Analytics Is The Limitless Key To HR Department Functionality

With analytics we can time travel for #NoHumanIsLimited

Today, most of the world citizens believe that there is #NoHumanIsLimited after the profound INEOS159. I am a subscriber to the same with a stronger belief that Analytics is the limitless key to HR Department functionality.

There are four basic categories of Analytics:
 


D1. Levels of analytics

NB: We shall focus on turnover to understand the levels of analytics

It is based on historical records and describes what happened. It is the most basic approach of analytics because historical data is always available.
 


Graph 1. Employee Turnover 2015-2018

Diagnostic analytics examine contents or data to answer, "why did it happen?" You have to delve into the content or data to understand the reasons why. At this level, the analysis of data is informed by the urge to know why.
 


Graph 2. Employee Turnover                 Graph 3. Turnover - Resignation

From graph number 2, it is clear that more employees are leaving on resignation as compared to gross misconduct and terminations. On further analysis of departures (graph number 3), it is clear that most employees that resign get absorbed by the competitor. Do we confront the competitor for snatching our employees? Do we write a policy restraining employees not to join the competitor? However, the Future HR have other smarter options for addressing the issue!

Our focus is to understand why employees resign to join our competitors. The reasons are attributable to the differentiating factors between a company and its competitors — for instance, factors such as working hours, medical cover, remuneration etc.
Remember, our aim is to reduce turnover (increased retention).

This form of analytics helps to predict unknown future events. Unlike diagnostic analytics that rely on historical data, predictive analytics looks into what is likely to happen using the available information. 

You can build regression equation for the resignation on different distinguishing factors between your brand and the competitor as below.
Resignation = 2 - 0.2*remuneration + 0.01*working hours – 0.03*medical cover – 0.003*Training
You can change the factors i.e., remuneration to see the impact on the resignations. With this, you are likely to know what is likely to happen.
Alternatively, you may use probability techniques to predict the future.

Easy or Risky? From the discussion between a CFO & CEO, "What if we train them & they leave, what if we don’t train them & they stay." I am still wondering where is the CHRO. Obviously, not in attendance at the meeting with the CEO. Now with ready information about the future, the CHRO can join.

Prescriptive analytics involves evaluation & selection of the best action to be taken to solve our resignation problem depending on our understanding of the predictive analysis of the situation e.g.
 

i. Increase employees’ salaries by 5% in 2019
ii. Introduce medical cover in 2019
iii. Allow flexible working hours
iv. Initiate on the job training for employees

I firmly believe that with analytics, we can time travel for #NoHumanIsLimited.