Putting the Right People to Work, Faster: Big Data for Recruiting Search and Match

According to Forbes, US corporations spend nearly $72 billion each year on a variety of recruiting services, staff, and products – and the worldwide number is likely three times bigger. In the recruiting industry, it’s a race to find the right candidate, making fast and accurate candidate matching is key.

The candidate search and match solutions on the market today are limited to basic resume parsing and a subsequent matching of the corresponding metadata to a job posting. Thus, most solutions fail to significantly improve open position fill rates or provide a simpler, better search experience for recruiters. This is where powerful big data recruitment applications can enable recruiters to know more and put people to work faster. It looks like recruiters agree: according to the Jobvite Recruiter Nation Survey results, 72% of recruiters say data analytics is important in the hiring process.

A look at our customer

Adecco is one of the largest HR companies in the world. They put more than a million people to work everyday, some at large banks, retailers and tech firms, others at small mom and pop shops. Since 2012, we’ve been working together to launch a customized search and match solution that will provide a competitive advantage in serving their customers. By combining advanced search techniques, the processing power of Cloudera’s Big Data platform based on Apache Hadoop, and the flexibility and scalability of the Microsoft Azure cloud, we’ve been able to provide sophisticated statistical and linguistic capabilities to understand applicants’ profiles and identify the best candidates for a specific job.

Adecco was invested in taking their digital strategy into the future and didn’t just want to reinvent the status quo. Their previous search system was inadequate to their needs: Boolean based keyword search and rudimentary in nature.

In order to generate high quality matches, it is essential to be able to compare each record to large volumes of unstructured data. This would allow Adecco to distil each record into a manageable set of key distinguishing features that can be used in order to find similar records.

In order to realize this goal, we built a custom application using, Cloudera Enterprise and integrated technologies for processing and preparing the candidate resumes for searching and matching. It is used to process and gain insight from the large volumes of data that are used to generate dictionaries of terms and their counts to discover semantic relationships such as co-occurring terms and common phrases. This provides key data to drive and improve the matching process and allows us to provide semantic resources in multiple languages without the need for our customers to maintain them.

This search and match solution runs in the cloud on Microsoft Azure. Adecco selected Azure for its global reach, ability to comply with regional data governance policies, and its ability to scale as data and workloads increase. Cloudera’s partnership with Microsoft ensures the data management and cloud platforms are well integrated and able to meet Adecco’s needs now and in the future.

Through this project and others, we’ve found that custom solutions are best for large recruiters because they each leverage their own unique strengths in terms of their existing application databases, company hiring histories, and preferences.


The resulting technology is a significant differentiator against other firms. By using Hadoop for deep processing of resumes and job descriptions, Adecco can unlock talent that was previously impossible to find. For example, in blue collar industries resumes are often one-line documents, or a short paragraph at best. The use of semantic expansion and the ability to statistically generate implicit skills and experiences based on industry or job title, enables recruiters to find candidates who previously would have been overlooked because they did not use the “correct” keywords in their profile.

In the first 90 days after implementing the “Search and Match” solution in Adecco’s North American Staffing Division, they saw a marked improvement in several business metrics including:

  • A significant improvement in fill rate (the percentage of job orders successfully completed), resulting in positive revenue growth
  • A 30% reduction in time to fill vacancies. Responding faster to customer needs with accurately matched potential candidates provides a crucial competitive edge.
  • A 20% reduction in job board spend due to better use of existing data.

The Search and Match solution is now a core piece of Adecco’s Front Office global strategy, powering both their recruiting and web solutions. Following the successful rollout of the project in North America, Adecco has extended the reach of this search and match application to Canada, Norway, UK, Switzerland, Germany, Japan, Italy, and Spain with more countries expected to come online soon.

Search Technologies is a Cloudera System Integrator Partner. Click here to learn more about their Cloudera Consulting and Implementation Services. Adecco is one of a number of large enterprises that have had success running Cloudera on Microsoft Azure. Click here to learn more about Cloudera Enterprise on Azure.


This guest blog is written by Susanne Ross, Solutions Marketing Manager at Search Technologies, a Cloudera partner specializing in the design, implementation and management of search and big data applications.

The post Putting the Right People to Work, Faster: Big Data for Recruiting Search and Match appeared first on Cloudera VISION.

Leave a Comment

Your email address will not be published. Required fields are marked *