Outcome
+90% Satisfaction Score
0% Error Rate
My role: UX / Product Designer & Researcher
Team: Data analysts, marketing managers, executives (CEO), system administrators
Product: PolyAnalyst 6.5 by Megaputer Intelligence Inc.
Year & Duration: 2025, 3 weeks
Overview
PolyAnalyst is a powerful analytics platform supporting workflows from data ingestion to advanced visualization and reporting. However, its first-touch experience, the initial screen users encounter after login, created significant friction, particularly for new users and intermittent users returning after periods of inactivity.
This case study examines how information overload, weak visual hierarchy, and unclear onboarding cues increased cognitive load and delayed task initiation, and how a role-aware, scaffolded information architecture improved usability, learnability, and adoption.
Problem Scope
Through user interviews and goal mapping across multiple user roles, I identified systemic UX issues in the existing home experience.
Core problems identified:
Weak Onboarding Cues
The interface did not guide users toward a clear “next step,” increasing time-to-action and uncertainty—especially for first-time users.
Cognitive impact: High extraneous cognitive load at entry.
Unprioritized Visual Hierarchy
All application tiles appeared equal in visual weight, despite significant differences in usage frequency and importance. Users had to consciously evaluate options instead of relying on perceptual cues.
Cognitive impact: Increased decision friction and slower orientation.
Information Overload
Multiple sections (Applications, Recent Projects, Recent Reports) competed simultaneously for attention, making it difficult to identify where to focus first.
Cognitive impact: Divided attention and delayed task execution.
Poor Whitespace Management
Dense layouts and insufficient spacing made the interface feel cluttered and harder to scan.
Perceptual impact: Reduced readability and visual comfort.
Research Framing
This problem intersects with several core HCI concepts:
-
Recognition vs. Recall (Nielsen)
-
Cognitive Load Theory
-
Progressive Disclosure
-
Novice–Expert Transitions
-
Information Scent and Visual Hierarchy
The guiding question was:
How might we design an entry experience that supports both novice learning and expert efficiency without fragmenting the interface?
Research Methods
To ground design decisions in evidence, I used the following methods:
-
Semi-structured interviews across roles (analysts, managers, executives)
-
Goal mapping to understand user intent at login
-
Heuristic evaluation using established usability principles
-
Task walkthroughs simulating first-time and returning user scenarios
-
Iterative usability testing of redesigned flows
Synthesis: User Intent at First Touch
Across roles, two dominant intents emerged:
1. Learning & Orientation
(“What can I do here?” “What’s new?” “How do I get started?”)
2. Task Resumption & Productivity
(“Continue my last project.” “Open a report quickly.”)
The existing design attempted to serve both simultaneously, resulting in overload.
Design Strategy
Separate learning from doing, without separating the system.
Rather than creating different products or complex personalization, I introduced intent-based spatial separation within the same interface.
Heuristic Mapping Table
This heuristic mapping demonstrates that usability improvements emerged not from isolated UI changes, but from aligning system structure with human cognitive constraints. By explicitly designing for recognition, progressive disclosure, and expertise-sensitive interaction, the redesigned entry experience supports both learning and efficiency without increasing complexity.
Design Solutions
1. Role-Aware Information Architecture Redesign
I restructured the home experience into two clearly defined spaces:
-
Getting Started Tab
An onboarding-focused area for:
-
Tutorials
-
Learning resources
-
“What’s New” updates
-
Conceptual orientation
This space supports knowledge acquisition and system understanding, especially for new or returning users.
-
Home Tab
A productivity-focused area for:
-
Rapid access to core tools
-
Recently opened projects and reports
-
Frequently used applications
This space minimizes friction for experienced users seeking efficiency.
HCI rationale: supports progressive disclosure and reduces unnecessary cognitive load for experts.
2. Streamlined Navigation for Project Resumption
I redesigned navigation between Recent and Favorite projects/reports to use a clear toggle instead of dropdown menus.
Impact:
-
Faster visual parsing
-
Reduced interaction steps
-
Alignment with common UI patterns
3. Recognition Over Recall (Usability Heuristic)
I introduced consistent iconography paired with labels for primary tools. These icons are reused across the platform to reinforce memory through repetition.
Result:
-
Improved scannability
-
Faster tool recognition
-
Lower learning curve
Landing Page_After
Evaluation & Impact
User Impact
-
Reduced cognitive load at entry
-
Faster decision-making
-
Improved efficiency for both novice and expert users
Product Impact
-
Higher engagement with onboarding content
-
Improved feature discoverability
-
Increased adoption of underused tools
System Impact
-
Expanded design system with reusable components
-
Established a scalable content structure to support future growth
90% Satisfaction Score
0% Observed Error Rate
in tested onboarding flows
4. Visual Hierarchy & Layout Refinement
I applied:
-
Increased whitespace
-
Clear typographic hierarchy
-
Visual emphasis on primary actions
-
De-emphasis of secondary tools
This reduced perceptual clutter and improved scanning efficiency.
Landing Page_Before
Landing Page (Home Tab)_After
Landing Page (Getting Started Tab)
Key Takeaways
Designing for Scalability
This project reinforced that scalability is not just visual—it is systemic. Every new component must integrate into a coherent structure that supports future expansion without increasing cognitive burden.
Designing for Transitions
Supporting users across novice → intermediate → expert stages requires intentional scaffolding, not one-size-fits-all interfaces.
Research Implications & Future Works
This work suggests broader opportunities:
-
Adaptive onboarding based on usage patterns
-
Context-aware guidance that fades with expertise
-
Longitudinal studies of learning curves in analytics platforms
Reflection
This project deepened my understanding of designing for cognition at scale, where success is not measured by visual polish alone, but by how effectively an interface supports human thinking, learning, and task resumption over time.
