Smartwork Redesign

TIMELINE

May 2025 - Present

PLATFORM

E-Learning Mobile Platform

MY ROLE

UX/UI Designer

1M+

1M+

students gain access to mobile and tablet devices

100%

100%

WCAG 2.1 AAA accessibility compliance achieved

#1

#1

W.W Norton's most
popular product

Overview

Over one million students use Smartwork — an online learning platform for students and instructors across higher education. For the past 17 years, Smartwork has been desktop-only, limiting when and how students could complete assignments.

As student expectations shifted to mobile-first experiences and competitors introduced native apps, Smartwork began falling behind. At the same time, institutions required accessible learning tools, putting additional pressure on the platform to meet modern standards.

This project focused on transforming Smartwork into a fully responsive and accessible experience while preserving the familiarity required by a large, existing user base.

The Problem

W.W. Norton's legacy product was not accessible or mobile-friendly, blocking a significant portion of students and failing to meet institutional requirements. The platform needed a full responsive and accessible overhaul — without disrupting the familiar workflows of over one million existing users, and within the tight limitations of a legacy system.

Users

Smartwork operates as a multi-sided system, with distinct user needs.

Instructors

Course educators

  • Customize assignments by curriculum

  • Monitor class performance & analytics

  • Maintain control over content structure

Students

Primary learners

  • Complete assignments efficiently

  • Resume progress seamlessly

  • Interact with complex question types

Content Writers

W.W. Worton Internal

  • Align content with textbook material

  • Maintain consistency across disciplines

  • Support scalable content creation

Research & Audit

Given the scale of the system and its limitations, research was focused on high-impact areas rather than broad exploration.

To understand the scope of the problem, I conducted a full audit of 100+ student-facing screens, covering: assignment homepage, question player, all 13+ question types, edge cases across interactions.

Then to understand how to approach responsive behavior, I also conducted a targeted competitive analysis of table interactions and dense data layouts. This was important for the Assignment Homepage, where traditional table structures failed on smaller screens.

It became clear that maintaining strict structural fidelity to desktop patterns would not work on mobile and that the experience needed to adapt to prioritize readability and interaction.

image here

From this audit, several patterns emerged:

  • Component inconsistency
    Buttons, inputs, and layouts varied significantly across screens


  • Responsiveness breakdown
    Tables, interactions, and layouts failed on smaller screens


  • Accessibility failures
    Missing labels and ARIA attributes, failed color contrast, and non-navigable for screen readers


  • Cognitive overload
    Excessive icons, dense layouts, and unclear hierarchy

Design Problems

1

Broken UI

Critical information like the score and date are buried in low-contrast colors that blend into the background, making it difficult to scan at a glance. The icons alongside them add no value and only contribute to the visual noise.

2

Color choice failures

Using red to highlight a neutral deadline sends the wrong signal. Red universally communicates error, danger, or urgency. This leads the user to feel they did something wrong or missed the deadline.

3

Accessibility gaps

A play button icon was used to link to individual questions, not start the assignment, which is immediately misleading. With no label or tooltip, users had no way to understand its purpose, and those relying on a keyboard or screen reader couldn't access it at all — failing basic WCAG accessibility standards.

4

No Mobile support

The application wasn't supported on mobile, disabling the entire assignment and blocking users from continuing. This excludes a large portion of users from completing their work.

5

High cognitive load

The original design was cluttered with icons. A lot of these icons added clutter without adding any clarity. On an already busy screen, decorative or misleading icons only overwhelm the user and get in the way of the actual content.

6

Broken components

Most components in the original design were not built for smaller screens, either breaking entirely, or becoming too small to use on tablet or mobile. This created many issues for users and blocked them from completing their assignment.

System Constraints

Design constraints of a legacy product

1

Updating the header required a full custom rebuild - which was out of scope. 

2

Assignment layout and behavior had to stay the same - including all visual cues and information.

3

All existing information in the question table had to stay and be visible to students.

4

Question layout and behavior had to stay exactly the same for the 1M+ existing users.

5

Question language for buttons, links, and descriptive text had to stay familiar.

6

Bottom navigation had include all existing buttons, links, and text.

Designing for Smartwork required navigating tightly coupled technical, organizational, and behavioral constraints.

The platform was built on a legacy system, making even small interface changes difficult to implement without significant engineering effort. As a result, developers were cautious about modifying existing components, and stakeholders were resistant to large-scale visual changes that could introduce risk. At the same time, over one million students were already familiar with the platform. Any major changes could disrupt established workflows and increase friction during assignment completion.

This created a constant tension throughout the project: improving usability, accessibility, and responsiveness while preserving familiarity and system stability. Rather than approaching this as a full redesign, the work focused on making targeted, high-impact improvements that could evolve the product without breaking it.

Challenges & Design Strategy

Strategy Execution

Given the scale of the work, I structured the project into six sequential phases, beginning with the Assignment Homepage and Question Player, which are foundational to all student interactions. This approach allowed updates to be rolled out incrementally without disrupting active users. It also enabled design decisions made in earlier phases to be reused and scaled across more complex question types later in the project.

Design Phase

Wireframes & Iteration

Early exploration focused on simplifying dense layouts, reorganizing content hierarchy, and adapting key flows for smaller screens. Wireframes played a critical role in validating structural decisions before applying visual design, allowing me to test different approaches to layout, responsiveness, and interaction patterns while working within system constraints.

image here

Designing for Responsiveness

One of the most critical transformations was enabling Smartwork to function across devices.

Before
  • Desktop-only layouts

  • Tables and dense interfaces

  • Broken mobile interactions

After
  • Card-based layouts for mobile

  • Progressive disclosure of information

  • Touch-friendly interactions

old design

new design

Complex structures like tables were reimagined as sortable, scannable card systems, preserving functionality while improving usability.

Accessibility Transformation

All Smartwork questions needed to be AAA WCAG compliant. They failed our … They were not usable with screen readers and not functional on mobile.

Solution
  • Increased touch targets across all interactions

  • Full keyboard navigation support

  • ARIA labels for all interactive elements

  • Improved color contrast ratios

  • Clear focus states and navigation cues

  • Accessible modals, tooltips, and inputs

Rethinking Question Interactions

Drag-and-drop interactions posed a major challenge. They were not usable with screen readers and not functional on mobile.

Solution
  • Replaced drag-and-drop with structured selection patterns (drop downs):

  • Enabled keyboard and screen reader compatibility

  • Supported all edge cases

  • Maintained learning integrity

Drag-and-drop interactions posed a major challenge. They were not usable with screen readers and not functional on mobile.

Labeling Question Type

old design

new design

Ranking Question Type

old design

new design

Design System Improvements

Rather than visual updates alone, this work strengthened the system:

  • Standardized components across 100+ screens

  • Aligned UI with Norton Design System

  • Reduced visual noise and redundancy

  • Improved scalability for future features

This created a more maintainable and consistent product foundation

image here before and after of components

Cross-Functional Collaboration

Given the scale of the project and the constraints of working within a legacy system, close collaboration across accessibility, product, and engineering was critical throughout the entire process.

Collaboration with Product & Engineering

Due to technical and organizational constraints, collaboration with product managers and engineers was continuous and iterative rather than a one-time handoff.

Throughout the project, I worked closely with engineers to understand system limitations and feasibility, while aligning with product managers on priorities, timelines, and scope. Many design decisions required tradeoff discussions, balancing usability improvements with implementation complexity.

Each phase included multiple rounds of review, where I presented interactive prototypes, walked through design decisions and edge cases, and aligned on implementation details before handoff.

Accessibility Integration

Accessibility was not treated as a final validation step, but as a continuous part of the design workflow. After each design phase, I conducted dedicated review sessions with our accessibility partner to evaluate all updated screens and question types.

During these sessions, we validated screen reader compatibility, keyboard navigation, labeling, focus states, and contrast compliance to ensure alignment with WCAG 2.1 AAA standards. Designs only moved forward once they passed accessibility validation, allowing issues to be identified and resolved incrementally rather than at the end of the project.

This approach was especially important given the complexity of the question types, where interaction patterns varied significantly and required careful consideration to ensure all users could complete tasks independently.

Handoff & Implementation

For each phase, I provided detailed annotated designs in Figma along with summary documentation outlining key updates. I also walked through the designs with developers to ensure clarity around interactions, behaviors, and edge cases.

During development, I remained closely involved by answering questions, reviewing Jira tickets, and conducting QA on implemented features to ensure design fidelity and accessibility compliance.

Final Design

The final experience preserves familiarity for existing users while introducing a more modern and consistent interface. It works seamlessly across desktop, tablet, and mobile, ensuring that interactions remain intuitive regardless of device. Accessibility considerations were fully integrated, enabling all users to navigate and complete assignments effectively.

image here (link to the full prototype)

Outcomes & Impact

The redesigned Assignment Homepage has been launched to over one million users, with additional sections rolling out progressively. This work established a strong foundation for mobile-first learning and contributed to Smartwork’s expansion into additional disciplines. It also significantly improved accessibility across all question types, aligning the platform with modern standards.

LinkedIn

Discord

Twitter

LinkedIn

Discord

Twitter

LinkedIn

Discord

Twitter