Redesigning Kalibrr's Job Board

Mockup of the Kalibrr mobile web job board


At the time, Kalibrr's business model was that companies had to pay to "unlock" a job applicant's contact information. This meant that revenue was directly tied to how many job applications there were on the platform.


How might we increase the number of job applications on the Kalibrr platform?


Duration: 3 months
Research: User feedback, user interviews, analytics
Prototyping: Sketch mockups + Invision prototype
Testing: In-person usability testing, iteration w/ stakeholders
Strategy: Product strategy
Visuals: Consistent with platform's visual design strategy


Job board applications increased by 20%
Mobile job board applications increased by 38%



Below is a screenshot of the Kalibrr job board before the redesign.

Screenshot of the old Kalibrr job board


We had a lot of jobs in the platform but the average number of job applications each user made was around 2.5. Increasing this meant a significant increase in the possibility of generating revenue.

The hypothesis was that if we made it easier to find jobs that were relevant to them, each jobseeker would generate more job applications.

More job applications meant the jobseeker had an increased chance of getting hired and Kalibrr had increased possibility of earning from that job seeker's contact "getting bought".


Research indicated that users were more prone to search via the search bar than to browse.

Since a Kalibrr visitor's primary job-to-be-done was to find a job, we decided to double-down on the behavior of searching/filtering versus browsing. When searching, we enabled the user to narrow down options to jobs that were more relevant to them. This might mean less time on the site as well but it might lead to a faster accomplishment of his job-to-be-done.

With the goal being to increase the number of views each job post receives, the team's premise was that a more accurate search will lead to more views for job posts that the user deems relevant.

We made two improvements to the job board search experience. First, we added auto-complete search suggestions for companies. This made it easier for users to find the correct company names.

Second, we also improved the filters to allow multi-select filtering. The old design only allowed using one filter at a time. Allowing multiple filters to be selected reduced the number of combinations a user had to try to get good search results.


Research showed that a lot of job views through mobile were not converting to applications. The use of a tabular layout was not mobile-friendly and, at that time, a growing majority of Kalibrr jobseekers (~70%) were using their mobile phones to view the website. This prompted a reconsideration of the table design pattern to something that responds better to mobile phones: a card list layout.

Compared to a table layout, the card list layout was more mobile-friendly and was easier to scan. However, moving to this layout may diminish some sorting abilities that was natural with a table layout. (This was another good motivation to improve the filtering and sorting abilities of this page.)

In designing the cards, we conducted several quick tests to determine which information was critical to be displayed in each "job card". We talked to some of our users (via Intercom) and asked what information made them interested to know more about a certain job.

At the end of this quick study, we learned that users were more interested in a job if it had a job title, the company's name and logo, the location, the job level, a brief description of the role, the hiring deadline, and the job activity—i.e., how active or stale the job was. The last one was new for us. We didn't show this at all in the old version of the job board. According to our users, they were more likely to be interested in a job if the recruiter was active.


Below is a screenshot of the actual Kalibrr job board after the redesign.

Screenshot of the Kalibrr job board