The whole of science is nothing more than a refinement of everyday thinking.
—Albert Einstein
This quote from Einstein highlights the impact of our everyday thoughts and reasoning, and applies perfectly when it comes to translating traditional business analytics into People Analytics. Like many other disciplines, People Analytics, also called Talent Analytics, simply builds on the principles of traditional analytics—something that we have all grappled with at some level, whether it was in negotiating the price when buying a house or evaluating whether to relocate for a job offer. In fact, analytics forms the basis of our logical reasoning process: It helps us weigh our options and take factors such as cost, commute, happiness, and convenience into account.
To ensure that you have everything you need to smoothly migrate from business analytics to People Analytics, in this chapter we will cover:
In order to best navigate today’s globally connected, competitive marketplace, which is propelled by an explosion of digital information, companies have been embracing analytics at different paces to help sift through and derive strategic insights.
From an industry perspective, analytics adoption has followed the classic adoption curve: There have been some early adopters who pioneered the adoption of analytics into their business processes; there are those, called the late adopters, who followed the pioneers; and then we have the laggards, a group that trails far behind both. What we’ve found is that while some businesses are struggling to move beyond basic reporting, the majority of HR managers and human capital leaders do not even know where to start with their analytics journey.
In Win with Advanced Business Analytics,1 we described how the insurance and financial services industries have been pioneers in collecting, managing, and leveraging predictive analytics to create actionable insights from their data. Given the mandatory reporting environment of these fields, not to mention the direct correlation between accurate data and their revenue, these industries first began using analytics in the 1800s to price life insurance and underwrite marine insurance. They have since continued innovating, undergoing many waves of improvement, including the use of neural networks to optimize premiums; introduction of credit scores to evaluate a customer’s financial position; and application of various behavioral, third-party, and social media data sets to supplement their forecasts and future predictions.
Over the years, the success stories have been numerous, and financial institutions that are not using predictive analytics today in product pricing and underwriting would be seen as obsolete and doomed to failure.
Even if People Analytics and predictive modeling in HR are fairly nascent, interest in HR metrics started before the current trend of HR and human capital analytics. As a result, 10 major events occurred:
In 1911, Frederick Taylor, a mechanical engineer who sought to measure the productivity of workers and improve industrial efficiency, pioneered the scientific management of time-and-motion studies. He developed a method for capturing and measuring the effectiveness of an organization’s employees’ work, and summed up these efficiency techniques and best practices in his book The Principles of Scientific Management.2 Taylor believed in transferring control from worker to management and focused on the distinction between mental labor (planning work) and manual labor (executing work).
To demonstrate these principles in action, Taylor used “The Parable of the Pig Iron,” which applied his time-and-motion studies to illustrate the role management plays in determining the correct workload for each employee that keeps him or her the most efficient and happy.3 Despite a lot of criticism of his work and principles, Taylor spurred a whole movement of analyzing the data behind physical work, which was first implemented in industrial engineering.
Two years later, in 1913, Hugo Munsterberg, a psychologist and an admirer of Taylor, extended these ideas into applied psychology, including industrial and organizational, legal, medical, educational, and business settings. In his book Psychology and Industrial Efficiency,4 he addressed many topics of industrial psychology and claimed that it is not only the physical strength of a worker but also his or her psychology that defines business productivity. He introduced the industry to the notion of worker selection based on a scientific approach of worker testing and job analysis. This later became known as an assessment center.
Munsterberg also outlined a new science in his book—a blend of modern laboratory psychology and economics. He believed that the question of selecting the best possible person for a particular vocation comes down to making a process very scientific (for instance creating tests that limit subjectivity), and that it comes down to fitting the person with the correct skill set with the correct position to maximize productivity.
During World War I (1914–1918), the U.S. Army started large-scale testing called the Army Alpha and Beta Tests of World War I,5 which were the first mental tests designed for the masses. In developing these tests, psychologists proved that one could be quite intelligent even though illiterate or not proficient in the English language. This led to the creation of two subsequent tests: the Army Alpha for literate groups and the Army Beta for the illiterate, low literate, or non-English-speaking individuals.6 Both tests were based on the theoretical assumption that intelligence was an inherited trait, and the assumption was made that native intelligence was being assessed. These tests also demonstrated how quantitative analytics could be leveraged for resource selection.
In 1920, E. L. Thorndike published his psychometric view of social intelligence, dividing intelligence into three facets: abstract intelligence, or one’s ability to understand and manage ideas; mechanical intelligence, or one’s ability to manipulate concrete objects; and social intelligence. In his classic formulation, Thorndike stated, “By social intelligence is meant the ability to understand and manage men and women, boys and girls—to act wisely in human relations.” Similarly, Moss and Hunt (1927) defined social intelligence as the “ability to get along with others.”7
In 1921 and 1923 in English, Carl Gustave Jung, a renowned Swiss psychiatrist and one of the founding fathers of modern-day psychology, published his book Psychological Types,8 in which he proposed that people are innately different, both in terms of the way they see the world and take in information, and in terms of how they make decisions. In this tract, he expanded the concept of social intelligence into a science of work, and taught that it is not only our individual skills that create productivity, but also our personalities and how we get along with each other.
Jung’s work in social psychology spurred a global evolution in personality testing. In 1943, driven by a desire to help people understand themselves and each other better in a postwar climate, Isabel Myers set about devising a questionnaire that would identify which psychological type a person was. To do this, she enlisted the help of more experienced psychometricians, and her work, which was first implemented in industrial engineering, was later endorsed by professors from the Universities of York, California, Michigan, and Florida. In 1943, she and Katharine Briggs published the first Myers–Briggs Type Indicator questionnaire, which is still in use today as an assessment tool.
While the roots of analytics are firmly planted in the late nineteenth and early twentieth centuries, most of the HR metrics in use today were developed following World War II. With the essential building blocks in place, much progress was made quickly in HR analytics.
In 1978, Dr. Jac Fitz-enz published the first HR metrics. And in 1984, he implemented the first benchmark metrics in HR and published his findings in the reference book How to Measure Human Resources Management.9 In it, he created 30 HR metrics that were developed through a joint effort between the Saratoga Institute and the American Society for Personnel Administration that later became the current Society for Human Resource Management (SHRM) in 1989. Some of the original 30 HR metrics identified where predictive analytics can be applied to derive actionable insights are: voluntary separation, involuntary separation, voluntary separation by length of service, time to fill jobs, time to start jobs, revenue per employee, expense per employee, hire as a percentage of total employees, cost of hire, absence rate, human capital return on investment, turnover rate and cost, and vacancy rate. Employee attrition, also called voluntary separation, happens when an employee decides to leave your organization.
One way to apply predictive analytics is to address questions such as: Who are employees at risk of leaving your organization? When will they leave? Why will they leave? To answer those questions People Analytics harnesses all the data available from internal source HRIS (human resource information systems) data to external talent and market data such as publicly available talent data, labor market data, and competition data to anticipate “talent at risk to attrite” and proactive actions to put in place to save them. There is also a body of People Analytics opportunities arising from a combination of 2, 3, or 4 of the 30 HR metrics to better understand some specific workforce challenges you may face.
In 1995, Rutgers University professor Dr. Mark A. Huselid’s work on high-performance systems demonstrated that the systematic management of HR was associated with a significant difference in organization effectiveness. This work provides evidence that the SHRM did, indeed, have a strategic potential. Shortly thereafter, in 1996, Drs. Robert Kaplan (Harvard Business School) and David Norton introduced the balanced scorecard. And, in 2001, in their book The HR Scorecard, Brian E. Becker, Mark A. Huselid, and Dave Ulrich highlighted how HR scorecards, which demonstrate the alignment of HR activities with corporate strategy and activity, improve organizational outcomes.
In 2005, the first talent management system (TMS) was created. TMSs are integrated platforms that automate and improve the key major processes of talent management, including recruitment, performance management, learning and development, and compensation management. They are tools that also store a wide variety of employee data, from recruitment data, learning and leadership data, performance and compensation management data, and resume profile data to job postings performance and work data. Today, TMS platforms can also store and manage social media data, as well as other digital footprint and talent behavior data.
It was only around 201010 that predictive analytics began to appear in the HR departments of most leading companies. And, according to Josh Bersin,11 only 4 percent of Fortune 1000 companies are using predictive analytics and, in doing so, this group’s stocks outperformed their peers on the Standard and Poor’s 500 by 30 percent.
First adopted and implemented by financial and insurance companies in the 1800s, and then within the HR industry throughout the twentieth century, data analytics has been gradually and meaningfully changing business practices and improving organizations’ decision-making abilities.
In many organizations, HR, the so-called intuitive and experience-based group, still shares the same challenges as the marketing department. Oftentimes, the HR department feels misunderstood and is being constantly requested to be more data-driven and become a strategic business function at the C-suite table.
Over the past 10 years, we have had the opportunity to speak at conferences across the world and to meet with industry leaders in staffing, talent management, and HR from organizations of all sizes on a regular basis. The most common challenge we consistently hear in People Analytics is that industry leaders are inundated with data from a variety of sources and are hampered by disconnected tools and systems. They are seeking the best analytical practices in order to create innovative talent life cycle management processes and optimize their HR teams.
While some are aware that HR will undergo a seismic change with Big Data analytics, those of us who work in the field believe the odds are high that the analytics revolution, which invaded the marketing industry in the late 1990s (with the beginning of the World Wide Web), is coming in HR. So how do we ready our teams for the future?
Having worked in business analytics for many years, enriched with interviews and research with those leaders, and sharing with them business analytics success stories in other functions of the organization, we have found that one of the best ways to tackle the migration from business analytics to People Analytics is by using the lessons we learned when the marketing industry adopted analytics. Why? This is because HR and marketing have significant common denominators, and they both used to be non-data-driven functions and cost centers for most organizations.
By reviewing how marketing enhanced its role and practice over time, we can then derive best practices for the HR field.
Before 1990 (the beginning of analytics use in marketing), marketing was basically a discipline of creativity, art, “gut feel” judgment calls, and experience. In search of becoming a strategic business partner and more accountable in a growing competitive marketplace, the industry underwent a massive change by integrating analytics into its business practices. Searching to strategically attract, segment, acquire, grow, retain, and reward customers, marketing opened up to analytics in order to fully understand its customer base, target demographics, the 360 degrees of its different customers, and the market, and to optimize the entire customer life cycle management and customer relationship management. Thus, marketing analytics was born.
Marketing analytics ushered in a new era. The old days of “spray and pray” and shooting blindly to acquire and retain customers quickly became obsolete. Top-performing companies gradually began to sunset those old gut-driven practices and started using analytics. For the majority of successful companies, marketing coexists with marketing analytics, and constantly leverages the power of data intelligence, which enables them to become a balance between a disciplined art (experience, judgment, and instinct) and science (the intelligence of data).
Talent management and HR teams can learn a lot from their marketing colleagues, adding or building a People Analytics program in order to proactively embrace the analytics revolution. Similar to traditional marketers, HR professionals are generally not known as “data geeks.” The field of HR has traditionally leveraged intuition, judgment calls, instincts, and experience. Although some managers may still rely on isolated metrics from HR dashboards to make major decisions, the field is not historically known for using predictive analytics as a systematic part of the decision-making equation.
Dr. Jac Fitz-enz12 rightly points out that “HR managers have done a poor job teaching the C-level executive how to achieve a high rate of return on employee investment. On top of that, there is a serious perception issue that the HR function has to fix as well.”
In today’s globally connected and competitive talent marketplace, HR is facing challenges from multiple fronts, and some HR leaders we have spoken with often say their departments are perceived as:
Let’s take a closer look at some of these challenges, so that we can begin to address and rectify them.
HR managers have done a poor job of teaching C-level executives how to achieve a high rate of return on employee investment. Analytics can be used to demonstrate the value of these investments and illustrate the important implications for the organization’s bottom line—for instance, how a 2 percent turnover would impact sales and profitability.
There was even an article in Forbes claiming that companies should fire their HR departments.13 In it, a group of economics researchers conducted a study on 2,500 resumes either with or without photos of the applicant and found that being good-looking (in the case of women) did not help an applicant find a job. According to the article, “attractive” women faced an uphill struggle to get a chance at a job, because 93 percent of the HR staff deciding whether to call someone for an interview were female.
Leaders want to leverage the power of advanced analytics and data intelligence to make more informed decisions. They are seeking to harness the power of Big Data analytics to improve performance in their business practices, and address core business challenges and core talent management questions such as:
These questions are similar to the ones that the marketing industry was able to address by implementing business analytics and predictive modeling into their practices.
At this point, you are probably wondering how HR can piggyback on marketing to undergo its own analytics revolution. To manage human capital for tomorrow, you need HR metrics that are inherently predictive and that will have a high business impact. To move from business analytics to People Analytics, we can simply replace the word customer from any marketing analytics strategy with the word talent or employee. This will provide us with a basic migration of marketing analytics to People Analytics (talent analytics or workforce analytics), as seen in Table 2.1. This is possible because the common denominator between marketing and human capital management is human behavior.
Table 2.1 From Marketing Analytics to People Analytics
Marketing | Human Capital and HR (People Analytics) |
Customer life cycle management | Talent life cycle management |
Customer relationship management | Talent relationship management |
Customer 360-degree analysis and understanding | Talent 360-degree analysis and understanding |
Over the past decades, HR has heavily invested in tools and technologies; however, it is time to move beyond simple dashboards, scorecards, and isolated talent metrics and embrace advanced business analytics to derive true value for organizations.
The following sections describe how People Analytics can use advanced business analytics paths and mainstream impact to address human capital management challenges.
Advanced business analytics starts with a business goal or question, integrates disparate data sources, creates predictions for the future, and leads to strategic actions with measurable results. It is powered by predictive modeling, which has helped marketing teams attract, acquire, engage, grow, retain, and reward their most valuable customers—goals that are very similar to those of HR departments looking to maximize their talent management cycles and that will require HR to think like a marketer.
Depending on their level of analytical maturity and their most pressing business objectives, companies have embraced analytics to create business value at different stages; predictive analytics is becoming a pervasive and competitive differentiator across different industries and business functions:
It’s clear that analytics is being used in nearly every aspect of our lives and across myriad industries. So, how do we apply it in HR to help solve current challenges?
People Analytics starts with a talent management business question or goal, and then integrates disparate data sources together to create predictions for the future, which can then be used to outline businesses’ actions with measurable results. Let’s take a closer look at the major components of People Analytics.
A good talent management business question or goal will be directly related to workforce planning, talent sourcing, talent acquisition, talent onboarding, talent engagement, talent turnover, talent development, retention, safety, or employee well-being. For instance:
Now that you have a talent management business question in mind, how do you go about integrating disparate data sources? Typically, data sources can be broken down into three categories: talent data, company data, and labor market data.
By identifying these different data streams and linking them to your talent management business question, you will start to see which data points are key to creating an actionable plan.
Creating business predictions that lead to strategic actions and measurable goals for the future requires you to use predictive models to anticipate what will happen with your employee base and to enable you to put proactive plans in place, addressing:
While these questions are critical for any top-performing organization, you need to translate them into measurable actions that can be presented as a business case to your executive team. This means you need to connect this talent data with business data. For instance, consider:
The entire process of creating actionable insights with People Analytics is summarized in Figure 2.1, People Analytics Virtuous Process, where you start with your talent management business question, and then integrate data and use analytics to gain actionable insights.
Traditionally, predictive analytics has helped companies to address the basic business questions of who, when, and why. However, when applied to the staffing industry, predictive analytics can help to anticipate and optimize:
To process, manage, and optimize the exponential growth of resumes and other talent data coming from multiple sources, staffing firms have to leverage Big Data intelligence technology to fully understand and maximize their recruitment metrics. The benefits of performing this type of deep-dive analysis include:
Predictive recruitment analytics and Big Data intelligence tools are changing the way organizations view, analyze, and harness their talent data. Leveraged efficiently, predictive analytics allows staffing teams to create economic value from their talent data, helping them become more competitive and, ultimately, more successful.
So, by now, you’re getting comfortable with the concept of advanced analytics and its associated insights, which include helping with:
Armed with this knowledge, we can begin to build connections between our data analytics and our insights through what we call the “actionable insights bridge.” One of this book’s authors, JP Isson, built and uses this model to illustrate to executive teams how actionable insights can be created using both little data and Big Data (see Figure 2.2).
Earlier, we discussed that the migration from business analytics to talent analytics happens simply by switching the word customer with the word talent. With that in mind, Tables 2.2, 2.3, and 2.4 will help us map preliminary reporting analytics to predictive analytics. This begins by translating your business question into a talent analytics question. For instance, a marketing question about proactive customer retention could be translated into proactive talent retention, and can leverage similar analytics techniques.
Table 2.2 Mapping Business Analytics and Talent Analytics for the “What Happened?”
Analytics Question | Business Analytics Solutions | Talent/People Analytics Solutions |
What Happened? |
Customer dashboards and customer reports | Talent dashboards and talent reports |
Customer scorecard and customer key performance indicators | Talent scorecard and talent key performance indicators | |
Cubes | Human capital cubes | |
Triggers and alerts | Triggers and alerts |
Table 2-3 Mapping Business Analytics and Talent Analytics for the “What's Happening and Why”?
Analytics Questions | Business Analytics Solutions | Talent/People Analytics Solutions |
What's Happening and Why? |
Customer segmentation | Talent segmentation |
Customer profiling | Talent profile analysis for turnover and retention | |
Market research | Market research employment conditions | |
Customer satisfaction survey | Employee satisfaction survey | |
Voice of customer analysis | Voice of candidate analysis Voice of employee analysis |
|
Competitive intelligence | Competitive intelligence: Who is hiring? Where and when? | |
Social media analytics | Candidate social media analytics | |
Text analytics | Text analytics | |
Mobile analysis | Candidate mobile analytics | |
Social graph analysis | Candidate and employee social graph analytics |
Table 2-4 Mapping Business Analytics and Talent Analytics for the “What Will Happen?”
Analytics Questions | Business Analytics Solutions | Talent/People Analytics Solutions |
What Will Happen? |
Predictive Models Customer and prospect predictive models Customer acquisition Customer targeting and selection Customer retention Customer reward and loyalty program Customer upgrade Customer attrition (voluntary churn) Customer involuntary churn (termination due to bad payment) |
Predictive Models Talent and candidate predictive models Talent acquisition Talent targeting and selection Talent retention Talent reward and incentive program Talent promotion Talent attrition (voluntary churn) Talent involuntary churn (termination due to bad performance) |
Forecasting Forecast market opportunities Forecast number of new customers Forecast number of existing customers Forecast number of lost customers |
Talent Scorecard Forecast market opportunities Forecast number of new employees Forecast new and existing employees Forecast number of lost employees |
|
Forecast churn by business function Forecast bottom line |
Forecast turnover by business function Forecast bottom line |
|
Optimization Customer life cycle management Customer relationship management optimization |
Optimization Talent life cycle management Talent relationship management optimization |
Creating a talent analytics center of excellence requires prudent investments in people, processes, and technology.
The overall data scientist labor shortage prevalent in other industries is even more acute in the talent management and HR field. In a recent Wall Street Journal article, Jeanne Harris and her colleagues rightly assert that “It takes a team to solve the data scientist shortage”; one of this book’s authors, JP Isson, provided insights to Wall Street Journal for his team solution approach at Monster.14 Similarly, it takes a team to build a People Analytics center of excellence to solve that shortage.
So, how do we do that? In addition to utilizing existing analytics resources, tools, and technologies to minimize your cost, building a multidisciplinary team of people with a variety of backgrounds will enable you to view your data from different perspectives and derive even greater insights from it. Such a skills-diverse team might include:
The main goal of every organization, and arguably the reason you are most likely reading this book, is to create a high business impact. In order to do this, you need to ensure you have the right structures in place to properly pull your HR analytics.
Thus, in 2012, Josh Bersin (Bersin by Deloitte), who is one of the leading thinkers in HR analytics, created the HR Analytical Maturity Model to explain the different levels of HR analytics adoption. He defines these stages based on four maturity levels:
With Bersin’s HR Analytical Maturity Model in mind, let’s explore the IMPACT Cycle that will help you quickly navigate across the aforementioned four stages to create high business impact.
In speaking with industry leaders, experts, and business partners, we realized that the data challenges they were facing could be addressed by leveraging advanced business analytics, and the same approach and methodology used in traditional business could be applied to talent life cycle management. In fact, there are critical success factors that organizations leveraging talent data analytics are using to drive better business outcomes. Companies such as Accenture, Deloitte, CGB Enterprises Inc., ConAgra Foods, Microsoft, Google, Goldcorp, FedEx, Xerox, Hewlett-Packard, SAS Institute, Bloomberg, Bullhorn, Sysco, Shell, CISCO, General Electric, Johnson & Johnson, Dow Chemical, and Harrah’s have invested in People Analytics to optimize their human capital and improve business performance, and some have even gained global recognition for being a Best Place to Work.
So, how do these companies do it?
The short answer is that they are focusing on the IMPACT. Coined in our previous book, Win with Advanced Business Analytics, IMPACT is a framework for creating actionable insights at every stage of the talent management cycle. While data is a necessary component of every business, it alone is not sufficient to unlock value for your organization—for this, actionable insight is required.
We speak with business leaders on a regular basis about their data assets and challenges, and it never fails—a sentiment we hear consistently is that their organizations are drowning in data but are lacking in understanding and actionable takeaways from that data. Based on our experience leading and building analytics teams from the ground up, as well as taking into account input we received from researching successful analytical organizations, we developed the IMPACT Cycle to help guide analysts through the process of ensuring they are insightful business partners, rather than just purveyors of data. (See Figure 2.3.)
It’s not always an easy task to get your analysts to pull their heads up from the data and focus on the business; in fact, it is a bit of both an art and a science. The IMPACT framework is made up of the following steps:
The IMPACT Cycle can help you guide discussions on some key talent management decisions, including:
Over the past four decades, HR has shifted from mainframe computers and manual payroll systems to cloud-based technology, software as a service (SaaS), and client–server solutions. In fact, the emergence of TMSs has enabled companies to capture, store, and manage a variety of employee and HR data, including social media data and talent digital footprints. As we see the amount of data increase, we see the number of HR tech firms who create these solutions also increase. In fact, this industry represents roughly $4.5 billion in business.15
But, because this is a journey, it is always best to start small and first master the human capital information at your disposal. As we mentioned, analytics is fairly new in HR, so leveraging experts in the field of traditional analytics technology can help you map out the best approach for your organization.
According to the 2014 HR trends survey conducted by the Information Services Group,16 there are three major benefits that companies expect to realize from their HR investments:
When investing in talent management tools and selecting service providers, there are some basic considerations to keep in mind:
Before getting started, there are also a baker’s dozen best practices to keep in mind for building a successful People Analytics journey.