Addressing attrition through analytics
For companies of all sizes, attrition has come to represent one of the most important risks to productivity. The cost of losing employees is rising in many industries, owing to a paucity of suitable replacements and the increasingly collaborative nature of many jobs. In this context, having mature processes in place to understand, anticipate and prevent attrition is all the more important.
Although it is one of the most pressing issues facing many companies today, attrition is a very difficult problem to solve. Employees, after all, are human beings: they juggle work, family, aspirations, emotional and social considerations and many more things when they decide to stay or leave a company. However, even if they can’t prevent it, companies can prepare themselves for the eventuality of employees leaving and mitigate the risk it poses.
Here are some considerations to keep in mind in order to be better prepared to meet the challenge:
1. Get visibility of all aspects of the employee life-cycle
Employees generate huge amounts of data during their time with companies, not all of which is recorded on a typical HRMS: while HR-related information (payroll data, attendance data, training data etc.) is easy to retrieve, it doesn’t cover all aspects of an employee’s time with the company. Operational and project management-related data can be just as useful in understanding what employees are working on, what kind of hours they have, or whether they enjoy working with their team, all of which can influence an employee’s decision to leave. Companies can also get data on how employees use other benefits like health insurance, which can also lead to an understanding of the employee’s life events. Integrating data from diverse data sources comes with some complexity: applications can have very different data architectures and standardizing across multiple databases can be a difficult and time-consuming process. Companies need to set up extensive and complete ETL processes in order to integrate all this data before any analysis.
2. Establish processes for aggregation and anonymization
In any process involving employee data, it is essential to ensure compliance to privacy laws. The downside of having a holistic view of employees’ career paths, as described above, is that such analyses can risk violating employees’ right to privacy. To mitigate this, companies need to establish rigorous policies and processes to aggregate and anonymize data. Aggregating data removes the risk of analysts identifying individuals through their behavior, instead providing information about trends among groups of people. In addition to this, policies need to be put into place to anonymize employee data wherever necessary, including textual references to individuals in texts, e-mails or web forms. This can range from a simple substitution of employee names with other identifiers, to more complex cryptographic anonymization techniques.
3. Understand the true impact of attrition
Resuming attrition as a simple percentage can hide its true impact. The role of every employee is unique and their departure can have different kinds of effects on the organization. HR professionals need to design better metrics to understand the impact of attrition on the company. For example, it might be more useful to analyse employees’ contribution to the company’s productivity to understand which employees are most critical to the organization. Network graphs can also be used to understand which employees in the organization serve as critical ‘communications hubs’ or repositories of knowledge in the company’s ecosystem. By going deeper into attrition data, HR professionals may be able to gain better insight into the impact of attrition, which can be masked by using shallow or incorrect metrics for measuring it.
4. Create actionable insights
While post facto analyses of attrition (including processes like exit interviews) do play a role in highlighting areas of improvement in a company, they do not provide sufficient information or insight to address the problem of attrition sufficiently. In order to make a meaningful difference, companies must be able to anticipate conditions that make employees more likely to leave. Key performance indicators (KPIs) are metrics which can be used to track various aspects of an employee’s personal and work life and can give HR professionals an early warning about employees likely to leave. KPIs can range from simple metrics like the average number of hours worked, or average time off during a year, to more complex KPIs which can track the employee’s engagement in the company, or their sentiments about their workplace.
As the amount of data companies collect and store about their employees continues to increase, it is becoming a strategic imperative for organizations to leverage that data to understand employee behavior and be forewarned about possible adverse outcomes. In a world where businesses increasingly rely on their employees’ skills and expertise to increase productivity, ensuring workforce continuity can become a vital competitive advantage.