THE PROBLEM: The current Texas DPS website is visually cluttered with poor information architecture and is hard to navigate.
THE SOLUTION: Using Conversation-as-a-Service (CaaS) to determine what service the user needs and help them successfully complete the task without having to read too much text.
MY ROLE: User Research, Wire-frames, Prototype, A/B testing and usability testing.
TOOLS: Adobe XD, Miro, Optimal Workshop
TEAM: Janice Fabunan, Vibhanika Ravichandran and Anupama Udaykumar
Since the user's experience on the website as well as at the DPS centers were found wanting, we decided to split the project into two - (1) Website Redesign; and (2) Experience Design at the center (which will be covered as a separate case study).
HEURISTIC EVALUATION OF THE CURRENT WEBSITE
The original website looks dated and we found that it violated a number of heuristic principles.
RESEARCH
58 people surveyed
Survey conducted on Texas Reddit Group
SURVEY RESULTS
1 Reason for website visit
98% - Driver License/ID related services
Insight: This is the most popular service and needs to be emphasized
2 Ease of Scheduling Appointment
50% - not easy
Insight: Appointment scheduling needs to be more straightforward and user friendly
Onilne Services
In addition, Texas DPS found that the biggest reason for long wait times and over-crowded centers was that people who were eligible to complete services online still visited the centers because of a lack of awareness about what services could be done online and if they were eligible.
Insight: Users need to be made aware of what online services they are eligible for without having to read lengthy text.
EMPATHIZING WITH THE USER
With the research insights in mind, we worked on Miro to empathize with the user via an Empathy Map and we used the Nielsen Norman Group Journey Map template to develop a Journey map to trace the user's steps and discover pain points and opportunities.
HOW MIGHT WE...
From the heuristic evaluation, research and the journey mapping exercise we found that the most important issue that needed to be resolved was the information architecture of the current website.
How might we improve navigation and clarity on the website by using Conversation-as-a-Service. - we looked toward "TurboTax" to mold our interface based on their website's format to get information from the user and lead them to where they need to go.
How might we streamline information on the website by segregating content. - Since the website is for the Texas Department of Public Safety, the content of the website other than that related to Driver License/ID, needed to be retained but in a more orderly fashion. We decided to do a Card Sort on Optimal Workshop to understand how users would sort the information in the website, for the most intuitive navigation.
DESIGN
WORKFLOW
We started the design by creating a workflow to flesh out the user touch-points and interactions. We wanted to create an interaction that would minimize the need for the user to read instructions and be able to accomplish their tasks efficiently.
Sketches
We then individually sketched out options for the landing page. This sketch shows a minimalistic design using buttons with icons for visual clarity. The content is categorized based on the card sort for a more streamlined information architecture.
Wire-frames
From the sketches we developed low-fidelity wire-frames. I worked on the Conversation-as-a-service part. I used simple and straightforward questions along with buttons with icons for a smoother interaction with the website.
High-fidelity Prototype
In comparison with the landing page of the current website in which, different categories related to Public Safety are mixed up leading to visual clutter, in our proposed landing page, we have separated the categories based on the card sort, for easier navigation.
In the current website, the users need to read 3 paragraphs of instructions in order to determine if they are eligible to get the service done online. This is time-consuming, frustrating and needs the user to remember a lot information. In our proposed redesign, there is a simple questionnaire that would determine if they can get the service done online and if yes, they are directly led to where they can complete the service.
You can access the prototype here: https://xd.adobe.com/view/e9fc02f0-2477-4904-a1d3-ea72515313aa-bcaf/?fullscreen
TESTING
A/B Testing
We conducted A/B testing between the current website and the proposed redesign, to understand how users thought they compared.
8 users
Tasks Tested
1. Schedule an appointment for a New Class C Driver License
2. Change the address on your Driver License
Key Findings
1. The proposed redesign has a more contemporary UI.
2. It is easier to navigate than the current website.
Usability Testing
In addition to the tasks for the A/B testing, we included scheduling an appointment for a particular center, date and time for the usability tests.
Positive feedback:
1. Pretty UI
2. Easy to navigate.
3. Checking eligibility for online services is so much easier than having to read all the criteria.
Negative feedback:
1. In the Confirm services page, the buttons were confusing and almost all of the users tested, assumed that they were clickable.
2. In the appointment scheduler page, users found the "see more" timings option hard to locate.
WHAT'S NEXT
Right from the initial discussions, we determined that this design problem needed a service design approach as the user pain points extended beyond the website to the actual experience at the DPS centers. We tried to incorporate some features in the website that would support that endeavor. For example, leading the user to the online services, if they are eligible, would reduce the number of unnecessary visits to the centers and in turn reduce wait-times for others.
I will be covering the Service design of the Texas DPS experience in the next case study.
