Problem: Jobs experience on Indeed was traditionally designed with small and medium businesses in mind. Those experiences did not accommodate complex needs of Enterprise employers.
Solution: We've created a simple experience with layers of complex functionality available if needed. It gave employers tools to efficiently manage a large number of jobs and make changes to get the number and quality of candidates required.
Impact: "This will change my life! I will no longer be dreading logging-in to Indeed."
My role: UX Designer in collaboration with UX Designer, Sr. UX Researcher, Content Designer, Product Managers, Engineers, and stakeholders.
More on ProcessMore on final resultDuring research rollup, we've learned that complexity is the norm for enterprise employers. Some recruiting teams have up to 20 team members, each managing up to 100 jobs at a time. All those jobs go into the same jobs list. Therefore, efficiency and organization are critical. The existing experience did not accommodate those needs. As a result, it was challenging and time-consuming to navigate and manage those jobs.
We've also learned that the hiring process in large organizations is a team effort. There were different people involved, but our existing job experience was not accommodating for those collaboration needs.
"It's just annoying. I'm dreading it come Monday, because it's the first of the month, and I'll be reposting jobs. It's annoying to have to keep sorting and sorting to find a job that I’m looking for."Employer about existing job management experience
After stakeholder interviews and initial research, we've realized that there were many problems to solve. To avoid trying to boil an ocean, we prioritized key areas based on potential impact and effort as well as the feedback from customer success representatives:
Next, we've defined Jobs to be Done so we can understand what was important for users. We tested them with users to make sure our scenarios were realistic.
As a recruiter, when a role opens unexpectedly and I'm feeling stressed and under pressure, I need to find a way to fill that position quickly with a qualified person, so that we can continue running a profitable business smoothly.
As a recruiter, when I'm just getting my day going unsure what to work on that day, I need to know what needs my attention, so that I can efficiently achieve my hiring goals.
During research roll-up, we've defined How Might We's that we later used in the brainstorming session. The brainstorming session included key stakeholders and cross-functional partners. Together, we came up with ideas that could help us solve the key problems that we identified. Some of them were:
For more innovative ideas, we did a conceptual test first to see what users thought about them. Then, based on the feedback, we either changed or eliminated them.
The first concept was around collaboration. It gave a recruiter ability to send the job description to their colleagues for review. After they received the changes, they were ready to finalize the job post. During the research, we've learned that recruiters typically don't collaborate on writing job descriptions. Instead, they needed to explain why a particular job description decision was made and support their decision with data.
The second concept was around tracking how different changes to jobs affect performance. In the storyboard, the recruiter sees that a job isn't performing well. So, he makes a change (let's say changes a job title) and waits a few days. Later, he comes back and compares how the job is performing now vs how it was performing before the change. Users thought this concept was helpful. Multiple users brought up an ability to A/B test different job titles.
The third concept involved guiding employers by using all the data we had. When an employer posts a job, we will show them an estimated number of applicants they would get with the job title, salary, location, and other details they have selected. We will tell them if they will get too many or too few candidates and what they could do to balance out the candidate flow. In addition, we will inform them how their most recent change affected the estimated applicant number.
The feedback on this idea was positive. Employers thought it would be helpful to have all this information if the estimated numbers were close to the actual results they would get.
After the concept testing, we created low-fidelity mock-ups. We tested them with users to get feedback on how those experiences worked. Then, we did a few rounds of testing that helped to identify and eliminate any usability issues. We iterated a lot and quickly and eventually got to designs that worked well.
To help recruiters navigate their jobs efficiently, we've introduced saved searches as an easy way to recall and focus on a specific group of jobs.
We also introduced an ability to star important jobs and assign a recruiter. That gave a manager a quick way to see who is working on a particular job. And recruiters could now quickly filter a long list of all job openings down to their jobs.
Finalized Design
To help recruiters quickly identify what jobs need their attention, we've introduced the concept of an applicant flow. It allows recruiters to see which jobs are lacking candidates and which ones have too many. By tapping the label, they can see a suggestion on what to do to improve the applicant flow.
Job can fluctuate day to day. Adding rules can help employers manage this dynamic list at scale. For example, they can automatically assign all new jobs in Austin, Texas to a specific recruiter.
