Article: People Analytics: Hype vs Truth

HR Analytics

People Analytics: Hype vs Truth

The hype vs. truth spectrum ranges from advocacy for splitting HR into two functions of people analytics and people administration to uberization of HR in the future
People Analytics: Hype vs Truth
 

HR function’s ability to use Big Data for HR will be dependent extracting ‘Value’ (V) from ‘Variety’ (V) of HR Big Data

 

There is a lot of hype around People Analytics and in any tide of hype, a lot of ‘meaningless’ boats get prominence while large ‘meaningful’ ones remain in the backdrop. This analogy aptly captures the current picture in the world of people analytics. This hype vs. truth spectrum ranges from advocacy (Bernard Marr) for splitting HR into two functions of people analytics and people administration to uberization of HR in the future. In between, there are lot of prophecies such as ‘People Analytics is a new fad marketed by consultants’ or ‘People Analytics is the way forward for making HR accountable like other business functions.’ 

Elsewhere David Green in a recent post wrote that the HR has suddenly discovered a gold pot in people analytics and every CHRO is running after it! 

Here is my take on hype vs. truth in people analytics:

HYPE (H): Data has value and is like an asset which can be monetized for powerful insights having impact on business outcomes. Hence data is has to be treated as ‘Strategic’ in nature.

TRUTH (T): There is no doubt that data analysis can provide powerful insights but truth is that organizations treat data as ‘Transactional’, making it unfit for any kind of analysis and creating a familiar situation of ‘garbage in garbage out’.

(H): People Analytics can help in identifying drivers of Organizational Performance, making it easy for the HR to align HR Initiatives and Programs with those drivers to maximize organizational performance.

(T): The reality is that we know a lot about individual level drivers of performance like pay; training; rewards; communication etc. but how these individual drivers of performance are connected to organizational performance is not clear as all individual performance gets aggregated into a project and several projects add up into a function or unit or account and thereafter light of contribution gets lost. So we don’t know Organizational Level Individual Performance Drivers.

(H): Big Data in HR will help HR in becoming a truly ‘Strategic Function’ in coming years.

(T): HR function’s ability to use Big Data for HR will be dependent extracting ‘Value’ (V) from ‘Variety’ (V) of HR Big Data. These 2 Vs (of 4 Vs of Big Data) are central to success of People Analytics. Analyzing, Interpreting and making sense of data is not easy. 

(H): Buying a high end Technology solution marks the successful initiation of People Analytics function in the organization. 

(T): Technology at best can be an enabler of People Analytics. There is whole lot of ground work which needs to be done within HR function/team and organization to harvest the benefits of People Analytics. One of the major reasons why People Analytics doesn’t take off in many organizations is simply buying a highly marketed technology solution and then assuming organization is ready for People Analytics. And also technology is expensive and when doesn’t give promised results, disillusionment sets in. 

(H): Large data sets based on years of employee data are needed for high quality insights.

(T): No doubt that large data are needed but tiny data sets and anecdotal evidence can’t be ignored and can be source of powerful insights. Water cooler or corridor chats or food court murmurs – all can be equally good data points for insights on effectiveness and impact of HR programs and policies.

(H): Models and Algorithms bring perfection and rigor to the practice of People Analytics by bringing out patterns from large amounts of employee data which is impossible for an average person to visualize.

(T): It is well known that even the Best Model and Algorithm with the best data can be imperfect as there are human limitations to include everything while creating a perfect model or an algorithm for capturing everything of a ‘whole man’ that matters and may get missed out in a model or algorithm and also a no perfect algorithm can give a perfect outcome.

(H): People Analytics will eliminate ‘human biases’ as everything becomes number based and numbers speak for themselves.

(T): People Analytics has its own biases because it measures what can be measured. However, people are people, and so far there is no exact science to exactly measure and predict what and why people do by uncovering all the biases and motives.

(H): Job performance data can be converted into objective measures making measurement simple and easy to capture each employee’s contribution to the job and organization.

(T): Over-objectification or over-simplification of the measurement of performance, creates the risk of  missing the richness of what makes that job special—or complex—or what makes each person’s contribution unique. For example, an HR Manager spending time on a difficult employee or client facing manager soothing down an angry customer, all are complex tasks not amenable to easy measurement.

(H): People Analytics will help in eliminating errors in decisions related to people management matters like hiring, promotion, potential, rewarding, deployment etc.

(T): It is true that errors will get reduced but we run the risk of facing unintended consequence of committing ‘fundamental analytic error’. In social psychology, we often commit fundamental attribution error, meaning we attribute causes to intrinsic factors (within individual) rather than to extrinsic factors (context or situation). Hence People Analytics runs the risk of unfairly making individuals scapegoat based on data, who are not performing well while ignoring the weaknesses and constraints of the system where individual is working. Truth is that system is always powerful than an individual and most of the time overpowers even the best talent in the world. 

(H): People with HR domain background are best suited for HR Analytics.

(T): Deep HR domain knowledge surely adds value in making sense of data patterns from ‘people’ perspective but a broad based People Analytics team comprising people with knowledge and background ranging from business, financial, statistics (data scientist), sociology, psychology, philosophy; IT, and data specialist makes a deadly combination of a dream team for People Analytics.

(H): HR function moving from charting, dash-boarding, reporting to predictive analytics signals the completion of installing People Analytics in HR function.

(T): Doing the analytics piece is half part of the People Analytics equation. True value and proof of pudding lies in converting People Analytics Insights into meaningful actions for Business outcomes, which is often tough.

Hype is vicious. Though it has generated lot of interest and likely to make People Analytics a Multi-Billion business and fuel demand for People Analytics professionals in the coming years yet it is possible that 20 – 30 years down the line all this hype around People Analytics may sound rubbish since the HR challenges being addressed are the same as 20 or 30 or more years back (who will be a best hire Or who is/will be the best performer). Whether People Analytics is Hype or Truth, jury is still out! 

Read full story

Topics: HR Analytics, HR Technology

Did you find this story helpful?

Author

QUICK POLL

How do you envision AI transforming your work?