Okay I admit it. I am a data geek, and a firm believer in the power of analytics. It is amazing the amount of intelligence that can be gleaned from publicly available data. I recently attended the HR Technology Conference & Expo in Las Vegas and learned how some providers are using predictive analytics and aggregated data to hone their talent acquisition solutions.
The concern I have is that as we continue to pioneer new solutions that impact candidate selection, there should be best practices with respect to transparency, degree of accuracy, intermingling of data sources, and delineation of what constitutes the original candidate profile, as indicated in a resume, and what constitutes the added enhancements. Solution providers must keep their algorithms private, but it would be good to know they are using their “secret sauce” in accordance with industry best practices. Okay, I know I am opening up a can of worms with this statement, but let’s not be too afraid to dive in and see what we can and cannot accomplish, using job-posting distribution and candidate sourcing as examples.
Job Posting Distribution: Several providers are offering solutions that automatically select or suggest the best sites to publish job postings, based on past success metrics and/or past job distribution behavior. Examples of some of these providers are: SmartRecruiters, Gild, and Smashfly. Smashfly automatically distributes jobs to the boards always used for certain jobs without recruiter involvement. SmartRecruiters uses predictive analytics to identify the best job boards.
Candidate Sourcing: Solution providers are using a variety of internal (to the client) and external data sources to help employers identify the best candidates by adding “intelligence” to the candidate profile or the sourcing/candidate selection process itself, such as likely time-to-fill a position based on market parameters, likelihood of a candidate succeeding through the organization’s hiring process, likelihood of a candidate changing jobs, candidate demand, skill ratings, and other metrics. Some of these providers using a variety of approaches include: Gild, Entelo, Monster’s TalentBin, PeopleFluent, and Randstad Sourceright’s Talentradar – as part of their Recruitment Process Outsourcing solution.
Monster’s TalentBin, which focuses on the technical talent market, provides a view of each candidate’s professional and social interests based on that candidate’s social footprint. Gild uses its technology to gather data from many sites and automatically rank prospects based on their skills.
Some Philosophical Questions
I think we are seeing just the beginning of how data is being used for talent acquisition. As I mentioned before, my concern is mostly centered on how this new trend impacts candidate selection. Computers are not perfect (though very close to it, I admit) and not all candidates have a social footprint, and those who do can have varying levels of information available. Are candidates who have a limited social footprint at a disadvantage when a solution provider enhances those candidates’ resumes? How is that handled by the solution provider? Which types of data sources should be excluded from these algorithms? How much information should be transparent to the employers and candidates?
Employers will always make subjective decisions about who they hire. Solution providers are helping employers improve their decision-making process, but let’s do it in a way that is somewhat transparent to the employer and the candidate. I look forward to seeing the future of data analytics in talent acquisition and how the industry will manage the process.