How predictive analytics can help prevent employee attrition
How long do you plan to stay in your current job?
It's estimated that people in the United States have an average of 12 different employers throughout their career. As companies battle to attract top talent, learning how to keep employees happy is key for any business wanting to prevent attrition.
Why is employee attrition bad?
With the recent popularity increase of freelancers and changing landscape of the average American workplace, some might wonder if losing an employee is “as bad” as it once was. As most CFO’s would readily tell you, employee attrition is still very costly. It usually involves:
• high financial burden
• lost employee knowledge
• expensive hiring process
• lengthy training of a replacement employee
• new hire time to effectiveness and their understanding of internal company procedures
• low morale for the remaining team members
These all lead to considerable impact on business continuity and often more profound impacts on profitability.
So, what can companies do to understand employee behavior and prevent attrition?
One option is to create a predictive HR analytics dashboard.
Where does dashboard data come from?
In the context of a dashboard about employees, data could come from:
• Employee timesheets (clocking in and out, PTO records, etc.)
• Company-wide surveys
• Virtual assistant messaging history
• Performance records
• Sales figures
• Quarterly or annual reviews
• Content management systems
Predictive analytics and employee attrition
When employee information is brought together holistically into a business dashboard, predictive analytics can be applied. Using machine learning algorithms, data can be sorted and displayed to show estimated behavior and/or figures. KPI cards can emphasize essential information.
Employees leave for many reasons, and it’s important to follow-up on dashboard data with a human conversation. Whether the person is unhappy with something about work or something outside of work is important for creating a remediation plan.
Let’s look at an example.
This business dashboard focuses on the staff of the fictional Acme Corporation.
This dashboard shows us a broad overview of employee attrition at Acme. The KPI’s across the top portion highlight the expenses associated, as well as current attrition rate and average risk.
The remaining charts and graphs give us a more detailed look at which departments are in trouble, reasons people choose to leave, and which employees are at especially high risk.
At Acme, it seems that the operations department is struggling. We can see that managers and senior managers have especially high scores for attrition risk, and—based on the dot graph—that there is one individual who is very likely to leave and expensive to replace.
As we hover over this dot, the dashboard changes and we can see some new information.
In the pop-up that appears, we see:
• The employee’s name and designated number (John P., 3542)
• The action required (Email John’s supervisor to schedule meeting, advise immediate PTO)
• Necessary information for fulfilling the action (Supervisor’s name and email address)
• Probability of preventing attrition
We can also see John’s information highlighted in the same gold color in the “Employees at risk for leaving” section.
Using all this information, it’s clear what next steps are necessary. HR can schedule a meeting with John and his supervisor to understand the situation and create a plan.
This is just one example of how analytics can be used to predict employee behavior and alert the right people to take preventative action. This type of dashboard can also be used in other industries, such as education.
Interested in predictive analytics for your organization?
Check out how Logic20/20 can help.
Nick Kelly is the Director of Visual Analytics at Logic20/20. He is a hands-on leader in analytics with over 16 years of international experience in analytics and software development, deployment, adoption, and user experience.