Rebuilding a Robo-Advisor Dashboard for Emerging Investors for Vanguard
Overview
Usability testing on Vanguard's Digital Advisor prototype showed that emerging investors struggled to complete core tasks in outlining parameters with only a 50% task completion rate for risk adjustment and interpreting performance.
This not only affects customer retention; it also highlights underlying issues with investing mental models and AI trust. Throughout this study, I developed a redesigned dashboard that more closely aligned with new investor expectations and needs.
Role: Lead UX Researcher and Designer
Duration: 4 months
Methods: Heuristic audit, moderated usability testing (two rounds), semi-structured interviews, card sorting, prototyping
Tools: Figma, Adobe CC, Kardsort, UserTesting.com
Process
Due to limitations in time, users, and access to Vanguard's data, testing was constrained.
I conducted two rounds of usability testing, supported by user interviews and a closed card sort to diagnose and address:
Task completion (signing up, risk, goals, tax setup)
Time on task
Hesitation (pauses, avoidance of actions, verbal feedback)
Comprehension (ability to explain outcomes)
Each method was selected to isolate a specific issue:
Usability testing → identify behavioral breakdowns
Interviews → understand mental models and confusion
Card sort → resolve navigation mismatches
Addressing Jargon
Problem: Financial data was not interpret-able → Presented a perceived expertise barrier.
Decision: Users struggled to understand investing terminology, descriptions, and visuals during tasks.
I used plain-language terms, outcome focused explanations, and progressive disclosure (tool tips, “learn more”)
Validation:
Follow-up testing revealed participants were able to accurately describe what different features did for their goals and 100% expressed confidence in their comprehension.
Risk and Investments
Problem: Risk was defined in terms of "risk attitude" and stock-bond allocation, but was very difficult for participants to understand or explain. Participants were hesitant to change this setting due to uncertainty, therefore unable to complete tasks.
Decision:
Explain how risk translates to outcomes.
Set investing strategy to "low risk" by default.
Tie risk to tangible outcomes: how it affects Robo-advisor decisions.
Validation:
Participants completed allocation adjustments with less hesitation and easily understood implications.
Navigation Redevelopment
Problem: Users did not understand navigational headers.
Decision: Reorganized information architecture (IA) around user goals, including renaming categories.
Example: "Investor Profile" became "Settings".
Validation:
100% of users located tax settings, risk, help, and other features without difficulty.
Help
Problem: Users expressed concern with the apparent lack of support.
Decisions:
Persistent Help: Button on the navigation bar.
Clear escalation path: Users may directly contact customer service via phone or email.
Validation:
100% participants easily located the appropriate help.
Prototypes
Key user flows included: Dashboard clarity, recommendations, and help.
Dashboard
Problem: Single hurricane chart, no clear timescale or allocation representation.
Decisions: The redesigned dashboard includes…
Plain-language summaries and metrics.
Investments and gains tracked per month, view can be adjusted (previously hidden on a separate page and view).
Allocation of funds is clear.
Validation: Reduced ambiguity while maintaining transparency. Users could interpret outcomes without removing the integrity of financial data.
Investment Customization
Problem: Users were only able to adjust investments according to risk levels.
Decision: Introduced recommendation feature:
Goal-aligned fund suggestions
Clear risk labeling
Links to further information
The goal was to preserve and encourage user agency and regulatory compliance while guiding exploration.
Persistent Help
Problem: Help was buried, reactive, and fragmented.
Decision: Help became infrastructure to assist users encouraging exploration and learning.
FAQ library
AI assistant entry point
Customer service channel
Contextual inline tooltips in high-friction areas
Validation
Follow-up usability testing with UserTesting.com:
Tax settings located without confusion.
Risk adjustments made in confidence.
100% of users were able to comprehend complex terms.
"If this was around when I was younger… I could definitely have used something like this. I would recommend this to anybody who's starting to save for some goals and want's to invest“ - User #2
Reflection
Key Takeaways:
Mental models as a barrier: Complex interfaces like investment portfolios require intimate knowledge of investing terms and UIs, but this doesn't work for novice investors: users succeeded, in some cases boosting task completion from 50% to 100%, when the platform was built in accordance with user interaction methods and plain language.
Business Impact: Ease-of-use improves…
Likelihood of account funding and trust (confidence-driven behavior).
Reduced drop-off in high-friction flows.
Stronger foundation for long-term retention and conversion to additional services.