Redesigning core flows to improve discoverability and usability

DESIGN FOR WEB SAAS
Highlights – Comments
Highlights – AI Generated
B – 678
B – 677
B – 679
B – 679a

Overview

CoGrader, an AI-powered grading platform for teachers, faced low feature discoverability, inconsistent interactions, and limited user understanding of its most valuable tools. In late 2024, I was brought in to redesign the platform’s core flows and make key features like AI insights, grammar corrections, plagiarism detection, analytics, and AI-generated rubrics easier to find and use.

MY ROLE

UX Designer. Responsible for colaborating with CoGrader's tech manager and stakeholders, creating high-fidelity screens, user flows, visual design, and design documentation.

CLIENT

CoGrader

PLATFORM

Web

CATEGORY

EdTech

YEAR

2024-2025

Impact Highlights

Streamlined feedback experience

Redesigned the teacher feedback screen to make AI-generated insights more intuitive and actionable.

Improved feature discovery

Increased visibility of analytics and rubric creation, helping teachers leverage the platform’s full potential.

Consistency across flows

Reduced interaction inconsistencies, leading to a smoother, more predictable user experience.

Key Contributions

 

Received tasks from tech manager based on qualitative and quantitative research, including teacher interviews, analysis of support channel feedback, and behavioral tracking via PostHog.

 

Mapped current vs. ideal user flows, identifying friction points in discovery and comprehension.

 

Collaborated with the tech manager to align changes with business priorities and technical feasibility.

 

Delivered redesigned interaction flows with clear entry points, consistent navigation patterns, and improved visual hierarchy.

OUTCOME

Creating new flows resulted in...

Teachers now finding it easier to discover, understand, and act on key features, leading to higher engagement with AI tools and a more effective grading workflow. The redesign strengthened both usability and business alignment, setting the stage for future feature adoption.