Research Goals
Understand how users interpret the financial metrics and visualizations on the dashboard. Identify usability issues that hinder quick decision-making. Explore how users interact with key actions like "New Loan" and "New Customers." Evaluate the clarity and usefulness of the loan application table. Measure user satisfaction and perception of data clarity.
Research Methods
Heuristic Evaluation - Conducted a UX audit using Jakob Nielsen’s 10 usability heuristics. Focused on visual hierarchy, consistency, error prevention, and match between system and real-world expectations.
User Interviews - Interviewed 5 internal users (loan officers & customer support staff). Asked about their daily tasks, pain points with current dashboards, and data interpretation needs.
Usability Testing (Moderated) - Task-based sessions using a prototype of the dashboard. Tasks included interpreting conversion metrics, identifying top-performing loan types, and processing a new loan.
Survey (Quantitative Feedback) - Follow-up survey with 15 users. Collected feedback on satisfaction, data clarity, and dashboard elements' perceived usefulness.
Key Research Insights
Information Overload - Users felt overwhelmed by multiple data points displayed at once without enough context or summaries. They desired more storytelling with the data (e.g., highlights, changes over time).
Unclear Progress Indicators - Loan completion percentages were confusing without explanation. Users didn’t know what 60% represented or what actions were needed to progress.
Low Discoverability of Interactions - Elements like “Read more” and KPI cards didn’t appear clickable. Users skipped them because they lacked visual affordance (e.g., hover, underlines, icons).
Desire for Filtering & Sorting - Users managing large data sets need filtering and sorting features in the loan applications table. Manual scanning was slow and error-prone.
Visual Appeal vs. Accessibility Gap - While users liked the modern design, many pointed out issues with small fonts, unclear contrasts, and the lack of colorblind-friendly cues.