Back to home

Indeed Job Seeker

The job seeker team owns the search engine results page (SERP) and job listing page. These two pages are at the core of the job seeker journey. My work helped millions of job seekers be more successful in their search.


Matching Signals on Job Cards

Why do the work
Job seekers wanted to see more relevant jobs and the business needed to deliver more qualified candidates to employers. Job seeker preferences were identified as something we could leverage to achieve both goals.

How I started
I teamed up with a UX researcher to roll-up the most relevant insights from several past studies on preferences.

Research informed a new hypothesis
Insights lead us to believe that if we show job seekers how their preferences and the job aligns, we will see an increase in key job seeker metrics (specific metrics omitted) in the same amount of time.

The challenge
One of the challenges was cross functional (xFn) collaboration. One team was in charge of the collection of preferences, our’s was in charge of surfacing them. Weekly syncs were set up to ensure we were all on the same page.

The biggest challenge was finding the right balance between information to display while ensuring we didn’t lose out on wins from current job card features.

Exploring design directions

In order to combat this, I brought in our content and brand team to collaborate on ideas and made sure teams who ran experiments on job cards were involved in the process.

The outcome
This work led to an overhaul of our job cards. A system on when to show certain signals / data was established to make sure we kept job card info lean and effective while ensuring we didn’t lose the wins from other features.

We released the updated job card and as a result we saw increases across all key job seeker metrics. Job seekers were finding the most relevant jobs faster and we were able to deliver more qualified candidates to employers in the same period of time.

Matching Signals on Job Listings

A fast-follow to the job cards work. We needed to figure out a good way to display matching signals on the job listing page to help job seekers make a decision to apply or not.

I conducted an audit of the different sections and meta-data permutations which helped get buy-in to take a more holistic approach to the design.

The biggest challenge was choosing between two initial directions: surfacing match signals in a "highlights" section or showing them in context of the existing job details section.

I collaborated with other teams and UX research to help identify the best path forward.

As a result, I introduced a more unified design. As with the job card work, we saw increases in job seeker success rates. This also provided more established guidelines which helped teams feel more confident in running tests on the page.

Conversational Filtering - India

India has a very unique segment of people who have only been using the internet for 1-3 years. Most are unfamiliar with basic features such as filtering which lead to irrelevant SERPs and high bounce rates.

The team already had seen some success with a WhatsApp chatbot, but wanted to do something new on SERP. I proposed we use learnings from the chatbot experience and lead with that instead.

It wasn’t as straightforward as I had hoped. There were issues around data storage and data fidelity. In the end there were some slight compromises to the experience I had envisioned.

Despite all that, the feature was released and we saw immediate gains in job applications and a large reduction in bounce rates.

This project was particularly meaningful to me. Not only were we seeing wins, we were also helping a large segment of job seekers to get familiar with technology.

Back to top