Accurately assessing candidate fit and skill level is critical to making a great hire.
Unfortunately, it’s also one of the biggest challenges of talent acquisition. Employers have their tactics, but they don’t seem to be working terribly well: More than 50 percent of voluntary turnover happens during an employee’s first year on the job, and as much as 80 percent of turnover may be due to bad hiring decisions. No matter the techniques they deploy, companies are by and large failing to assess talent accurately.
Enter technology. Thanks to advances in HR tech, recruiting teams have already made their processes more efficient, automating away the menial tasks to increase speed and optimize results. Now, new technologies are offering unprecedented insight into candidate performance and fit, thanks to prescriptive analytics.
What Are Prescriptive Analytics?
Sometimes called “predictive analytics,” prescriptive analytics uses statistics, modeling, machine learning, and artificial intelligence to analyze current data to make predictions about the future. For a talent acquisition team, this means using existing employment data to predict which candidates are likely to be top performers who stick around for the long haul.
To understand prescriptive analytics, it is essential to first understand intelligent automation. “Intelligent automation” refers to data-driven algorithms that help talent acquisition teams speed up processes. Though these algorithms are more efficient than humans when it comes to analyzing data, the real competitive edge of automation comes from the lack of human bias.
Recruiters and hiring managers are great at forming relationships, but not so great at making unbiased decisions. Researchers have studied hiring discrimination for years — and the resulting picture is a bleak one. According to one study, white candidates get 36 percent more callbacks than black candidates and 24 percent more than latinx candidates. These results suggest that hiring discrimination has not lessened much if at all over the past 25 years.
This goes to show that even the most experienced professional with the best of intentions can suffer from unconscious bias. Machines, however, are less prone to bias — and they can analyze significantly greater amounts of data while they are at it. This frees HR professionals up to focus on the high-touch, human side of talent acquisition.
Making Hiring Decisions Based on Prescriptive Recommendations
Using predictive analytics, recruiters can predict the likelihood of a candidate’s future performance based on historical data. Predictive analytics models use known results to develop a prediction value for new data. These models offer recruiters predictions that represent the probability of the target variable — for example, a loyal candidate — based on estimated significance from a set of input variables.
While predictions are nice, simply predicting that a person is a good fit for your company is only one small step of the equation. What if they don’t accept the offer? What if they are highly likely to leave within the first 90 days? Recruiters focused on high-quality candidate outcomes need to know who is most likely to accept an offer and stay with the team long enough to make a substantial impact. This is where prescriptive recommendations come in.
Prescriptive recommendations map qualified candidates against the talent DNA of top performers in order to offer precise recommendations that help recruiters identify the candidates who are most likely to succeed and most likely to stick around. Using prescriptive recommendations gives you the ability to find the right talent for your company while measuring the likelihood of that talent’s future success.
A version of this article originally appeared on the Oleeo blog.
Jeanette Maister is managing director of the Americas for Oleeo.