5 HR Analytics Questions every HR Business Partner Should Know the Answer to - Analytics in HR

5 HR Analytics Questions every HR Business Partner Should Know the Answer to

You’re a business partner and there’s been quite a buzz about HR analytics. Strap in, here are the five HR analytics questions you should know the...

You’re a business partner and there’s been quite a buzz about HR analytics. Strap in, here are the five HR analytics questions you should know the answer to. We start off generic and make it progressively more specific for you, as a business partner, towards the end. 

 

What is HR analytics?

HR analytics is a contraction of Human Resources and Analytics.

Traditionally, HR departments have been data averse. HR professionals got into HR because they wanted to work with people – not numbers. Reducing people to numbers just feels… wrong.

In reality however, HR analytics is not about reducing people to numbers. It is all about supporting HR decision making – just like talking to people, in the end, helps to make better personnel choices. To read more, read this post, titled: What is HR analytics?

 

What’s the deal with predictive analytics?

Predictive analytics is probably the most famous part of HR analytics. Using the right data, an HR analyst can:

  • Predict who will leave the company
  • Make a salary estimation for new functions – just by looking at the job description
  • Predict the performance of new hires

And much more.

These examples are all very exciting but it’s not what analytics is really about. Predictive analytics is a tool in the HR analyst’s inventory.

 

What are the other tools of the HR analyst?

The HR analyst is someone who is doing organizational development using primarily data.

The analyst first needs to talk to the business to find out what the real problems in the company are. This is a consulting skill set.

Second, the analyst needs to talk to the employees to really define the issue. Senior management can think one thing, but employees often have a very different perspective. Combining these different perspectives is important, because it will give clues as to what the relevant variables are that should be included in the analysis.

Third, the analyst will take a look at the available data. Maybe all the data is available and the analyst can start immediately. Oftentimes, analysts need to generate data themselves by creating questionnaires or establishing other measurement methods.

When the data is available, it needs to be cleaned and filtered. Is Jake, who is a “Sr. Manager” in the same role as Jill, who is a “Senior Manager”, or Bill, who is “Sr Manager”? These little differences in data prevent analysts from putting the same functions on one large pile for analysis. This process is called data cleaning.

After cleaning the data, the analyst can do the actual statistical analysis. This can be a predictive analysis, or a simpler correlation analysis. There are multiple kinds of analysis and depending on the problem the analyst will deploy the one that suits the situation best.

After the results come in, they need to be presented to management in order to trigger action. Analytics without action is called analysis paralysis. This is something you want to prevent.

Often, the results show things that are hard to explain. When this happens, the analyst needs to go back to the employees that were analyzed in order to get an accurate picture.

For example, the analyst finds that people who are redeployed in department X are likely to leave within the first 6 months. However, the data might not say why. Is it because of an unhealthy culture, the high work pressure or because the senior manager is a bully? This kind of data is oftentimes not measured and the only way you can find out is by talking to people.

Click here to continue reading Erik van Vulpen’s article on DigitalHRTech.

 


Join the Conversation