Employer Hire Survey

UX / UI

Indeed needs to know if their employers are successfully filling their positions, but the existing way of measuring was unrefined and returned messy data. It started as a task to add a button; in the end, I rethought the entire flow and collected more, cleaner data for the whole company.

Employer Hire Survey front screen.jpg

My role

I was the sole designer on this project, working with project managers, content strategy, the design systems team, and the engineering team.

The challenge

The senior leadership team had declared better hiring data to be a crucial goal for 2019 (along with addressing the “black hole”). We relied on employers and candidates to tell us when a hire took place. The primary medium for reporting this info was the “post-hire survey,” an optional, bare-bones, one-size-fits-all modal that appeared when employers paused or closed a job.

Bonus challenge: During my research, I discovered that employers were using the “other” input to tell us about all of the different problems that were causing them to close their jobs. How can we help them solve those problems instead of closing their jobs?

The solution

When an employer changes the status of a job, provide a survey that is appropriate for their situation. Make that survey quick but not-optional. And if they’re closing their job in frustration, provide resources to help them find success instead of giving up.

The approach

The survey had been tweaked for months (or years) before I was involved, and this time around I had been asked to simply add a button to first answer “did you make a hire?” and then ask for a reason. That didn’t seem like a particularly effective improvement for getting more users to participate, so I explored other possibilities. After all, this was our primary instrument for understanding if our product was successful at its primary purpose – we needed those responses!

› Listen to the users

During my research, my product manager linked me to a treasure trove of “other” responses from the employers who had filled out the existing survey (over 1 million of them). I scoured the responses and (unscientifically) identified common complaints and problems – technical bugs, billing problems, accidental job posts, etc. Those responses guided the options presented in the survey, better equipping the data science and customer service teams to track them.

› Educate and clarify

I updated the content and hierarchy to explain what happens when an employer changes their job status – details that were nearly impossible to find, forcing users to guess what “closed” and “paused” mean. I also applied the current design system, which was long overdue and made the survey consistent with the rest of the pages users has been seeing.

In qualitative user research sessions, employers found the survey to be easy to understand and responded that they had no significant problem with the survey being mandatory.

The outcome

The reworked hire survey was a success. Measured hires increased by 15%. The data science team reported that they were excited to have cleaner, more actionable data about job statuses and hires. As planned, the survey options continue to be refined as we learn more about employers' habits. The survey is not yet mandatory for employers changing the status of their jobs, but there are plans to test that part of the solution.