This case study was conducted as part of a group project during the winter term for CSC318: The Design of Interactive Computational Media at the University of Toronto.
Tools
Figma, Canva
Skills
Wireframing, Prototyping, User Research
Role
Designer & Strategist
Duration
January – April 2025
The Problem
The online job search and application process can be overwhelming, especially for students managing academic deadlines and stress. Many undergraduate students struggle with navigating job platforms, leading to frustration and, in some cases, avoidance of job searching altogether. Current job search platforms often rely too heavily on keyword-based matching, lack intuitive filtering options, and provide limited transparency in job recommendations. Additionally, tedious, manual application processes further discourage engagement.
Our goal is to streamline the job search experience by making it more efficient, user-friendly, and personalized. We aim to reduce friction in job discovery by addressing issues such as poor job-matching accuracy, lack of filtering options, and high application complexity. By improving search relevance, enhancing filtering capabilities, and introducing AI-driven personalization with clearer job-matching explanations, we hope to create a more accessible and stress-free job search process for students and busy professionals alike.
How can we make the online job search process less overwhelming and more intuitive for students and busy professionals?
Stakeholders
Primary Stakeholders
Job seekers, including university students and busy professionals, who find the job search process overwhelming due to competing priorities. They struggle with identifying relevant roles, tedious applications, and a lack of streamlined tools.
Secondary Stakeholders
Employers who struggle with filtering through large volumes of irrelevant applications. Improved job matching can help them identify qualified candidates more efficiently, benefiting both employers and job seekers.
User Persona
Research Process
Formative Studies
Interviews
Primary Stakeholders
- Networking effectiveness varies significantly.
- Customizing applications is tedious.
- Job seekers use multiple platforms to maximize opportunities.
Secondary Stakeholders
- AI-generated applications create low-effort submissions.
- Referrals are preferred for hiring.
- Job postings across multiple platforms serve different purposes.
Questionnaire (Survey Form)
Primary Stakeholders
- Participants commonly use a range of platforms to maximize their exposure to potential job opportunities
- Tailoring cover letters specific to each company as well as finding jobs that are a good fit are some of the most challenging aspects of the job search process.
- Research outside of the job posting
Understanding the User
Job Stories
| Situation | Motivation | Expected Outcome |
| When preparing cover letters and resumes, | I want something to help me identify keywords/buzzwords for the job I am applying for | so that the chance of my application being automatically filtered out is minimized |
| When I have found an online job posting I am interested in, | I want something to help me prepare a cover letter and fill out the online application | so that I can avoid application fatigue |
| When I have found an online job posting I am interested in, | I want to learn more about the employer and company beyond the information in the job posting description | so that I can make a better informed decision if the job is right for me, as well as prepare a better interview/application |
| When searching for job postings, | I want to be able to easily cast as wide a search net as possible through multiple platforms | to maximize my ability to find jobs that I want, as well as increase my callback rate |
| When searching for job postings, | I want to conveniently find the postings for jobs that match my skills, interests, and stipulations | so that I can focus my energy on opportunities that I am both excited about and more likely to succeed in |
Understanding the User
Experience Map
Defining the Solution
Design Requirements
Based on insights from our field studies, we have identified five key functional requirements that our system must achieve to enhance the job search experience. These requirements address job seekers’ primary frustrations, as identified in our interviews and questionnaire responses, and aim to improve efficiency, reduce cognitive workload, and provide relevant information for informed decision-making.
- Help identify keywords/buzzwords that are relevant to the jobs being applied for
- Aggregate job posting information from multiple platforms
- Help users find job postings that match their specific background and goals
- Provide additional information beyond the job posting description
- Track applications, deadlines, and statuses
These design requirements directly address job seekers’ frustrations and inefficiencies in the job application process. By reducing cognitive workload, aggregating job listings, enhancing search personalization, providing additional employer insights, and streamlining application tracking, our system aims to create a more seamless, efficient, and informed job search experience.
Low-Fidelity Prototype
Our paper prototype simulates the AI-assisted job search/application website named “Job Cruise,” meant for PC users.
Storyboards
High-level Storyboard #1: Resume & Cover Letter Customization
This storyboard illustrates how the system streamlines the process of tailoring resumes and cover letters for job applications. It reduces the cognitive effort typically required by automating aspects of writing and customization. At the same time, it offers users the flexibility to choose how much control they want to retain, allowing them to balance efficiency with personalization based on their individual preferences.
High-level Storyboard #2: Enhancing Job Search with Intelligent Matching and Insightful Support
This storyboard illustrates how the system streamlines the job search and application process by matching users with job postings that align with their background and goals. It also allows users to access deeper insights by asking detailed questions to an AI model about each posting. By presenting relevant information efficiently, the system supports more informed decision-making while saving time and reducing cognitive effort.
Screen-level Storyboard #1: Job Browsing with Smart Preferences & Unified Search
This storyboard outlines the process of setting up Smart Preferences and Personal Information to browse job postings tailored to the user’s background. The system allows users to filter and personalize job listings based on their skills, experience, and preferences, while also consolidating postings from multiple platforms into a single interface. This streamlines the job search process and helps users focus on the most relevant opportunities.
Screen-level Storyboard #2: Tracking Job Applications with a Centralized Progress Dashboard
This storyboard showcases how the system helps users monitor their job application status by categorizing them into backlog, in progress, review required, and accepted. The comprehensive application tracker allows users to stay organized and keep track of their submissions, upcoming deadlines, and interview progress—all in one centralized location.
Screen-level Storyboard #3: Informed Job Search with Integrated Research Assistants
This storyboard demonstrates how the system supports users in exploring job opportunities by providing access to additional information such as salary estimates, company culture, and employee experiences. The Research Assistant feature streamlines information gathering from multiple sources and presents it in a single, unified interface. This helps users make well-informed decisions while saving time and reducing cognitive effort during the job search.
Observational Research Process
Think-Aloud Study
To conduct our Think-Aloud study, we gathered four participants from our CSC318 course, who fall under our target user base of ‘job seekers’. We had each participant complete the 5 following tasks with our paper prototype, while narrating their thoughts out loud:
Task #1: AI generate a resume
- AI-generate a resume from scratch and preview it.
- Go into the “Personal Info” page and then upload a professional certificate called certified_consulting_accounting.pdf into the system.
- AI-generate a resume from scratch again, preview it, and observe any changes and updates in the newly AI-generated resume compared to the previously AI-generated one.
Task #2: Update phone number in “Personal Info”
- Go into the “Personal Info” page and access the system’s raw memory.
- Update your phone number into the system.
- Access the system’s raw memory again and note any changes to it.
Task #3: Browse job recommendations
- Go to the job recommendations page and browse the jobs that are displayed.
- Click on a couple to open up the expanded view for the job
- Pick one job and find out more about why it was recommended to you and find out more about the company
Task #4: Apply to a job
- Apply for a job of your choosing manually
- Go to settings and turn on auto apply, but do not require user review and apply to another job (Not
Task #5: Review Job Application Tracker and check for notifications
- Go to the Job Application Tracker page from main menu
- Observe the Kanban board with four sections: Backlog, Doing, Review, and Accepted/Rejected.
- If Auto-Apply is enabled, jobs move through the stages automatically.
- If Auto-Apply is disabled, user manually move a job through its stages
- Return to the Menu Page and check for a notification about recent job applications.
Key Findings:
- Hyperlinks/shortcuts make navigating the site a lot easier
- The nature of what “Auto-Apply” did was unclear to participants
- Uneven interpretability of the job tracking system
Observational Research Process
Heuristic Evaluation
After each of the four participants had conducted a Think-aloud Evaluation, we asked them to complete a Heuristic Evaluation based on their experiences during the Think-aloud Evaluation. We asked them to rate the severity of heuristics, which can be assessed according to frequency, impact, persistence, and market impact, based on the severity rating scale below.
Heuristics:
- Visibility of system status
- Match between system and real world
- Aesthetic and minimalist design
- User control and freedom
- Consistency and standards
- Error prevention
- Recognition rather than recall
- Flexibility and efficiency of use
- Recognition, diagnosis, and recovery from error
- Help and documentation
Severity Rating Scale:
0 or N/A = Bright spot, not a usability problem
1 = Cosmetic problem (fix if extra time)
2 = Minor usability problem (low priority to fix)
3 = Major usability problem (high priority to fix)
4 = Usability catastrophe (must be fixed before release)
Key Findings:
- Current UI design feels cluttered, inefficient, visually unappealing, and difficult to navigate
- Lack of sufficient documentation and guidance – users struggle to understand and efficiently utilize certain features
- Usability and discoverability issues hinder key system functions, emphasizing the need for intuitive design, clear guidance, and user-centered features that support effective navigation and decision-making
High-Fidelity Prototype
Coming soon in April 2025!
Usability Study
Coming soon in April 2025!
Final Prototype
Coming soon in April 2025!
Reflection
Coming soon in April 2025!