In my last article, I wrote about the definition of HR Analytics and the skills needed to be successful in this field. In this article, I want to discuss two key benefits of HR analytics to the HR function in an organization and to the business: Evidence Based Decisions and Reducing Human Bias.
HR professionals want to be strategic partners with business leaders, not simply a cost center designed to maintain policies and procedures. While these policies are important, analytics provides HR with a means to demonstrably improve the efficiency of a company’s people resources. It does this in several ways.
Evidence Based Management Decisions
Through its dependence upon data and facts, HR Analytics delivers evidence, and evidence trumps intuition. To support these benefits, I’ll ask two questions:
Is your interview process optimized to find the best candidate for a position?
If you have ever participated in an interview process from the hiring perspective, you may be aware that at many companies, interviewing candidates can be an informal, non-standardized process. At worst, interviewees are simply asked by HR “What did you think?” More sophisticated HR methodologies define a standardized process for who the candidate meets and what questions are asked. At each stage, feedback is collected and quantified, typically in the form of ratings. Are these ratings predictive of future performance in the job role to be filled? HR Analytics can tell you the factors that are predictive of high performers in certain job roles (or tell you that you don’t know and that you should either change your process or collect different data points).
Does internal employee training improve company performance?
Most HR professionals would say ”Yes, employee training is a good thing and we need to do it.“ Many top companies spend precious resources to train their sales staff or send aspiring leaders to leadership training. Does this training have a material impact on performance? On the company’s bottom line? HR Analytics aspires to quantify that benefit. To do this, we may need to pull together data from several systems, such as on-the-job performance data, financial data and data collected during the training process. We should define the performance metrics that are most important in that job role. We must also consider a baseline of performance (i.e., comparable employees who were not able to take the training). By taking a more scientific approach, we can quantify the benefit and produce evidence of impact. We may also demonstrate that certain training is ineffective.
Reducing Human Bias
If you have read Michael Lewis’s book The Undoing Project, then you know about the work done by psychologists in the last 50 years to explain how bias interrupts the human mind’s ability to perceive information. Literally, our personal bias leads us to see things that simply are not there. We all have expectations, and these expectations are based upon hard won human experience—most of which has served us very well in life. But in the case of making HR judgments, or indeed any judgement requiring us to process large amounts of information, bias is quite detrimental.
In the questions/examples provided above, we see the opportunity for human bias to creep into common HR processes and potentially undermine them. First, let’s examine the interviewing process. As people, we may have expectations about how a qualified candidate dresses, how they speak, and which personality traits are most prominent in a good candidate. These are likely informed by our own experience, and colleagues who may have made a deep impression on us. Just as likely, information contradicting the same bias is dismissed. This means that our human minds are not able to process large amounts of information in a uniform and objective manner. When applied correctly, HR analytics can do this much better. For example, an HR analytics team would consider data collected during the evaluation phase and performance data for successful applicants; in other words, before and after hire. Hopefully, many applicants become very successful at your firm, but you also know that many do not. We can apply a label certain to each candidate profile, recognizing that the candidate either was or was not successful. We can then train our analytics algorithms to learn what a successful employee will look like, mathematically, at hire time and reduce our human bias. Bear in mind that bias can still creep into the process, if interviewers fail to recognize the need for standardization and quantification.
Similarly, as it relates to evaluating training against performance, we see an opportunity for bias to lead to conclusions that are false, or at least for which there is no evidence. Business leaders can (and should) demand this evidence from HR, so that they know that capital is being deployed correctly in support of the firm’s financial well-being. To be clear, it can be very difficult to prove causation between training and financial ratios (i.e., that training causes an increase in Net Income). However, HR should be able to provide evidence demonstrating correlation between employees who perform well on the job (be that metric in sales figures or on-time delivery) and those who attend certain training activities. When HR provides evidence of this correlation, it becomes a strategic partner with business leaders, helping them see and understand the patterns in human behavior.
See Differently, Know the Facts
Analytics offers HR professionals an opportunity to approach decision making differently. Measurements and quantification of candidate and employee characteristics and performance can provide evidence of correlation between the policies HR is supporting and the outcomes the business seeks to drive. By thinking differently about HR, we can reduce our propensity to see things that are not there, replacing that vision with a clear eyed, scientific, data-driven approach.
Want to learn more about the world of HR Analytics? We are speaking at this year’s CBIA Human Resources Conference on the topic. We hope to see you there!
About Brian: Brian Berry leads the Microsoft Business Intelligence and Data Analytics practice at BlumShapiro. He has over 15 years of experience with information technology (IT), software design and consulting. Brian specializes in identifying business intelligence (BI) and data management solutions for upper mid-market manufacturing, distribution and retail firms in New England. He focuses on technologies which drive value in analytics: data integration, self-service BI, cloud computing and predictive analytics.