Problem

International students find it hard to find work authorization information on online job searching platforms.

Work authorization is essential for international students to work in the U.S. However, the current online job hunting platforms mostly lack of this information on their job postings page. This not only causes international students' stuggling of finding the information, but also a waste of time for both employers and employees to process without reaching an agreement on whether to provide work authorization.

Market Research

Existing platforms each has its merits and limitations.

Users mainly use three kinds of platforms for job hunting: job posting platfroms, companies' websites, university career website. We choosed LinkedIn Jobs, Google Career and Cornell Handshake to investigate.

• LinkedIn Jobs: Valuable network, large amount of job postings. Low HR responding rate. No work authorization information.
• Google Career: Reliable job postings. Low HR responding rate. No work authorization information.
• Cornell Handshake: High HR responding rate. Very limited job postings. Some provide work authorization information.

Hypothesis

Improving the transparency of work authorization information will make the job hunting experience happier for students and also help companies to lower the costs of hiring.

User Research

Understand what are the pain points through the user journey and what are some core problems to solve.

Interviewing 5 participants who are all STEM major international students at Cornell and are all actively looking for jobs, we introduced our persona Lilly Li and instilled our opportunities:

Initial Approach

Provide a job posting platform that has work authorization information.

After discussing around 100 ideas that the group came up with, we agreed on creating a job posting platform specially for international students. Our initial design has 4 basic ideas:

• Visualized Educational Content of Work Authorization information
• Job posting list that are searchable and filterable by sponsorship and E-Verify information.
• Let users to "crowdsource" their experience of E-Verify and sponsorship on job postings.
• Provide a browser extension to allow users check E-verify information on job listing page on other platforms.

Test the Initial Approach

Users liked our web browser extension but were skeptical about the reliability of our sources of infomarmation and also how we could compete with other existing successful platforms.

1. "How can you ensure the job postings on your platform are reliable and up-to-date?"
   We realized it's a lot of work to operate a platform that covers all the postings.
2. "Are you going to contact HR to verify the sponsorship information? Can I trust your source?"
   Unfortunately, sponsorship information can change base on different candidates and different positions inside a company, and thus it is hard to verify without the company's announcement.
3. "How can you stand out from so many existing platforms?"
   With the sponsorship feature being unsure and the E-Verify being confusing, our platforms seems lost it's attractiveness to international students.
4. "I really like the browser extension, can you develop it right away?"
   I am glad we finally got some compliments.:)

New Approach

Realizing our initial design didn't work well, we revisited our interview notes to find new opportunities.

We redefined our goals: to facilitate users' applications on existing platforms and help them to manage the job hunting process. The browser extension is essential to achieve the new goal.

The new platform works like a job "Pinterest". As users viewing jobs online, they are able to check the work authorization information at any time. By "Pin" the job, they could manage interested job postings at our platform.

Test the New Approach

Is the new idea in a right direction? Is the information architecture and interface easily understandable?

Considering the time limitation, we decided to create a higher fidelity prototype using Balsamiq to test if our new approach is in the right direction and the basic information archintecture together.

We tested the prototyple with 5 users. They were all excited about being able to check work authorization information through the browser extension anytime. They also understood the information architecture of "Browser Extension - Personal Board - Public Board". Their concerns about the platform were:

1. "People may be unwilling to share job opportunities to other candidates. In this case, the public board doesn't make sense."
   The ideas of pinning to the public board and crowdsourcing the E-Verify and sponsorship information are based on building a community where people help each other. How do we design to encourage people to contribute to the community?
2. "How can I trust the vote for E-Verify? They can be out-dated and wrong."
   We decided to have users write comments and provide details about their information sources instead of having them to vote.

The Balsamiq prototype was not enough for testing the detailed interactions. We created a high fidelity prototype to test the interactions.

High-fidelity Prototype

Exploration of the main features go in this phase.

Usability Testing

Voice from user helps us jump out of our own brain.

Through the Heuristic Evaluation, the system with basic UX problems excluded has worked on our own brain, but we need the voice from the user. We recruited 5 users to test the usability of our product. Users were to conduct 4 major tasks:

1. Sign in to the system and go through the welcoming education information.
2. Check the E-Verify status of a company in the extension, contribute their own comment on the  E-Verify status on the company, and see other people’s comments.
   
3. Pin a job that sponsors employee in their own job board, and also make it available in the public board.
4. Organize the jobs in “My Board” by creating a new folder and put the first job posting into the folder, and then delete this job posting from their board.

Following are the top 3 UX problems we identified from the usability testing:

1. My Board: Need an improved user interaction for organizing job postings.
   All participants getting stuck on organizing job postings. Some could not find the Select button, and some did not know how to select each item on the list after clicking Select. 3 of them could not find the Delete button. They expected a single action bar containing all actionable items for each job posting and suggested implementing a select all feature.
2. Browser Extension: Need confirmation after pinning.
   Most participants got confused after the pinning a job posting because nothing happens after the action. Since they were unsure whether the job had been pinned successfully, they didn't know what to do next.
3. Educational information: Need to be clearer and more cleanly presented.
   Participants thought the work authorization tutorial during on boarding being too text-heavy. They had trouble understanding some details of the content provided and did not like the idea of having pages with different lengths, so they have to scroll down to continue. 
   

We made three major changes based on users' feedback:

Final Design

Go Through Educational Information

Pin a Job to User's Personal Board

Manage Job Postings on Personal Board

Bottom-to-top Public Job Board & E-Verified Company List

Future Work

We haven't spare enough thoughts on how to help people to manage job application process more effectively.

All my teammates were happy about the result we got and aspire to work together to implement it. But I want to ask ourselves two more questions before we move on:

1. What are some more effective ways to help people manage the job application process, other than just creating different folders?
   According to our user research, users are using excel and bookmark to manage their job application. Does our tool has all the feature they use in excel and browser bookmark?
2. How might we encourage the sharing of the more sensitive sponsorship information?
   Our design has solved the E-Verify problem, but the sponsorship information is more sensitive and can be different from individuals. The best way to solve it is to have a more conversational and interactive environment for users to share the information.
   

I look forward to go back to work on those new opportunities soon.😁