Predictive Analytics in Human Resource Management

A Hands-on Approach

Shivinder Nijjer, Sahil Raj

• Publisher: Routledge India
• Year: 2020
• ISBN:
  9781000208139
• Edition: 1
• Topic: Business
• Subtopic: Human Resource Management
• Index: Business

This volume is a step-by-step guide to implementing predictive data analytics in human resource management (HRM). It demonstrates how to apply and predict various HR outcomes which have an organisational impact, to aid in strategising and better decision-making.

The book:

• Presents key concepts and expands on the need and role of HR analytics in business management.
• Utilises popular analytical tools like artificial neural networks (ANNs) and K-nearest neighbour (KNN) to provide practical demonstrations through R scripts for predicting turnover and applicant screening.
• Discusses real-world corporate examples and employee data collected first-hand by the authors.
• Includes individual chapter exercises and case studies for students and teachers.

Comprehensive and accessible, this guide will be useful for students, teachers, and researchers of data analytics, Big Data, human resource management, statistics, and economics. It will also be of interest to readers interested in learning more about statistics or programming.

1
Analytics in HRM

After reading this chapter, users will be able to understand the following key concepts:

• Understand the meaning of business analytics and its need and value in business management
• Understand the meaning, scope, and context of human resource management
• Understand the meaning, need, and role of HR analytics in business management
• Understand adoption of systems approach/process view for application of analytics
• Understand steps in the application of analytics to a business problem
• Identify business problems through an illustrative example of the issue of turnover and retention management in the Indian IT industry

Opening case

Human resource management (HRM) has always been concerned with getting the best out of people, empowering and motivating them, and driving workplace performance. But the question has always been how to empower and motivate them, or make them perform well. Leading organisations today are adopting human resource (HR) analytics – a set of sophisticated analytical tools and techniques – to find an optimal answer to this question. One such business firm, Harrah’s Entertainment, an American firm functional in gaming hotels and casinos, applies analytics to understand how HR-related decisions and outcomes translate to customer services and therefore revenue generation. It uses analytics to determine an optimal number of staff required at the front end and back desk operations. Further, through HR analytics, the company understands what organisational factors and benefits make healthy and happy employees, since it directly has an effect on employee behaviour with the hotel guests and visitors. To improve the health and wellness of the employees, Harrah’s frequently conducts wellness programmes and then uses metrics to evaluate the impact of these programmes on employee engagement. Analytics assesses how the improvements in employee engagement metrics correspond to an increase in firm revenue. Similarly, other major companies value employee engagement and use analytics to understand how it contributes to firm revenue. Best Buy (Davenport, Harris, and Shapiro, 2010), a retail chain, has also deployed analytics in this context and shown that a 0.1% increase in employee engagement levels leads to a US$100,000 increase in revenue of that store. Analytics has found widespread adoption in many other areas of HRM as well, for example, to predict employee churn, or to understand information flow across the organisation. All these insights aid the management to curb the problem areas proactively. For example, Sysco (2019), another American multinational firm, involved in marketing and distributing food products, kitchenware, etc., has been able to target immediate retention strategies for its delivery associates using analytics and bring an improvement from 65% to 85% in their retention rates. This directly corresponds to savings in hiring and training costs for the firm. However, although the application of analytics can impart unprecedented insights for effective decision-making, the success of its implementation depends on a number of factors such as availability of data, leadership, and commitment of the management and so on. The case, therefore, highlights how HR analytics can aid in decision-making and provide insights into the relation between HR programmes and revenue generation. This chapter builds an understanding of business analytics in general, and the need for HR analytics and predictive analytics, among the readers of this book.

Introduction

As Alexis Fink, General Manager (Talent) says, “Data-driven HR is a mindset – a philosophy about how we make decisions” (Feffer, 2018).

For years, business firms have been making substantial investments in human resources through a variety of HR practices such as compensation, training and development, and career planning, to name a few. The evolution of HRM from the traditional personnel function, which was a standalone entity functioning in an organisation and treating employees like other resources of the firm, to the present role of being a vertically and horizontally aligned strategic business partner, is marvellous. The basic premise of this change is the need to address the dynamicity of the business environment, which demands an organisation to sustain its competitive advantage despite environmental uncertainties. In the present era, human resources are the only unique resources an organisation can have which guarantee a competitive edge since all the other resources – physical and organisational (resource-based view, RBV) – can be replicated in the firms. However, this also poses a consistent challenge on the firm’s ability to maintain its workforce, since competing firms are forever on a lookout to ‘poach’ away its competitive talent. Management consultants worldwide agree that when organisations look at their talent as a strategic business objective and work towards strategic integration of HRM in their business, organisational success will be automatically guaranteed.

In the 2017 Deloitte Human Capital Trends, we found that 39% of business people believe their company has ‘very good’ or ‘good’ quality data for people-related decision-making and 31% understand what ‘best-in-class’ people analytics looks like.

(Bersin, 2017)

Organisations today are moving towards a quantitative analytical approach to management, from the qualitative approach. In this context, HR practitioners are also adopting an analytical approach which is data-driven and fact-based approach to addressing business problems. People data is growing rapidly, and organisations are keen to develop and gain insights into people data. Resultantly, firms are eager to invest in the application of workforce analytics and the key for firms is to create an organisation that can routinely define hROI, human return on investment. Successful human resources executives in the 21st century prove their worth not by downsizing their own departments, but rather by telling a success story with each investment in human capital – and validating that story with numbers, or hROI. This quantification of HR data marks the trend towards the use of analytics in organisations for business management.

Business analytics refers to the use of statistical and analytical tools and techniques to interpret and explore business data (regardless of the data structure) to provide data-driven, factual business insights to management, to assist in decision-making.

Business analytics is often viewed as a subset of business intelligence (BI). BI is an umbrella term encompassing analytics, reporting, and database management, in addition to the applications, infrastructure, tools, and best practices used to gain information, improvise on its extraction, and aid in optimal organisational decision-making. The general term analytics comprises of both business analytics and data analytics, with the difference stemming from the data used to gain insights and the relevance and context of insight for a business situation. Nelson (2017) defines analytics as “the scientific process or discipline of fact-based problem-solving”. Davenport and Harris (2007, p. 7) define analytics as “extensive use of data, statistical and quantitative analysis, exploratory and predictive models, and fact-based management to drive decisions and actions”. Wilder and Ozgur (2015) define it as “the application of processes and techniques that transform raw data into meaningful information to improve decision making”. Finally, INFORMS recommends the definition: “Analytics is the scientific process of transforming data into insight for making better decisions” (Boyd, 2012). Although the term analytics was coined late in the 1990s, the concept of analytics has roots in application of traditional statistical techniques, and ever since the concept of quantification of work was proposed by Taylor (2006) in the scientific theory of management.

The widespread growth and adoption of business analytics in organisations can be attributed to a number of factors. The exponential rate at which data is growing is the primary reason for the parallel growth in business analytics applications. This is because firms have understood that these massive amounts of data contain much hidden information for the decision-makers, which only analytics solution can decipher. Further, the proliferation of a variety of data sources and the massive capabilities of the technological solutions to integrate and analyse such data has also prompted firms to tap into the utility of business analytics. The analytics products today are able to capture and analyse data from wide variety of sources, irrespective of the data structure, such as RFID (radio frequency identification) data, sensor data, social networking data, newsfeed data, data from web crawlers and weblogs, ad clicks, IoT (Internet of Things) data, and so on.

A number of factors contribute to the adoption of analytics in organisations such as rate of growth of data, the advent of new sources of data and technological solutions, techniques to capture data regardless of its structure, and so on.

Exhibit 1.1 Corporate examples of analytics on emerging data sources

Analytics applied to RFID data has seen numerous applications by retailers for on-shelf inventory management. This application of analytics provides a way of visualisation of product movement, sales traffic, and number of sales in real time.

Data from sensors are widely used for predictive maintenance in manufacturing firms, tracking customer behaviour, and detection of anomalies. For example, Reliance Power uses predictive analytics applied to sensors placed on power generation machines to determine their condition and whether they need maintenance services.

Insurance companies are now using facial analytics to determine age, gender, and body mass index (BMI) of a user based on the uploaded image of their face. Then various insurance options can be suggested to the customer.

Weblog data of the users is used by organisations to determine browsing patterns of the users and predict and suggest possible web pages based on their preferences. This stream of web analytics is widely in use in the retail and marketing industry for advertising web pages based on the browsing history of the user.

Digital marketing firms use ad analytics to discover which keywords in their digital advertisements appealed most to the viewers, based on their response to their ad. The response is determined by tracking the number of people who visited the webpage, who viewed a complete advertisement, who converted into an actual sale, and so on.

Primarily, the use of business analytics can be categorised based on the outcome of the dataset – into classification, clustering and association (Raj, 2015). Further, the categorisation can also be made based on the capability of the analytics solution – into descriptive (basic narration of analysis and events, mostly to be followed by subsequent analysis), predictive (forecasting of future events based on analysis), and prescriptive (highest analytical capability with narration, prediction and recommendation) solutions.

Primarily, three classes of business analytics exist: descriptive, predictive, and prescriptive analytics.

Exhibit 1.2 Corporate examples of types of analytics

Predictive analytics

• Dow Chemicals mines historical data of approximately 40,000 employees and has been able to forecast rates of promotion, internal transfers and labour count required.
• Experian (2020) predicts employee attrition, based on the computation of employee risk factors using predictive algorithms. The prediction was based on approximately 200 attributes like the performance of the supervisor, commuting distance, etc. The firm has been able to reduce turnover by 2–3% over one year of implementation.

Prescriptive analytics

• Predictive modelling of turnover and retention at Nielsen (2020) prescribes retention strategies by years of tenure of its employees. For example, for the first year of new hires, the model prescribes that establishing critical contact points is crucial for retention. The managers of new hires, therefore, are reminded to contact new hires during their first year.
• Optimal staffing was determined at a busi...

Table of contents

• Cover
• Half Title
• Title
• Copyright
• Dedication
• Contents
• List of illustrations
• Foreword
• Foreword
• Foreword
• Preface
• Acknowledgements
• List of abbreviations
• 1 Analytics in HRM
• 2 Looking for data
• 3 Modelling the business problem
• 4 Predictive analytics tools and techniques
• 5 Evaluation of analytical outcomes
• 6 Predictive HR analytics in recruitment and selection
• 7 Predictive HR analytics in turnover and separation
• 8 Predictive HR analytics in other areas of HRM
• 9 Emerging trends in predictive HR analytics
• Subject index
• Company index