There was a time when data-based predictions powered decision-making for marketers and financial analysts on the back of detailed insights, be it regarding demand forecasting or predicting defaulters in loan repayments.
Today, Predictive Analytics (PA) is becoming a trend. In HR it means a technology which learns and draws insights from existing data and uses it to map future trends and behavior to enable effective and efficient HR management and decision-making.
By identifying relevant trends and patterns, PA empowers organisations to understand what they do not yet know and gain insights on various aspects of management. The PA-ledhuman resource management can, thus,enable HR managers to make better strategic decisions while dealing with various incumbent challenges faced by the workforce.
The importance of this technology can be illustrated through some HR problems which managers face daily. One of them is that of absenteeism and typically costs organisations a huge amount of money. Identifying excessive absenteeism, diagnosing the reasons, and predicting the same for the future becomes critical.
Another example is the process of analytics-driven recruitment. Businesses across industries look not only for talented, qualified, and experienced candidates but also those who will make a great culture fit in their organisation. The PA can help determine which candidates are a better fit along with delineating various factors involved in such decision-making.
Predictive social media analytics and sentiment analytics are some of the solutions that can look at non-numeric data, such as feedback surveys, social media content and others to gauge the engagement level of employees.Leveraging the insights collected from these systems, organisations can come up with relevant strategies to bolster employee engagement while delivering superior employee experience.
All organisations incur recruitment cost and it goes waste when the selected candidate does not join the workforce. In addition to losing money, it can also negatively affect an employer's brand. The PA will assess historical data to unearth possible renege and help managers take the necessary decisions. It also helps in predicting the attrition before it happens. It will also look at an organisation's data to identify factors that have the most significant influence on an employee's flight risk. Managers can then use this data and address these issues before they cause an employee to leave. Google is one of the first companies to effectively utilise predictive analytics in its hiring process.
The above examples of predictive analytics are not limited to predicting the future trends but also predicting the category — whether the candidate is suitable to be hired or not. The PAs is used to classify the performance of employees as high, medium and low. HR managers analyse a large number of variables and with the help of technology are able to predict the performance of employees. This way HR managers can use statistics and technology to make better decisions. The PA where classification is done, may use past known outcome and derive rules for prediction or group the employees with similar behavior when the outcome is unknown.
Thus,HR managers can classify the performance of employees in three categories: high medium and low performing, without resorting to the traditional normal curve.
The companies who have started using PA in HR use one or more statistical procedures like Regression-Linear and logistics, Decision tree, Random Forest, Cluster Analysis, Neural nets etc.
In view of this, there is no doubt that analytics can be a powerful data-driven tool for organisations to anticipate performance outcomes and develop pro-active strategies in Human Resource Management.
The writer is Sangeeta Shah Bharadwaj, Professor, Information Management Area, MDI Gurgaon