AI-driven predictive analytics tools are a groundbreaking way to process large volumes of data, helping HR staff identify the right candidate for a role, spot high performers, or pinpoint retention issues. Predictive analytics is a component of nearly all modern ATS systems. But does it have hidden drawbacks?

The High Cost of a Bad Hire

Every recruiter dreads making the wrong hiring decision. A new employee who lacks the right skills, doesn’t perform to expectations, or doesn’t work productively within their team can be worse than having no employee at all. Some of the consequences of a bad hire include:

  • Lost productivity within their functional area, as they fail to meet expectations
  • Lost productivity among supervisory and training staff as they invest time and energy supporting the new hire to be more successful
  • Negative effects on team morale, company reputation, or customer service, depending on the role
  • And, inevitably, lost productivity within hiring and recruiting staff as they have to start the same talent search all over again

Estimates indicate that a bad hire can cost 30% or more of that role’s annual salary to replace, but a highly-ranked poor performer can cost much more. 

One often overlooked impact of a bad hiring decision is its effect on the recruiter themselves. When they have made a mistake, especially if they strongly advocated for the candidate, they may lose confidence in their professional abilities. A bad hire can even damage the reputation of the recruiter who placed them. Recruiters and HR staff have deeply personal and professional reasons to avoid making the wrong hiring decisions.  

The Role of Predictive Analytics

Modern candidate and human resources software systems often include a wide range of AI and machine-learning tools. These tools perform a wide range of functions, from answering candidate questions to pre-screening resumes to measuring employee success. Predictive Analytics is one of the most popular features of these software packages. 

Predictive Analytics does what the name implies: it analyzes vast amounts of data, and uses that analysis to predict future outcomes. Predictive analytics may look at industry-wide data to identify the skills, education, or experience that is linked to performance, review internal and external metrics to spotlight high or poor performance or review resumes and social media profiles to determine the strongest candidates for a specific position. These tools can establish correlations and key determinants that may be overlooked by human analysis, and save time by selecting the most relevant people, factors, or metrics for consideration.

A Word of Caution on Predictive Analytics

While predictive analytics greatly improves productivity and streamlines recruitment processes, it is critical to consider that these tools only measure past successes in order to predict future ones. As such, there are two main drawbacks to over-reliance on these tools:

  1. Perpetuates bias. Analysis of top historic performers may cause the AI to arrive at conclusions like “white men who attended Ivy League colleges make the best CEOs.” Global data and analysis is full of certain biases and stereotypes. Analysis of a company’s own internal historic data will reveal the ideas and preferences of the leadership team, which may also generate a specific, biased, portrait of an “ideal” candidate. The truth is, hiring decisions are full of implicit bias, but human screening and recruitment staff can be made aware of these tendencies and strive to overcome them. When software preemptively eliminates candidates from even being considered for a role, it can be difficult to know when bias is being perpetuated. 
  2. Stifles innovation. When we rely too much on historic data to drive future decisions, we create a safe, stable, risk-averse culture where creativity and innovation cannot thrive. We cannot use predictive software to anticipate the unpredictable, the unexpected, the flash-of-lighting inspirational moments that spark innovation and cause transformation. In the past, it may have been true that some industries, like banking or insurance, needed to be stable and avoid change, but today every industry benefits from some amount of creativity and innovation.   

Predictive analysis may make the work of recruiting too easy. By simplifying the decision-making process, streamlining choices, and reducing risk, it may make recruiters too comfortable. They may never have to take a chance on an unlikely candidate, advocate for a new approach, or consider an unexpected perspective. 

Predictive analytics streamlines and simplifies day-to-day decisions and may pay off well for performance and retention over the course of a career. However, it may also stifle the future potential of a candidate, a company, or even entire industries. 

For balanced insightful solutions for your recruitment challenges, contact grapefrute today.