Google Cloud Jobs API: Smarter Job Search Using Machine Learning

Google Cloud Jobs API: Smarter Job Search Using Machine Learning

Joe Essenfeld

When it comes to job search, job seekers and employers might as well be speaking two different languages. The job titles and descriptions employers use are fraught with company lingo and not intuitive to job seekers.  This creates a language gap between employers and job seekers.

Making this problem worse, the job search on companies’ career sites is unable to translate jobs and search queries effectively enough to close this gap.  When Google released the Cloud Jobs API, they brought a new solution to market with the capability of effectively translating and closing the “language gap” between job seekers and employers.

Before machine learning was an option for job search on career sites, the traditional keyword search used was akin to my daughter’s ability to talk at 18 months.  She was getting good at using individual words and pointing.  It would take me however, a few tries to figure out what she wanted.  With some work I could get close and we would communicate.  The same is true for traditional career sites that just rely on keyword matching to determine the relevance of listings. Job seekers have to work hard to find a relevant job on a career site.  They need to try different keywords, use filters, and manually adjust the results to get to the right job they were looking for.  All of that takes effort and time.  That extra effort and time results in a lot of job seekers leaving a company’s career site before finding the right job.

When you swap out traditional keyword search with a machine learning algorithm, the communication between employer and job seeker becomes more seamless.  It reminds me of talking to my daughter now.  She is a fast growing 2.5 year old who speaks in full sentences. Even if all of her words aren’t correct I have enough context to know what she wants.  This means less time guessing or asking her to point.  We are able to communicate about more with less effort on both sides.  As a result there is less crying and frustration.  The same holds true for job seekers – even the crying!.

Like the example with my daughter, when you add machine learning to job search it is like the career site can speak in full sentences.  This happens because the machine learning algorithm assigns meaning to jobs by factoring in the information contained in the entire job description and job title.  With an actual understanding of the job, it can suggest more relevant jobs to job seekers even when they use vague queries to search.

This intelligent matching has a profound effect on recruiting efficiency.  Job seekers are more likely to apply to a job after performing a search backed by Google’s Jobs API and are less likely to bounce.

Solving the language gap between job seekers and employers increases the conversion rate of of job seekers starting an application after searching for a job. This increase in conversion also boosts the quality of applicants for companies. Our aggregated client data shows a company is almost two times as likely to hire someone that searches for a job on their career site versus someone who finds their job on a job board.  This metric suggests a company’s money is best spent getting more applicants directly on their career site versus casting a wider net on job boards.  Google Cloud Jobs API provides a better experience for job seekers that translates into more qualified applicants and lowers the cost and time required to fill positions.

Interested in the topic of employer branding? Check out our new eBook, The 2016 State of Employer Branding. This eBook is key for companies both starting or well into their employer branding journey:

employer branding ebookcareer site assessment