Overview
The University of Maryland Baltimore County has its own transit service which is
known as UMBC Transit also affiliated by Maryland Transit Administration(MTA).
We chose it as our annual project mainly because we were the user and felt it
needs improvement in the overall product User Experience. We decided to perform
the following UX practices and divided it into four phases:
Brainstorming
Based on our own experience as a student using transit web applications and talking
with fellow students, we noted down major drawbacks that need to be addressed for
the research and more user inputs.
Competitive Analysis
The first study we did was based on the list of transportation services provided
by university-owned transit in the united states. We chose four websites and analyze
them for the overall user experience and to find common pain points by highlighting
the most relevant features similar to our website.
Qualitative Interviews
In order to collect real user data, we followed a semi-structured interview process
with two students where one was an active user and another one was a novice user.
We also interview the transit manager as our main stakeholder of the website. The
interview environment was a natural field setting of the user.
Personas
Then we came up with these 2 personas to serve as a cumulative representation of all the
common user needs, goals, tasks, pain points that we derived from the Stakeholder interviews,
user interviews, and contextual inquiry.
Task Analysis
Based on insights we found that checking the bus schedule is the most frequent
task performed by all types of users so we did a task analysis of how to check
a specific bus schedule. For different scenarios, we also defined the path to
achieve goals as plans.
Heuristic Evaluation
Now to focus on most specific problems with high severity we quickly followed Heuristics
principles of usability to find out good & bad aspects of the website that needed improvement.
Participatory Design
To get more closer to the user’s ideas, we followed a participatory design
where we invited two users as our design collaborators to help use guiding
what can we do change in the existing interface to make it better and usable.
From a previous study, we found two most important problems that require less
cost and time in order to make the user experience better and decided to work on that.
Problem 1: HomePage navigation visibility(HE-06)
The task was to design home page navigation in such a way that it is clearly noticeable and easily accessible.
Problem 2: Excessive details above route table(HE-04)
The second activity was to redesign the data above the route table that helps users to choose the right schedule for the bus without spending much time.
High Fidelity Prototypes
After previous exercise results, we finalize to convert user sketch designs into high
fidelity prototypes for usability research. For both problems, we designed each variation
in Adobe Photoshop.
A/B Testing
To narrow down the best solution we needed criteria with actual results. A/B testing
with two users with a different set of prototypes linked as a clickable system was
used and think aloud technique was performed to measure loopholes. Below are the
results of both users and their inputs:
From the above quantitative results, we chose to move further prototype-3 for the
first problem and prototype-2 for the second problem as our main priority.
Usability Testing
We used Tobii eye-tracking software for analyzing how impactful the changes are in the prototypes.
The study was conducted in the user study lab with time-based tasks given to two different users.
Below are the eye-tracking results for both prototypes which clearly show the improvement of the
user’s task procedure technique.
Project Learnings
- Keep users first
Although I and other designer both were active users of the transit website,
we didn’t let ourselves decide the end results, we followed the qualitative
and quantitative data gathered from the actual users.
- Transition matters in interaction
It is important to explain to users that we are evaluating the system not them.
Most of the time due to a lack of understandings people don’t open up what they
are actually feeling.
- Cost Factor
Summarizing the list of problems needs proper severity levels to proceed further.
The most frequently encountered problem by users with the least improvement cost
is prioritized always.