJ Transit provides nearly 270 million passenger trips each year across 253 bus routes, 12 rail lines, and an assortment of light rail systems. It's mobile app boasts over a million downloads for Android and iPhone users, which allows users to access transit information and manage tickets. However, web pages relating to NJ Transit or its mobile app often complain about the design of the app, some preferring to use Google Maps or alternatives. Considering its user base, I decided to work on a redesign to show possible improvements to the transit app's user experience.
In and effort to figure out how to improve the user experience, I needed to gather information from the current users.
I conducted a series of 6 interviews for NJ Transit users who have used the mobile application in the last 6 months. These users were recruited by direct message from the NJ Transit subreddit. The average interview lasted 39 minutes.
4 of the interviewees primarily used NJ Transit to commute, while the other 2 used it primarily to get to events at urban areas.
- 4/6 said they enjoy NJ Transit services, but only 1/6 said they enjoy using the mobile application.
- 5/6 said it is hard to find features in the application.
- 5/6 said that
In order to get a better idea of where these problems were occurring, I did a set of 6 think-aloud user tests with users unfamiliar with the mobile application. The participants were given a scenario where they were asked to plan a route from Kinnelon to Willowbrook Mall and purchase the corresponding tickets.
3/6 users could not plan the correct trip using the application and the other 3 could only do so with difficulty. Those that did complete the task did so in an average of 2 minutes and 44 seconds. Major areas of issue related to finding user features or needing to switch back and forth between sections of the app. One user (NB, 21) claimed that it was confusing how the search functionality was split up. Another (M, 29) spent 2 minutes trying to find out how to link to ticket purchasing from the trip planner because he saw price totals.
Following the test, the users participated in short interviews where asked if they would use the application if others were available.
Following the initial research, some pain points were found consistently across the application.
1. Navigating the app is unintuitive.
2. App sections are segmented in a way that is not user friendly.
3. It is hard to plan a journey without already knowing the transit system.
4. There are few quality of life supports, like text-correct or adaptive searching.
NJ Transit Users need a clear and streamlined layout for planning and paying for trips so they can use public transportation without confusion.
1. Create a straightforward workflow for finding, planning, storing, and tracking trips.
2. Set up a concise way of purchasing and managing tickets.
With the problem statement and goals clear, designing can begin.
I started by identifying and defining the required features:
1. Planning a Trip
2. Purchasing Tickets
3. Saving Locations & Trips
4. Managing Tickets
The first set of wireframes was built to incorporate elements of Google Maps and other transportation apps (like Uber and 9292). A major idea was to make navigation a bit more visual and search-oriented. This makes the user need to think less about any given intermediary steps and just need to know their first and final points. Additionally, I aimed to make it easy to access information about trips within a given user flow. I also opted to add a few redundant methods to access and purchase tickets depending on how the user wants to go about it.
Following some whiteboard drafts, the designs were translated into Figma. White crafting it, some design design elements were fleshed out in the wireframes. I intentionally used colors of NJ Transit and those related to it to make sure it had a coherent style. While not fully elaborated in the wireframe, I made sure to incorporate favorite elements of the existing app, like notifications for real time updates, on the home screen.
The focus of the design was on the search flow, purchasing tickets, managing tickets, and demonstrating the benefits of some adaptive functionality. One example of this is in ticket activation- if the user just wants to active a relevant ticket, the app is designed to just use the oldest valid and relevant available ticket. I also color coded the ticket button so that users could have a visual indication on whether or not they had a valid ticket for a searched trip without needing to search for one first.
Following the completion of the wireframe, a series of 6 usability tests were conducted with users unfamiliar with the NJ Transit mobile application. Similarly the tests before, the users were given a scenario to map a trip from Kinnelon to Willowbrook and speak aloud as they completed the tests.
1. 5/6 users found the application easy to navigate.
2. 6/6 users said they would have no problem using this on a daily basis.
3. 6/6 users said they found purchasing tickets straightforward.
4. 2/6 users explicitly mentioned that they liked the trip overview functionality.
This time, all users completed the test (with an average time of 1m 48s).
Despite the general sense of satisfaction, a few inefficiencies in version of the design:
Ticket Buttons: While users realized that the ticket buttons in the trip search were related to the tickets, they did not find the purpose of them intuitive. I fixed this by adding clarifying text to the buttons.
Managing Tickets: Users found it somewhat tedious to go through and click on the clusters of text to search for tickets that met certain criteria. I alleviated this by going from list format to card format and adding a search function.
Searched Information: Users sometimes lost track of the information of the task. While the task given to them was fairly simple and handed to them explicitly, some users went a screen back when talking about the task, meaning on-screen reminders were needed. I alleviated this issue by adding search information to the top of the trip details screen.
I once more performed a set of think-aloud usability tests on a set of 6 users who were unfamiliar with the NJ Transit mobile app. Tested users had to route a trip from Kinnelon to Willowbrook Mall, pay for it, and activate.
1. 6/6 users found the application easy to navigate.
2. 6/6 users said they would have no problem using this on a daily basis.
3. 6/6 users said they found purchasing tickets straightforward.
All users completed the test (with an average time of 1m 21s).
No users expressed any significant confusion during or after the test. After the test, users suggested new features when they could not think of problems with the current design.
After wrapping up the design, it became clear that there are some fairly objective improvements NJ Transit can make to the mobile app to improve it. Several participants mentioned that features like real-time updates and active timetables with the related stops. However, there was a consensus that the app was inconvenient to go through, especially if you don't already have the route you want to go memorized.
When comparing the usability tests from the current app design and those from the final design, the navigation issues cleared up and users found the updates fairly intuitive. The latest set of participants liked that they could reference relevant tickets from within a selected trip. They also liked could go straight from a search to purchasing the related tickets.
The differences present between the current design and the redesign showcase one possible method for improvement. However, the need for a streamlined search method and ticket management is nevertheless apparent. The data suggests that the redesign helped provide clarity when dealing with the task.
One major restriction this case study has is the limited amount of quantitative data available. I considered analyzing Reddit data, but found that the volume of posts and comments in recent months might be impacted by updates to the application and instead opted to lean into qualitative data. I did not have access to the large body of relevant user data that those at NJ Transit might have, which might provide evidence counter to my findings. Additionally, my app being a prototype means that there might be an under-representation of navigation issues due to the lesser ability to get lost. However, the ease that participants had during the think-aloud testing makes me think this issue is non-substantial.
Additional improvements might be beneficial to the updated design. One such function not mapped out was a map view, where progress on a trip could be demonstrated using maps that fill. It would also be nice to directly link with external map software, like Google Maps, which could help fully route a trip (even beyond the public transit journey).
