Not yet a Member?
HR Magazine is highlighting the next generation of HR leaders.
Is your employee handbook ready for the New Year? With SHRM’s Employee Handbook Builder get peace of mind that your handbook is up-to-date.
Get the HR education you need without travel expenses or time out of the office.
Join us in Chicago for the latest trends and technology in talent management, and what to expect in the future.
Two experts on HR metrics debate the issue.
To become full business partners, HR must be able to measure—and predict—business results.
HR leaders are quick to say “People are our most important asset.” But this statement is incomplete unless it continues “ … and therefore our biggest potential liability.”
Managing assets is a business process that requires effective analysis. HR’s desire to be taken seriously by business leaders can only become reality if HR is capable of demonstrating how it affects business results. Doing so requires predictive analytics.
Many HR organizations invest in talent acquisition systems and processes that harness significant employee data. Implemented effectively, talent acquisition methods provide HR with invaluable baseline information as to how new hires perform against incumbents in the same positions. Not only can this information be applied to the employee life cycle, it also serves as a baseline and comparative source for predictive purposes.
Here is a real case. My team and I led a five-year global study of a transportation company. We found that there was a 74 percent voluntary turnover rate among about 9,500 customer-facing employees. The annual cost translated to $68 million. The consequences were low employee satisfaction, which caused low employee engagement, which in turn resulted in low customer satisfaction.
Further analyses showed irrefutable evidence that voluntary turnover was impacted by four specific personality traits: resilience, adaptability, entrepreneurship and the ability to deal with ambiguity. Scenario planning based on predictive analytics suggested that hiring for these traits would reduce annual voluntary turnover by at least 28 percent. When the company revised its selection process, it reduced turnover costs by $27 million over three years. During the same period, employee satisfaction rose by 41 percent and customer satisfaction by 52 percent.
Additional modeling helped predict performance ratings, which were in turn instrumental in creating a more equitable rewards structure. It also helped identify emerging talent for succession planning, resulting in improved retention. The result was an incremental, annual productivity gain of $1.7 million. Applying predictive HR analytics transformed an HR liability into a real asset.
While some people claim that analytics can be used to discriminate, the reality is that companies with the intent to discriminate will do so with or without predictive HR analysis. But the benefits outweigh potential risks.
Predictive analytics position HR as an equal partner within the business. For one thing, they drive evidence-based decision-making, thereby shifting the dialogue between HR and business leaders from one focused on cost to one focused on investment. For another, they create accountability for business performance. HR’s success is no longer measured merely by the return on investment of HR programs. Rather, success is measured by HR’s impact on business results.
The real beneficiaries, however, are employees. Predictive analytics enable more-effective development structures; harmonize performance management; lead to more-efficient and more-equitable rewards systems; and lay a strong foundation for talent identification, retention and succession. In short, they create an environment in which employees can be their best.
Allowing employers to model employee data can lead to discrimination by algorithm.
Many analytics systems measure frivolous characteristics unrelated to individual job performance. For example, according to an article in The New York Times, Google has used an elaborate survey that explores applicants’ and employees’ attitudes, preferences and values on seemingly innocuous aspects of their personal lives. Individuals may be asked, “What magazines do you get? What pets do you have? What’s your favorite ice cream?”
Of course, in many cases, predictive data have been shown to be linked to relevant job indicators. For example, a job applicant’s hometown is a relatively accurate predictor of attrition. Applicants born and raised in big cities are more likely to leave their job than folks who grew up in small towns. But is it appropriate to rule people out on a data point over which they have absolutely no control?
Such data are dubious at best. I suppose the thinking goes that dog lovers from small towns are more loyal than cat people raised in an urban jungle. Or maybe employees who favor butter pecan make better leaders than plain-old vanilla folks.
HR, like most professions, is built around norms, values and ethical principles. The Society for Human Resource Management’s
Code of Ethics states that HR professionals “are ethically responsible for promoting and fostering fairness and justice for all employees and their organizations.”
Yet some well-educated members of our profession are having difficulty distinguishing between the law and ethics. For example, during a conference presentation in which I shared questionable data practices, an HR luminary asserted that he had no problem using predictive analytics that are legal and that don’t discriminate against a protected group.
But just because a practice meets the letter of the law doesn’t mean it is ethical. Is it fair to assess people by virtue of personal data that may be linked statistically to certain job behaviors but that does not, in reality, have any bearing on how a given individual will perform? Such “penalizing by preference” is the very essence of what it means to discriminate.
Moreover, in some cases the legal repercussions may not be immediately clear. At another gathering of HR professionals, an organizational psychologist shared that his company monitors nonexecutive employees who “dump” their stock. His company believes that such behavior is an accurate indicator of disloyalty and imminent attrition, and therefore it terminates some employees found to have engaged in it. A savvy attorney could make a case for wrongful termination by demonstrating that a systematic bias had been applied (i.e., that termination for stock-dumping was a pretext for discrimination).
Unfortunately, the genie is out of the bottle with regard to predictive analysis. It will probably take country-specific legislation to sort out what we can and can’t do with employee data. In the meantime, it is up to HR professionals to proactively address ethical quandaries and challenge questionable practices.
Salvatore Falletta is the president of Leadersphere and an associate professor for human resource development at Drexel University in Sacramento, Calif.
You have successfully saved this page as a bookmark.
Please confirm that you want to proceed with deleting bookmark.
You have successfully removed bookmark.
Please log in as a SHRM member before saving bookmarks.
Your session has expired. Please log in again before saving bookmarks.
Please purchase a SHRM membership before saving bookmarks.
An error has occurred
Recommended for you
Become a SHRM Member
SHRM’s HR Vendor Directory contains over 3,200 companies