Employee Turnover: Solve It Using 4 Levels of Analytics
Every organization has a problem they want to solve. It could be a small problem or a large one. The key to solving problems is:
- Gathering the right information
- Having a good approach
Often, organizations approach problem solving from the standpoint of “Oh, I know what the problem is…” and ultimately, they end up only addressing a symptom and not the real issue. I’m sure you’ve seen it too. In thinking about analytics, I could see how using the four levels of analytics could be a good approach for problem solving.
Here’s an overview of the four analytics levels and how it could be applied to address a challenge many organizations face: employee turnover.
DESCRIPTIVE analytics is focused on what’s happened. The first step in solving most problems is figuring out what’s going on. That’s descriptive analytics. The organization wants to collect information and data on the who, what, when, and where. The good news is with technology, you might be able to gather this information easily.
An example might be reviewing exit interview data to see if there are any trends for why employees are leaving, when they tender their resignation, and what positions / departments the employees work in.
DIAGNOSTIC analytics answer the question “why did it happen?”. Using the information from step one (descriptive analytics), the organization might be able to reach some conclusions about the situation. A technology solution might provide some assistance is filtering the data in different ways to explore new causes and correlations.
For instance, the organization might discover that managers tend to resign in March or April (right after they receive their annual bonus). Or a large number of new hires are leaving the accounting department within their introductory period. (Not making any judgements about accounting departments here. It’s just an example.)
PREDICTIVE analytics is about what will happen. I like to think of this step as “How will the situation play out?” Meaning if the contributing factors don’t change, what will the situation look like. Or if only one factor is changed how will that impact the outcome.
Companies can’t and probably don’t want to make lots of changes at one time. They want to implement change in a logical manner and use their resources wisely. It also helps individuals accept and embrace change.
PRESCRIPTIVE analytics points toward “What should we do?”. In the predictive analytics stage, the organization considers all of the options. In the prescriptive analytics phase, decisions are made.
Using our turnover and employee retention situation, the organization might decide to focus on stay interviews. Managers can ask a stay interview question during regular one-on-one meetings. HR might add a stay interview question to new hire onboarding check-ins and regular pulse surveys.
Organizations can use the principles of analytics to help them identify problems and brainstorm solutions using data as the driving factor. This is definitely one of those moments when we can put the power of technology to excellent use.
P.S. If you’re looking for more information about analytics, check out “Predictive Analytics for Human Resources”. I’m a big fan of Jac Fitz-enz and his work. You know that I recommend his book “How to Measure Human Resources Management” to everyone. His predictive analytics book is equally excellent.
Image captured by Sharlyn Lauby after speaking at the HR Change & Transformation Conference in London, England
21