5 Ways Companies Are Using Predictive Analytics In HR

By Dan Harrison, PhD - March 07, 2019

What if you could predict the future? In a business context, such predictive powers would allow you to see and address potential problems before they develop, and anticipate changing workforce needs. 

For HR, the ability to predict the future has arrived. While HR has for many years had the ability to use data to understand workplace factors that affect the employee experience, the data was primarily descriptive of the current state. The advance of technology has created a new horizon for HR: the capacity to not only explain the status quo, but to use predictive analytics to see what is coming around the bend. (Tweet this!) 

The use of HR predictive analytics began trending several years ago, but it is gathering momentum. This article will present examples of how predictive tools for HR are being used to improve processes and free up HR professionals’ time for more consultative work.

Perceptyx surveys give you the exact data you need to succeed with predictive analytics. See how it works—sign up for a demo today.

5 Ways To Use Predictive Analytics In HR

1. Recruitment & Hiring

HR predictive analytics can play a role in recruiting talent by helping identify where that talent may be found. In general, people analytics practitioners have a lot of information about the labor market; predictive modeling based on this information can suggest where an organization might have the most success in recruiting candidates with a particular skill. For example, if a company needs software developers, predictive modeling may highlight a strong workforce emerging overseas, where good talent is plentiful and less expensive than in Silicon Valley.

Beyond locating sources of talent, HR predictive analytics can be used to identify the individual characteristics and types of experience that are most predictive of success in the jobs that need to be filled. This information forms the basis of the sorting algorithm for computer screening of applicants; using such data and automated sorting processes, a pool of thousands of applicants can be easily winnowed down to a few dozen or less. The use of predictive data allows applicants to be sorted quickly and with more accuracy, giving HR more time to concentrate on the most qualified applicants.

Data for the sorting algorithm can be derived from analysis of the existing talent base or external data, to identify the key competencies and factors associated with success. High performers, identified by data or the organization, can also be interviewed to determine the qualifications and competencies that are suggestive of high performance. This information can then be used to “teach” the machine to sort and aggregate candidates based on how well they align with those predictors.

After the pool of candidates has been narrowed through this initial sorting, predictive analytics can again be employed for further screening exercises. Psychometric assessments can measure candidates’ numerical reasoning, abstract reasoning, and task-based problem solving abilities to see how well they align with the demands of the position. Research shows that personality, behavior, and intellectual ability assessments are also extremely predictive of future performance and engagement on the job. Through these further screenings, the machine can help to narrow the list further to include only those candidates who are the best cognitive, behavioral, and cultural fit for the organization.

In addition to identifying applicants who are the best fit for the position and the organization, computer sorting eliminates the problem of bias. Unlike human screeners, the computer has no unconscious or subconscious biases related to race, religion, gender, age, or other factors, preventing candidates who are good matches from being eliminated from consideration for reasons unrelated to their qualifications. This unbiased assessment can strengthen the organization in terms of inclusivity and diversity—both of which are important for innovation and business success.

2. Onboarding

Just as candidate experience surveys can predict a company’s future success in attracting new talent, information from employee onboarding surveys can predict the employee’s future success within the organization. Research shows that the employee’s experience in the first 30–90 days will color their later attitudes about the company—and their performance. Employees who had a poor onboarding experience—they were not set up for success in terms of resources, training, and support—are predictive of future turnover if those issues are not addressed. Data collected from onboarding surveys identifies the processes that need improvement to reduce future attrition.

3. Workforce Planning & Management

Just as predictive analytics can be used to help determine the best new hires, a similar process can help identify members of the existing workforce who are most likely to succeed in a different position.

Many times large organizations lack knowledge about the existing workforce needed to leverage their skills. With access to data about individual employees’ preferences, skills, and aptitudes, predictive modeling can reveal which employees are closely aligned to new roles and any knowledge or skills gaps that need to be addressed for a good fit. This type of information can be gathered through survey items designed for skill and aptitude assessment. Other survey data can identify employees looking to make a move in their careers, either up in the company or out of it. Tapping these employees for a new role allows them to advance in their development without leaving the organization.

Understanding the organization’s informal network of knowledge and influence brokers is also important. This information can’t be easily identified through hierarchical data, but with tools like Organizational Network Analysis (ONA) and survey items questioning employees about who they go to for knowledge and who they identify as strong performers, HR can identify informal leaders. This information can predict which employees would be influential advocates and opinion leaders through change, and who should be kept on board during downsizing or transitioned to a new role if their current position is being eliminated.

4. Predicting Attrition Risk

Perhaps the most widespread current use of predictive analytics in HR relates to predicting attrition risk through responses to engagement questions—specifically in response to the intent-to-stay question. (Tweet this!) Employees who score lower on agreement scales in response to the statement, “I intend to remain in my job for the next 12 months” are more likely to leave than those who score high, and the lower the score, the sooner they are likely to depart.

This measure can be used broadly, from rethinking job roles and opportunities for career growth to providing targeted nudges to management. These nudges raise the alert when employees are approaching the length of tenure correlated with higher attrition risk; managers then have the opportunity to schedule career development conversations with those employees to determine what will influence them to stay.

Using Predictive Analytics Beyond HR

Predictive analytics can also be applied to improve customer satisfaction. Since there is a positive correlation between high customer satisfaction and high engagement levels, identifying the relationship between key indicators of customer satisfaction and workplace culture offers predictive data on what needs to be improved in lower-scoring departments or locations to achieve a better outcome in customer satisfaction.

5. Employee Exits

Exit data can be predictive primarily in terms of raising the alert over aspects of the employee experience that failed those who decided to depart. Exit data predicts that, if these aspects are not improved, employees will continue to leave the organization as a result.

The annual census survey is often predictive of exit survey data. Typically the elements of the experience rated as subpar on the annual survey will be cited as a reason or influence on the decision to leave on an exit survey. Both the annual survey and the exit survey allow leaders and managers to get ahead of the curve in addressing improvements to elements of the experience most closely correlated with reasons for leaving, to reduce attrition and the associated costs and disruption.

Is your company ready to take advantage of HR predictive analytics?

Considering the promise of predictive analytics for improving HR processes, it’s unsurprising that it’s one of the hottest topics in HR circles.

At Perceptyx, we design surveys to address your biggest concerns and collect the data you need to predict future workforce trends. Our platform helps HR departments like yours achieve important goals such as reducing attrition and attracting the right talent; it also helps you get out in front of potential issues to influence outcomes. The predictive analytics power of the Perceptyx platform frees up HR practitioners to take on more of a consultative role—instead of devoting their time to putting out fires. Getting ahead of challenges before they emerge also empowers other stakeholders with the insights they need to improve decisions and outcomes for the organization—and raises HR’s profile and importance in the company.

Contact Perceptyx today and see how we can help you transform the way HR does business.

Download Now: Using Cross-Survey Analytics - Perceptyx


We promise that we won't SPAM you.