Workforce Analytics: Much Ado About Nothing… or Something?HR Operations & Analytics
Is there substance to this concept?
When faced with a major investment decision, how many organizations would bet their success on a gut feeling? How many would rely on intuition when taking a new product to market? Yet, when it comes to the workforce – one of an organization’s most expensive and valuable assets – too many executives rely on hunches without relevant data.
There is much information posted about Workforce Analytics and most of it is coming from vendors that sell large enterprise software packages who have much to gain when the concept gains momentum and runs rapid through your organization.
Workforce Analytics Defined:
Our team set out to learn a few things and to share what practical items we could find regarding the concept of Workforce Analytics. One key measure that was frequently referenced is the impact of turnover. This seemed to be a logical starting point for those pursuing Workforce Analytics initiatives. There are many other measures we discovered and will discuss later, but here are two key themes worth mentioning regarding Workforce Analytics initiatives.
1) Helps demystify some incorrect “gut” feelings
We have a few examples how when using your instinct or perception of what happens in your own organization can be misleading and if you compare yourself to others their reasoning may be completely different as well.
Let’s take turnover as a measure to see what some actual companies learned when they applied Workforce Analytics and actually aggregated and analyzed their own data.
Thrivent Financial in Minneapolis learned that their retention rates were higher for new hires in their Customer Service organization when the new hire came from another Customer Service job. Initially, the team felt they would have more success bringing people from other disciplines into Customer Service because they would have a fresh perspective and be able to excel in the job for several years. This was not the case when they analyzed the data. Retention rates were higher when new hires for Customer Service came from Customer Service disciplines.
Met Life discovered that employees within the company who had different job experiences across the organization had higher performance review scores and longer tenure. Initially, they believed an employee jumping from job-to-job was a sign someone was dissatisfied and searching for more. Prior to this analysis job changing was a flag for high probability to leave. After reviewing the actual data Met Life learned this was not the case. A variety of job experiences within the company resulted in better performance and longer tenure.
University of Southern California was worried and making plans for massive retirements because they have a high population of employees close to retirement age. Workforce Analytics helped them find out two things. First, non-tenured staff members are too young to start retiring. Second, tenured staff are approaching retirement but they’re not required to retire and most are not able to retire. They end up working longer into their 70s so aging is not yet an issue for this University and they were able to postpone some of their initiatives focused on planning for massive retirements.
2) Helps HR lead by example to create a data driven culture
The Human Resources team can lead by example to create a data driven decision-making culture. We have seen organizations use Data Visualization techniques to show top skills present for a given position that resulted in a hiring decision. We have also seen charts developed which map out hire and retention rates per posting source for the specific position. Additionally, some organizations calculate a retention score for each new hire when they first start their job. The score is based on demographics, job history, benefits history, and compensation history.
When HR comes to the table with real numbers from areas like pay, benefits, payroll taxes, and other expenses HR can more credibly argue its case to the CEO when forecasting workforce needs. This puts Finance and HR on the same ground in a world where the finance department owns much of a company’s data.
This type of analysis is also important because it can help identify processes and sources that are producing better results and these results can have an impact on the organization.
For example, the cost to replace employees is impactful.
Data Point 1 – Costs to replace an hourly employee – $5K – $20K
Data Point 2 – Cost to replace a salaried employee – 30-60% of the salary
It is important for HR to bring the overall cost and cost impact to the executive meetings, however the leading HR firms are taking it one step further and using analytics to build predictive relationships with the data to demonstrate baseline cost and then a proposal which identifies what can be done to improve the situation using data not “gut feel”.
A large Financial services firm was experiencing high turnover in its call center and retention was important to its service quality and cost as demonstrated above. This organization pulled information from its HRIS system to build predictive models to help understand the factors causing people to leave. The key drivers were identified as compensation, career development, experience levels, mobility, and work-life balance. One of the predictive equations that was developed form this analysis was as follows:
If an employee is promoted within the pat year (predictor X) he or she has a 30% lower probability of leaving within the next year (outcome Y). Remember this is not a gut feel statement this is something derived and supported from real data and the organization knows that it could cost close to 25% of the employees salary to replace them so looking at promotions and pay increases more aggressively was worth the investment and time to ensure quality service delivery at a lower cost.
Using Workforce Analytics takes the organization out of the guessing game and brings facts and data to the table to improve decisions and take actions to prevent costs from increasing. As we learned in the last example, I guess quality can be free. Maybe the 80s guru Deming knew what he was talking about.