The Conundrum of Talent Management Analytics, BIG Data and IoT


Few years back, when I was working for a Retail Organization, I was the functional lead for an HR analytics project using Business Intelligence Data Warehousing. Retail store analytics was of high interest to business leaders in a burgeoning and extremely competitive organized retail market.

We developed people analytics capable of reporting across several dimensions and at the most granular level. For example, while the attrition report shared the overall attrition number, one could navigate further.

Considering the overall store associates attrition was 25%, user could navigate further for stores say in Maharashtra at 40%, within Maharashtra store formats where you see highest for specialty stores at 55%, and within these stores for specific store at 70% and in this store for associates with demographics of tenure less than 6 months and gender and so on. It was faster than the time it took to read this paragraph with few clicks and available every day. This did support a lot for understanding real issues and action planning at specific store(s) level.

However, when I look back at the analytics capability, what I find missing is the market insight and linkage to business. What if we compared with market attrition for similar roles in similar stores in same geography, what if we could say higher attrition lead to reduced throughput, low off-take of new launches, private labels, Up and cross selling, number of loyal customers, customer satisfaction and so on leading to a 9-10% loss of store revenue and translate into a meaningful number in terms of lost revenue opportunity.

Sharing numbers without linking to business does not make much sense for business and business leaders connect better to numbers which in turn builds respect and credibility for HR professionals.

HR Analytics and BIG Data

Over the past few years, BIG data analytics has been revolutionizing the way companies do business with real-time, forward-looking and integrated analytics.  HR as well has progressed to deploy predictive talent models that can more effectively and more rapidly—identify, recruit, develop, and retain the right people. Today, we have organizations using people analytics across the spectrum from Reactive (what happened) to Predictive (what would happen next) to Prescriptive (what’s the next best action) aggregating data from many sources.

The desire to quantify, measure, and monitor ourselves has never been more than it is today. Reason why BIG is in italics is there is already lot of data available, may be not clean enough or residing in different systems, and we need integration of data, automation of processes while aspiring for forward looking insights. What I have experienced, the challenge in HR is to use this data and specifically ‘quality data’ for sharing insights and influencing and linking back to business. Data is required to shift the dial on perception that HR is not detailing the return on HR-related investments. Once you understand the business and the industry you operate, the competition, key growth drivers, the emerging market trends, link people actions supported by data to these insights.

How effectively are we using the data we have?

Would take two prosaic examples on usage of data to augment effectiveness of existing HR processes:

Talent Management

  • Which functions and roles you need to build talent for today and tomorrow for our business in context of a changing market and growth plans?
  • How is the organization talent pool changing every year- people moving in/out vs staying same?
  • If people identified as talent are changing frequently how do you continue andmeasure their development efforts and career moves?
  • If people were identified as top talent ready for next move and staying too long in same role, how do you continue to engage and retain them?
  • If you have more talent in non-business critical roles while the requirement is in critical roles, are these talents willing for moves outside their roles, how can you develop their skills?
  • What is the depth of talent pipeline for identified critical roles, how is that developing?

And how is the talent management process delivering results for the organization in terms of business growth and improved revenues, profitability, productivity and customer satisfaction. To answer the above questions, most data we need is already available, question is how are we looking at it and using it.

Certainly, BIG data can further augment creating talent profiles at risk with demographic profile, professional and educational background, performance ratings, levels of compensation, longer travel periods, missed weekends for meetings, changes in family status etc. These insights can support more customized approach for talent development.

Development Programs

  • Is there any improvement in efficiencies of equipment’s or processes with critical mass of trained people?
  • After training teams on a program say Negotiation Skills, what changed, did the teams improve in customer relationships in terms of securing better margins, improved credit periods, preference in terms of in-store displays etc.
  • If you are conducting assessments, how are you establishing linkage of assessment scores with development programs and key business metrics
  • What’s the impact on performance and career movements and retention of people joining these programs?

HR Analytics and IoT

With all approbation, IoT continues to evolve and transform from industrial applications with autonomous machines to a human-centric category with personal augmentation and “living services” that let people program and connect smart devices the way they want. IoT feeds on data and can generate an unprecedented amount of data associated with machines and people and how jobs are performed on a real-time basis.

IoT can pave the way for HR to perform its business by bringing in a broader set of verifiable metrics & data to measure the previously unquantifiable or invisible. A case for reference here is Deloitte Canada who redesigned their work place using the data with a set of volunteers who wear sociometric badges—measuring location, voice, and movement—to assess which aspects of work were positive and negative.

The need would be higher with the changing workforce demography and more Millennials and Gen Z entering the workforce who are well connected as consumers and would want to have the same experience at workplace, insights of which can be gathered with IoT. The watch out would be are you sweating your existing data enough to avoid drinking from another firehouse of data through IoT.

While I personally have a high proclivity for data based decision making, I do not leave everything about human behaviors to be understood with data alone. Numbers are important, in my experience the maximum insights I have got is by talking to, listening and observing people. So am careful about conclusions derived from numbers alone and use them along with my insights. Let not data become like giving a child a hammer and he would treat everything as if it were a nail.

To summarize,

  1. Sweat the data you have– Look at ‘More’ data and ‘More’ often.
  2. Link to Business to drive buy-in using data and insights and create value for the business.
  3. Change less by changing constantly in your journey of HR analytics and adopting emerging trends and continue to listen and observe.

AuthorAmit Singhal is currently working as Director Capability Development at Coca-Cola Amatil, Indonesia and prior to this he has served as Associate Vice President HR at Hindustan Coca-Cola Beverages Pvt. Ltd. Amit is an accomplished business HR professional with more than 15 years experience in leading culture and change transformation across businesses in major business cycles in leadership roles. He is an alumnus of Symbiosis Institute of Business Management, Pune and started his career with Tata Motors, then worked with Reliance Retail since start-up stage and last 9+ years with the Coca-Cola system.


Please enter your comment!
Please enter your name here

This site uses Akismet to reduce spam. Learn how your comment data is processed.