My Role: UX Researcher and Designer | Duration: 2 Weeks | Project Status: Complete
Project Overview
In this unit project, I worked both independently and in a group with Shraddha Jain, Lena Nazaryan, and Charlie Frenz to integrate a new feature into an existing mobile app. This project was designed for us to move through all phases of the UX double diamond design process.
Scope of Work
The COVID-19 pandemic has forced many companies to turn the majority of their recruitment efforts online, and it has also forced may job seekers to turn their primary job search efforts to online searches and applications, including on mobile devices. My group members and I were hired as UX designers to explore and discover a solution that would support the remote experience. Our assignment was to redesign one aspect of remote job seeking (i.e. scheduling, screen sharing, face-to-face communication, networking with recruiters, etc.) within an existing mobile app (i.e calendar, messaging, video conferencing, screen sharing, etc.).
Process
We learned how to interview users, extract insights from our interview findings, turn those insights into personas, ideate low-fidelity solutions for our personas, create a prototype in Figma, test that prototype with users, extract insights from our tests, create mid-fidelity wireframes in Figma, and present our designs to an audience in a way that explained and sold the value of our UX work.
RESEARCH PHASE (Discover + Define)
Goal of Research
Our group’s hypothesis was that with more jobs being available than ever to apply online, users tend to only apply for positions that are easy to apply for unless it’s a targeted job. We also believed that because of the pandemic, more people apply to jobs online and change jobs more frequently. We also came up with the following assumptions:
Candidates are more likely to apply if the process is time-efficient
Candidates are less likely to re-enter information they have already provided on their resume
Candidates don’t like to re-enter information they have already provided
Candidates will apply to more jobs if the application process is direct and quick
Our initial problem statement was that candidates need a more efficient method to submit their info to employers. We wondered: how might we make the job application process easier by adding a feature that automates their information from their resume directly into the required forms of an application?
Methodology
My research team and I began by considering potential issues that users may encounter when completing online job applications. After brainstorming our hypothesis and assumptions, we created an interview discussion guide. Next, we remotely conducted and recorded a total of 5 user interviews over the course of three days, after which we collaborated on creating an affinity map.
Synthesis » Affinity Mapping
We initially created ten groupings, but some of the groupings had so many data points that I had to regroup them when each group member went on to work individually. I regrouped my revised affinity map into 17 groups; by the time I made a second revision, I was able to narrow it down to 11 groups by shifting some of the data points around and removing some of the other data points altogether.
Findings + Themes + Insights + Takeaways The second revision helped me identify the following insights:
Users tailor resumes and portfolios to the job they’re applying for
Users get frustrated when the application process takes too long
Users don’t like repeating data entry for multiple job applications
Users like applying using as few steps as possible
Technology/access issues can pose challenges when applying to jobs
Users often get either fraudulent responses, or no response at all.
My revised affinity map
Persona
Meet Eve!
The initial group affinity map lead to my first creation of my user persona Efficient Eve. By examining the pain points that the users shared during their interview, I used that information to bring Eve, a representation of our users, to life. I found that the most challenging part of this process was determining how to construct and clearly illustrate Eve’s behaviors, especially in a way that could reflect how those behaviors might impact Eve’s job search. I was also challenged by figuring out how to articulate Eve’s needs, goals, and frustrations without repeating too many of the same pain points. Refining the group affinity map on my own and incorporating the feedback I received on the first submission enabled me to make Eve more detailed, hence Eve 2.0.
Meet Eve 2.0!
Revised Problem Statement
In our research, not only did we find that a significant number of users used LinkedIn’s mobile app to submit job applications, but a common, reoccurring pain point was that users didn’t want to have to spend a long time on job applications, especially when it came to having to reenter information that could already be found on a user’s attached resume.
My revised problem statement was that Eve needs a mobile-centered way to apply to jobs online so they can apply to high-quality jobs quickly.
I came up with the following “How Might Wes”:
How might we reduce the amount of times that Eve has to re-enter previously entered information on an application?
How might we improve Eve’s ability to submit multiple applications in a short amount of time?
How might we facilitate employer responsiveness to Eve’s application submissions?
How might we make online applications more mobile-friendly?
The “How Might We” I chose to focus on was, “How might we reduce the amount of times that Eve has to re-enter previously entered information on an application?”
Introducing FastApply
Based on the pain points users shared in their interviews, I began brainstorming for FastApply — a way to auto-populate job applications on LinkedIn’s mobile website. According to our research, a few users mentioned using a computer to do certain job application tasks depends on how much typing was needed or if a portfolio was involved. My idea was that a user could use a computer to log in to LinkedIn and enter their profile information, in addition to resume information that could be later used to auto-populate job applications via the mobile app. A user could also save search results and sign up for notifications for certain jobs and job types; the user would begin by tapping on the job notification to get to the application.
My first user flow completely missed the mark. I made it based on the user I interviewed instead of coming up with user flows based on my ideas.
User Flow #1
After working with teaching assistant Sonya Alexander Pevzner to gain further clarity and guidance on how to refine my user flow, I designed a second version.
Revised User Flow
Design Studio: Sketching + Concepting
I began this phase of design by downloading the LinkedIn mobile app, looking at the interface, and learning how to navigate around the app. The next step was to begin sketching screens based on the design, figuring out how I wanted to implement the changes visually, and determining what elements were needed.
Some of my initial sketches before wireframe construction
Mid-fidelity Screens
Transitioning from sketches to wireframes helped me get a better idea of how the screens could flow from one page to another, and it also gave me the idea of having confirmation pages and timestamps for when a user successfully submitted an application.
My first wireframes using my sketches
Design Iterations: First Round (Annotations/Screen Flow)
When I began creating the first prototype, I got the idea to add a confirmation time stamp for a successfully saved application.
My first design iteration
Final User Testing (What I tested, what I learned)
I recruited three users to test my prototype using the following plan:
Purpose: My purpose is to test a new auto population feature in LinkedIn’s mobile app.
Goal: Users will be able to submit an auto populated job application via mobile app in less than 5 minutes, with no more than one error.
Logistics: The test will be conducted via moderated remote tests.
Persona: 28, East Coast USA, Freelance Artist
Tasks:
Apply your profile information to the job application
Submit your job application
Metrics: Success rate, time it takes to complete each task, number of errors, users’ likes, users’ dislikes, and users’ questions
I also created a script that I could use as a guide.
Findings + Themes + Insights + Takeaways
Two out of three users made no errors when clicking through, and all three users completed the tasks under 5 minutes each (which was a time frame I estimated), However, 1 user made the majority of errors. One of the things I noticed while watching this user was that they were hesitant. There were times when they weren’t sure if they wanted to save the application. They wanted the opportunity to review certain data, including their auto-populated information. They also stated that they wanted to have the option to potentially edit that information. I also created an affinity map based on my usability testers’ comments and feedback, and here were some key points I pulled out: • “The process was very easy.” • “It seemed pretty straight forward.” • “I do like these pop-ups along the way, giving you timestamps and everything else.” • “I want something to signal that I can put my information there.” • “…At this point, where it says your application has been submitted, my question would be now how do I get out of this?”
Prototype Affinity Map
Based on the feedback I received regarding the color (or the lack thereof) and some layout questions, I wasn’t sure if I set the interviews up correctly or if I gave the users enough information before beginning the test. I also realized that I forgot to time each step and had to watch the video recordings I made of each interview to gauge the time. In hindsight, I attribute my forgetfulness to nervousness.
Final Prototype
Using feedback from the users who tested my prototype, I added more detailed text to the first page, as well as a page that would show users what would happen if they decided to save a job application for later at the beginning of the process. This brought my final prototype to 7 pages, and the prototype completion culminated in a 7-minute presentation of our research.
Images of full file and final prototype
Recommendations + Implementation + Next Steps
The following are my next steps:
Create another prototype iteration with more specific icons (to better guide the user’s next steps)
Determine navigation options after application submission confirmation
Consider implementing a mobile option to verify profile information before submission
One of my main takeaways from this portion of the process is the importance of including as many recognizable elements and icons as possible. The more readily identifiable certain elements are, the more easily the user will be able to navigate the product. I would also like to refine my usability testing process so that I can garner more rich data to help me further improve the product.