Taking A Data-Driven Approach To 2019 Hiring Plans With Machine Learning

Carlos Tobon
Carlos Tobon
Head of Demand Gen

  • 74% of hiring managers and recruiters say they’ve hired the wrong person for an open position according to CareerBuilder.
  • 75% of demand for new employees is to replace those who have left the company, making churn one of the most expensive Human Resource Management (HRM) factors to manage today according to recent research published in Undercover Recruiter.
  • Companies are losing an average of $14,900 on every bad hire according to a recent CareerBuilder poll.
  • 22% of failed hires are the result of having insufficient talent intelligence before recruiting according to a CareerBuilder survey.

The best resolution any HR professional can make is to eliminate bad hires in 2019 by closing the growing gap between the skills hiring managers need and candidates recruiters find. It’s a clear sign the recruiting process is broken and needs to be reset on a data-driven foundation.

Relying on the centuries-old approach that hasn’t changed much since Leonardo da Vinci created a handwritten resume from 1482, organizations on average select the right candidate only 26% of the time. Applicant Tracking Systems (ATS) automate recruiting, failing to find the best possible candidates because they’re more focused on analyzing resume keywords and less on what all available data says about a candidate’s potential to grow.

How Data-Driven Recruiting Finds High Potential Candidates

Chief Human Resource Officers (CHROs) and the hiring managers they serve need to redefine recruiting based on the quality of all candidate data, not just resumes alone. Aggregating all publicly available data, internal data repositories, HCM systems, ATS tools, and spreadsheets to create ontologies based on organization-specific hiring goals is the first step. Having created an integrated platform that provides a data-driven 360-degree view of candidates, recruiters and hiring managers can define their ideal candidate by position and build a pipeline, achieving personalization at scale.

The inability of hiring managers and recruiters to agree on a clear, succinct definition of just what skills the ideal candidate needs and what their potential is slows down recruiting. Without clarity regarding the ideal candidates’ profile, both recruiters and hiring managers churn through thousands of resumes using ATS tools when available.

Talent management platforms built using AI and machine learning algorithms are giving recruiters and hiring managers the flexibility of emulating the core capabilities of their current stars and searching for candidates that match star employees. Candidate pipelines for the most in-demand positions can be created in minutes. Eightfold’s approach to emulating an ideal candidate while providing a data-rich 360-degree view of where comparable high potential candidates can be found is illustrated in the following graphic:

Eightfold emulates and finds the ideal candidate

Comparing High Potential Candidates on Key Attributes

HR recruiters and hiring managers both miss the capabilities, unique strengths and innate skills candidates have when they only review resumes. Instead of relying on a 500+-year-old process that delivers just 26% successful hires, it’s time to capitalize on the many advantages a data-driven approach. Using an AI-powered platform, HR recruiters and hiring managers are evaluating high potential candidates on the following key elements in their profiles:

  • Career Growth Bell Curve – Illustrates how a given candidate’s career progressions and performance compares relative to others.

How a given candidate’s career progressions and performance compares relative to others.

  • Social Following On Public Sites – Provides a real-time glimpse into the candidate’s activity on Github, OpenStack, and other sites where technical professionals can share their expertise. This also provides insight into how others perceive their contributions.

A real-time glimpse into the candidate's activity on Github, OpenStack, and other sites

  • Highlights Of Background That Is Relevant To Job(s) Under Review Provides the most relevant data from the candidate’s history in the profile so recruiters and hiring managers can more easily understand their strengths.
  • Recent Publications – Publications provide insights into current and previous interests, areas of focus, mindset and learning progression over the last 10 to 15 years or longer.

Publications provide insights into current and previous interests, areas of focus

  • Professional overlap that makes it easier to validate achievements chronicled in the resume – Multiple sources of real-time career data validate and provide greater context and insight into resume-listed accomplishments.

A data driven approach to recruiting:

Resumes are rear-view mirrors that reflect the past while AI and machine learning provide data-driven insight that predicts future employee performance. Finding high potential candidates by emulating the star employees at a company can be accomplished in minutes, saving untold hours of recruiters and hiring managers churning through resumes and potential applicants. Evaluating candidates based on data and their similarity to star employees also enables greater diversity and inclusion. The bottom line is data is the great equalizer, enabling organizations to create a true pipeline of high potential candidates while minimizing the effects of conscious and unconscious hiring biases.

Related posts