A summary of the discovery interviews.
Planning & Discovery
Out of the myriad of BART issues, we started off the project not knowing the exact problem to design for. We relied on interviews with BART riders to help narrow down the most prevalent issues, with these goals in mind:
- To discover what motivates riders to take BART (pre- or during COVID)
- To discover pain-points or unmet needs with the current BART system
- Secondary research would be conducted once the problem was decided on
The participants’ BART usage ranged from riding it daily for >3 years to refusing to take it for their whole life in the Bay Area.
- To discover what motivates riders to take BART (pre- or during COVID)
- To discover pain-points or unmet needs with the current BART system
- Secondary research would be conducted once the problem was decided on
These are direct quotes regarding interviewees' fears with riding BART.
Although we found a myriad of problems to design for, these were our key findings:
- The majority of participants feared for their safety on BART, no matter what time they took it. We found this fear to be the most frustrating part of their experiences, so we prioritized the the project scope to focus on safety.
- Participants had various methods to help them feel safer. From self-defense objects, like pepper spray, or traveling in groups, most participants noted they always needed to have their guard up while commuting, no matter the time.
- Riders had a general distrust in the BART agency. Our participants cited general distrust in the agency, due to slow improvement and negligence of serious issues in the system. BART has also been found withholding crime evidence, to the extent that a third party site was created to release their crime reports for the public’s knowledge (Bart Crimes Website).
Competitive analysis. Starred post-its would influence the features of the second ideation session.
Ideate & Concept Testing
Crowdsourcing safety came from our first ideation session. It would involve users reporting incidents, rating station safety, and increasing the paper trail for BART police, similar to Citizen or Nextdoor apps. This idea would also fill the transit space that the other safety apps don’t currently cover. However, we were afraid the app would be more fear-inducing than empowering.
We gauged the crowdsourcing idea with 3 potential users. Two users did not think it would be fear-mongering, but would make them more alert and informed. However, one user said, “I feel like crowd-sourced reporting would open the possibilities of personal prejudice.”
We previously underestimated how serious prejudice would be in this setting and did secondary research. Her opinion was validated in our competitive analysis of similar apps: particularly Nextdoor reviews left on the App Store and BART news reports. We found that racial prejudice is a prominent flaw in several crowd-sourced safety apps and it made us reconsider the project direction.
When we realized racial bias was an inherent issue in crowdsourcing safety apps, we pivoted back to the idea of “safety in numbers.”
Messy whiteboarding process during the second ideation round.
Ideate, Round 2
We revisited the concept of group travel and “safety in numbers,” which means that being part of a group lessens the chances of harm happening. For the idea to be successful, we needed incentive for casual users to use the app for features other than just safety.
The second round of ideation led to a social transit app, where users release their location and passively update their statuses to other riders. There would also be achievement system that would incentivize even the most casual users to use the app. When concept testing this idea, users told us that the idea was “interesting and fun” take on safety and the achievement system “gamified the experience.” These insights would affect our branding and visual design.
User flows were made to narrow down the MVP design.
While still ideating on this new social concept, creating the user flows helped to fully flesh out features. There would be social components to talk to fellow travelers, as well as an option to ping nearby users that you don’t feel safe.
A quick competitive analysis with branding colors.
Design
Our primary research showed us that this app should be a separate identity to remove riders’ poor perception of BART. Plus, this opened the opportunity for the app to expand to other transit systems, with BART just being the first iteration. For the new branding, the name “Convoy” was chosen for its simplicity and its meaning of “a group of ships or vehicles traveling together for protection.”
Teal was chosen as the brand color because it was an intermediate of green and blue (representing safety and calmness, respectively). The color also differentiated itself from the true blue brand space that Waze and Google Maps occupy, while still being a reassuring color for users.
Waze and Google Maps also have a broad range of brand colors. We utilized the opportunity to create a soft gradient of less colors—teal, blue, and purple—to create distinct brand recognition for Convoy.
A preview of the Convoy style guide. Visual inspiration was taken from the BART payment cards and map of the railways.
Circular font was chosen for its ability to remain approachable and friendly, yet modern and stylistic. The app is meant to evoke fun, as our second round of concept testing showed, yet have the professionalism needed for an audience typically commuting to work or school. It was important for us to use avatars, so that people feel some sense of privacy in such a closed space.
A few screens depicting the splash screen, station selection interaction, train overview, messaging, and safety reporting features.
Animated fly ins were crucial to create speed bumps to help users read important information.
Usability Testing
We used a mid-fidelity prototype to catch early errors in the flow before fully creating the design in high fidelity. Tasks 1-3 were completed error-free; however, we got plenty of feedback for task 4, the reporting a safety issue feature.
During the mid-fidelity testing (left), users did not feel a sense of urgency, nor did they read all of the text besides the title. We modified this in the high fidelity prototype (right) by adding the highlight red to denote seriousness, as well as forcing the user to click an overlay when they first enter the flow.
Color was also added to denote the importance of this feature.
“Report an App Bug” feature on other apps. The button was moved up and changed visually to improve its hierarchy, given its important function. Unfortunately, we didn’t have time to gauge users’ opinion on this change.
When users were asked about their overall thoughts on the app idea, we got positive feedback. One said, “It’s an interesting idea. It’s nice to be reminded that there’s people like you that are riding BART because it’s really easy to just have your guard up.” Another participant said, “It’s a fun idea because transit is usually an unsocial activity and that element is brought back in.”
Link to Final Prototype
The process from whiteboard to user-informed high fidelity mocks :)
Concluding Thoughts and Next Steps
It’s hard to describe just how many times this project changed directions conceptually. My initial idea actually was to design a Clipper e-commerce app; however, user research revealed deeper and more serious BART problems. Solving for safety then resulted in fun colors and pup avatars that gathered a normally unsocial community. This was an unconventional approach to a serious problem, but I’m very happy with how it turned out.
During the usability tests, we learned that the participants were very intrigued by a social transit app; however, they still had so many questions to how it’d work. Many had privacy concerns, “Is it creepy that people see where you’re going and getting off?” and “Is there a way that these avatars can be verified, so that I know that they’re not stalkers?”
We had very brief consideration for privacy given the timeline, however, these were great questions to further develop the project in the next iteration. Also, these discussions always brought up an interesting question: just how much privacy are people willing to trade off for the ability to connect and seek safety with others?
