LinkedIn is a global leader in job networking and search services, backed by its exceptional reputation; however, we believed we could further enhance LinkedIn's market value and customer satisfaction by improving its job search and application process.
We engaged in two major processes to ensure we created a user-centric design that would truly help LinkedIn succeed, starting with exploratory methods before the redesign, and then various user testing to inform and test the redesign.

We began our exploration by creating a task:
“Find an internship or entry-level job and apply through Easy Apply”
We walked through this task as a new user would and broke this task into 12 steps to analyze in detail. At each step, we identified potential issues and ranked them based on severity to prioritize which issues needed to be addressed first.

With the issues we identified in the cognitive walkthrough in mind, we turned to LinkedIn’s competitors to assess the current effectiveness of LinkedIn and propose ways to improve it.

From both the Cognitive Walkthrough and the Competitive Analysis, we were able to identify and define many UI and feature recommendations that we would tackle in the redesign.

In our cognitive walkthrough, our team had identified that the features in the current information architecture may be confusing. So, we utilized OptimalWorkshop to conduct a hybrid card sort and allow users to move features into the categories that they felt were appropriate.


Focusing specifically on the Jobs page features, we found the following:

How did we know that the issues we had previously identified were actual issues users face though? In order to test our hypothesized issues and maintain a user-centric mindset, we used UserTesting.com to create tasks for users to think-aloud while completing and then having them answer a few qualitative questions about their experience.

With our hypothesized issues confirmed with the qualitative test, Mela and I were able to create our redesign to incorporate the new features and information architecture. Below is a breakdown of the changes we made on each page involved in the job application process.

Quantitative Test: Did our changes have the effect we expected?

After creating a Figma mockup mimicking the existing LinkedIn flow and another for the new design, we hoped to ensure that our new redesign was truly more intuitive to use. To do this, we utilized both Maze and Google Forms to conduct within-subject AB tests with 29 participants, randomizing which version the participants analyzed first and gathered their opinions about each and comparing them.

Our team chose the very ambitious task to improve upon such a well established site and it was a great experience working with an established site instead of one that needed a lot of work. This project also allowed me to dive deeper into the user research side of UX and define our own goals as we moved through the project. I was also able to experiment with so many research platforms such as OptimalWorkshop, UserTesting.com, and Maze. Thank you to my team members for all your hard work and participation in all the meetings we had.