Introduction

My choice of a surgeon was a selection decision, plain and simple. Though I did not realize it at the time, it was also a case study in data and intuitive decision making. I am a fact guy; it is really important to me to make as optimal a decision as possible, but I also have learned to trust my instincts. The data analyzed and research I did was critical for making an optimal decision, but just as important was my own intuition. There is still no software application, supercomputer, or A.I. tool that can touch our ability to assess certain intangibles.

The use of analytics has a long history associated with human capital management (HCM) decisions, but far too many organizations continue to use these tools for reporting simple descriptive statistics and correlations. Advanced analytics has been adopted by other business functions such as finance and marketing; however, it still has a long way to go to be fully utilized for HCM decisions. According to research conducted by IBM in which 700 chief human resource officers were interviewed, less than 25% are using sophisticated analytics to predict future outcomes and for decision making.¹

The underutilization of advanced analytics associated with HCM decisions is a problem because the jury is in: There is a real and direct bottom-line impact associated with getting these decisions right. If an organization wants to deliver the highest quality goods and services, superior customer service, and the most innovative products, effective HCM is required. Getting HCM right boils down to making many decisions and making them correctly.

The challenge and opportunity is that the entire range of HCM decisions (from where and how to recruit and hire, how to reward and motivate, and which policy and practice to use in a specific situation) is getting very difficult to make optimally. There are a huge number of different practices and policies and combinations to choose from and an ever-increasing amount of pertinent information useful for making these decisions. Fortunately, there is new research, insights, analytical tools, and processes associated with advanced analytics that can assist in making these decisions much more optimally. For example, companies like Xerox and Google are using predictive analytics to evaluate which characteristics are associated with good employees, and this information is used to help with employee selection.² The use of advanced analytics can help eliminate all forms of bias associated with selection and promotion decisions and also provide a mechanism for compensating and rewarding people in a more accurate and fair manner.

If biases are eliminated from the decision-making process, previously unconsidered possibilities will emerge. SAP, the German software giant, has announced that by the year 2020, 1% of its workforce will fall on the autistic spectrum. The company has found greater engagement and productivity in locations where they have adopted this hiring policy.³ Some of the most productive and capable computer programmers fall on the autistic spectrum. By undermining any prejudice and bias associated with autism, SAP is potentially developing a previously unrecognized HCM competitive advantage. Advanced analytics can aid in the process of identifying these possibilities by eliminating all extraneous factors from decision making so that only merit and potential is taken into consideration.

A number of factors are converging that make this the right time to start using data and other information to make more robust decisions. Technology has become more accessible, user friendly, and powerful. There have been recent advances in machine learning, natural language, and deep Q&A expert systems (for example, IBM’s Watson beating two former Jeopardy! champions). In addition, we know substantially more about what really contributes to organizational performance (for instance, balance scorecards and intangible capital), and we are also getting much better at modeling what is important to people and how people think and how they actually behave (for example, behavioral psychology, behavioral economics, and neuroeconomics).

Many of these paradigm-shifting developments have not been incorporated into our decision-making processes. It has long been held that we humans are rational decision makers who are very self-centered and selfish. Recent research has shown that we are rarely inclined to make rational decisions and that we are actually very cooperative, collaborative, and unselfish and want to be treated fairly and to see others treated the same.⁴ These finding have tremendous implications for how we manage the employment relationship. Equity matters because it matters to the primary input in all organizations’ output equation: human capital.⁵ Humans want to be treated and rewarded fairly. If they are not, they withhold value-creating information and effort, are more likely to be absent, quit, and sometime actively conspire to undermine the goals of the organization.

I refer to all these recent findings as “The New Human Science.” I integrate the recent findings on what motivates us, what influences our decision making, and what our natures are like, with recent advances in technology in order to assist us with making more optimal value-creating decisions.

This is not to suggest that advanced analytics will replace human expertise. Instead, I believe that it will complement it. In 1997, the chess master Gary Kasparov lost to the IBM computer Deep Blue. However, as Kasparov later reported, the most unbeatable champion is not a supercomputer. The most powerful computer can be beat by a good amateur chess player working with a standard PC. The optimal decision maker is not computer or human alone, but rather the combination.⁶ That is the position taken in this book. When well-seasoned human expertise is combined with the right advanced analytics, the decisions made will be much more likely to create value for everyone.

There is data and there is data. I will be talking about techniques, but equally important is to get the questions right. The tools have gotten really cool and the types of analysis that are now possible were not even imagined ten years ago. None of that changes the fact that data is really about stories. In the case of this book, stories are about what is going on in your organization—what (and whom) is working and what is not. Everything that is discussed here is meant to help us become better and more accurate data story tellers.

Some might view big data, advanced analytics, and data science as being sterile and potentially dehumanizing. I argue the exact opposite. The use of these tools, when coupled with the right kind of human expertise, can help us become much more humane decision makers. By humane, I mean fairer, inclusive, and merit based—ultimately making our organizations more equitable, collaborative, and successful.

One final note. This book is meant to be used in conjunction with its associated website, DecisionAnalyticsInc.com. The focus of the book is on what can and should be done with advanced analytics and optimal HCM decision making. The website will provide tools and more detail on exactly how this this optimal decision making is accomplished.