Machine Learning Tool for Research Analyzation

Project Overview

Context

The application was receiving poor user feedback, with a negative NPS score. With 30,000 users and 400-500 survey responses per month, it was clear that the user experience needed significant improvements.

The goal was to enhance usability, introduce new features, and refine the overall design to create a more intuitive and engaging experience.

‍

Objective

Enhance the user experience and user interface of the Machine Learning Tool for Research Analyzation by building upon existing base features, improving navigation, visual design, and adding new interactive capabilities to improve user engagement and satisfaction.

Problem Statement

The current user experience is inefficient, unintuitive, and lacked key interactive features, resulting in low engagement and poor satisfaction scores. A redesigned user experience and user interface is needed to improve usability, visualization, and overall effectiveness.

Design Process

Wireframing

• Developed high-fidelity wireframes to restructure key workflows.
• Iteratively refined UI elements based on user feedback.

Key Design Principles

• ‍Simplification: Reduced cognitive load by refining navigation.
• Interactivity: Introduced graph views and dashboard analysis for better engagement.
• User Empowerment: Enabled self-service data uploads and categorization controls.

Design Solutions

New Feature Enhancements

1. ‍Improved Landing Page
   • Redesigned for better clarity and accessibility.
   • Included quick access to key features.
2. Graph View for InteractiveAnalysis
   • Enabled visualization of key insights.
   • Allowed users to analyze categories, key points, and comments.
3. Self-Serve Data Uploads
   • Allowed users to upload datasets larger than 200 comments.
4. Human-in-the-Loop Categorization
   • Users could update and refine classifications to improve accuracy.
5. Enhanced Security & Compliance Features
   • Ensured data protection and user privacy.
6. User Guidance Tutorial
   • Introduced step-by-step onboarding to improve usability.
7. New AI Models for Text Analysis
   • Integrated AI-powered insights for conversational, ticketing, and textual data.

Implementation

• Collaborated with the platform squad to integrate new design elements.
• Tested wireframes and prototypes with real users for validation.
• Launched new UI/UX features in phases to ensure smooth adoption.

Results & Impact

Expected Outcomes

• ‍Increased NPS score, reflecting improved user satisfaction.‍
• Faster data analysis with interactive visualization tools.‍
• Reduced frustration, leading to higher engagement and adoption.‍
• Enhanced security and compliance, ensuring user trust.

Key Learnings

• Interactive features drive engagement: Graph views significantly improved usability.
• Self-service capabilities empower users: Reducing reliance on manual intervention enhances efficiency.
• Step-by-step tutorials are crucial: Proper guidance helps users adapt quickly to new features.