Using Workforce Analytics in a Talent Shortage

Using Workforce Analytics in a Talent Shortage

If you have been following any of the news around the workforce shortage epidemic, we know that we are in a period where talent is hard to find, and even harder to hold on to.  In nearly every industry, there are challenges in finding people with the right skills.  After a period of investment to build a candidate up, they leave to capitalize on their new abilities in a market hungry to take them.  It is becoming more and more competitive for companies to retain their skilled staff.

Many organizations have HR data that spans employee transactions, benefits, performance management, compensation, surveys, and learning.  These data assets are rarely put to good use in addressing the problem of employee retention.  This valuable data can be combined and used to make workforce analytics a key part of employee retention and growth.  Companies that are using data driven insights to make decisions around employees will have a key competitive advantage.

Combining advanced analytics, information discovery, data warehousing and business intelligence disciplines with the employee data assets, new actionable insights can be exploited.  For example, a key objective in a talent shortage would be to predict which employees are likely to leave, and then take action before they start to look for a new job.  If employers knew who was likely to leave, it would greatly improve the company’s ability to retain the investments made to develop these employees and reduce all the costs associated with replacing and upskilling a new employee.

Predicting Voluntary Turnover

Predictive modelling is the tool of choice when looking to identify which employees are likely to leave.  Predictive modelling is an area of data mining using statistical models to predict future probabilities and trends.  A predictive model uses predictors or factors that influence the future outcome.  For employee retention, those predictors can include employee performance ratings, commuting distance, geography, supervisor ratings, survey sentiment, years of service, age, and many other possibilities.

There is no one-size-fits-all when it comes to developing a predictive model for a given organization.  Considerations on the type of industry, employee factors, and geography play a big role in identifying the right predictors for identifying employees who are likely to leave.   People are not robots and the predictive model needs to take those factors into account.  Finding the right predictors, does require some skill and experience.

Running?  But we barely crawl!

Crawl, Walk, Run is a common idiom around going through a stepwise progression.  A couple of clients have said that they have trouble reporting on the basics around HR transactions and moving into prediction is a step to far.  Those same clients have mentioned that the consumers of this information would have trouble making action out of such advanced techniques.

Have no fear!  Although the techniques are advanced, it doesn’t have to be difficult to use and make proactive decisions to retain valued employees.  Once the predictive analytics is in production and in operation, all managers need to see for each employee could be as simple as (Low, Medium, High).

Probability of Voluntary Termination:

As you can see, even though the techniques to get the predictive result, could be more advanced than your current processes, using the results is really simple.  This simplicity, can make it easy for any manager to change the competitive struggle for a talent shortage.


Write a Comment


Contact Us Close