Invest stress-free with Vanguard's Digital Advisor, which personally customizes investment strategies for each user.
The Vanguard Digital Advisor is a new robo-service that provides automated investment assistance. I was tasked to create an accessible experience for those just starting their financial journey, targeting specifically young adults in my process.
My initial usability test was through this prototype created by Steve Schang, who is the Senior UX Researcher at Vanguard.
I created a list of usability tasks that I had my participant complete while using the prototype. My participant completed all of the tasks, but they reported confusion with some of the prototype's elements.
They recommended that the hurricane chart's scale should be narrowed down to something more realistic for the user and to add beginner-friendly elements, such as more graphics, to make the user feel less intimidated.
You can watch the full usability test below.
I consulted with my group to create a chart analyzing the key characteristics of users that would most likely use the Vanguard Digital Advisor service. We also used language learning models (LLMs) such as ChatGPT and Gemini.ai to curate ideas of typical users that would utilize this service.
Ultimately, we narrowed it down to younger users who do not have an extensive knowledge of investment and personal finances.
After consulting with my group, we developed a series of Likert Scales based on the common topics, problems, and values we felt appeared the most in our listening sessions.
Our final list was:
We discussed the pain points our participants experienced during our initial usability test and translated their emotional responses into values that would translate well onto personas. We charted these personas on our Likert Scales to better understand their values.
Then, I created a user persona based off my findings:
Ethan Smith is a junior at UNC Chapel Hill studying computer science. He comes from a very family-oriented culture. Due to language and cultural barriers, his family has been unable to pass on financial wisdom regarding generational wealth, so he is concerned about what his family will do regarding retirement. Ethan is conflicted about how to strike a balance between providing for his family and himself.
I proceeded to investigate Ethan's morals further by creating a user persona chart and an empathy map.
I used language learning models (LLMs) to generate a list of elements commonly featured on financial investment sites like Vanguard. I pared down the list by cross-referencing similar elements I saw from different LLMs and turned them into cards on kardSort.
I then asked 10 individuals to sort these cards into categories that they created themselves based on how they felt these cards should be organized. I graphed this data using Tableau and you can see my final graph here.
While some categorized elements by function, others used their familiarity with financial websites to categorize what they felt was "necessary" versus "unnecessary".
Based on the list generated from the LLMs I prompted, I created a tree test outlining the general structure of a financial website.
Financial websites have more security assurance features than other websites, such as security certification badges, user testimonies, and so forth. This is most likely due to the sensitive nature of the data they're requesting from the user– users want to know that their money is in good hands!
Ultimately, this helped inform my design choices moving forward by constantly asking, "does this process make the user feel secure?".
Below is my final tree test:
About/Informational
Necessary/Core Features
User Community
Home Page
Feature/Tools
Based on my tree test, I used Figma's built-in AI generator to create user flows based on inputted prompts. I had to reorganize the tasks a bit, but Figma's AI did a great job overall creating these user flows.
I created a wireframe/low-fidelity prototype based on the user flow for setting a goal. I chose this user flow because I felt this would be one of the most frequently used features that users would use when interacting with Vanguard's Digital Advisor service.
You can view the full wireframe flow here.
After building my wireframes, it was time to move into developing a higher fidelity prototype using Vanguard's branding. I created a UI kit using Vanguard's colors, my closest approximation of their typeface, and even their logos. This will smoothen my workflow when I build my high-fidelity prototype.
You can view the final prototype here.
After I completed my high-fidelity prototype, I submitted it to UserTesting where 3 anonymous users were assigned to complete the tasks using the prototype I created. I opted to include users from any background, not narrowing them to a specific age group despite targeting my redesign for younger users.
I discovered that older users experienced more problems with my redesign, while my youngest user (age 34) had no issues at all. My older users were ages 49 and 61 respectively and they experienced some pain points due to text size, confusion on the prototype's capabilities, and even the hurricane chart projecting investment growth.
Perhaps in my quest to make the Vanguard Digital Advisor more accessible to younger users, I ended up alienating an older user base. In my future redesigns, I would like to add options that allow users to toggle their view of the Vanguard Digital Advisor to be tailored to their exact preferences and needs.
“I was frustrated because I couldn't find 'Charts' in 'Contributions'. The font size of the text is very small. It's hard to read it.” Anonymous User, age 49.
Overall, this project was incredibly informative and I learned a lot about the design pathways UX designers undergo when tackling a project. It was incredible working with Steve Schang and the team at Vanguard for this project and I have learned a lot on how to create usability value through my designs. Fortunately, my redesign achieved a 100% task accomplishment rate across my participants.
However, I was admittedly overwhelmed with the scope of this project. I felt that the tasks demanded of us were better suited to be evenly distributed across a team or group rather than designed alone. For example, I intended to create 20 interactive panels for my final prototype, but I found that my perfect amount was 10. Considering the amount of assets and how closely I was trying to emulate Vanguard's style, creating 20 detailed panels was an insurmountable task that would be better distributed across a group. I truly learned why UX designers work in teams during projects of this scale.
