Before diving into the data each day, we would do well to a step back from our desks to review the nature of the problems our organizations are trying to solve. The omnipresence of data doesn’t make the management of our customers, teams or market shares any easier, for each data source constitutes yet another piece in the increasingly complex puzzle of business. Solving this riddle requires making sense of how the data reflects the larger picture of human nature, your business, and your market. What types of problems are we trying to solve, are these problems fundamentally different from those faced by managers in the past, what role can analytics play today?
If the data tells us anything, it’s that the Taylorian vision of “one best way” has long outlived its usefulness. Because business by its very nature is “messy”, operational management is largely about addressing processes, policies and decisions that didn’t work as planned. These challenges arise in part because consumer perceptions of value are based today more on their experiences rather than on the characteristics of a company’s products or services. This complexity also comes from the fact that these experiences are evaluated differently based on each stakeholder’s state of mind, cognitive biases, and context. This is also the result of the evolution of the information economy: the volume, velocity, and variety of data have influenced the veracity and visibility of the data at our disposal.[i]
David Snowdon and Mary Boone remind us that the nature of the problems we face today has evolved considerably over the years.[ii] They speak of levels of complexity that determine how managers evaluate the context of their business and where they look for data. They suggest that first level problems concern linear chains of events where process optimization provides pertinent and replicable solutions. Second level business challenges result from malfunctioning processes in which the answers can be deduced from management’s prior experience. Third level challenges require reformulating the problem, for neither the process nor prior experience provides sufficient data to deal with the problem. Snowdon and Boone argue that markets and organizations over the years have largely addressed the first and second level problems, leaving us today third level challenges of customer satisfaction, employee engagement and organizational effectiveness that are intricately woven into the way in which we envision work.
Addressing business challenges has less to do with the quantity of data at hand than our ability to use the data to incite collective action. Good, better, and great decisions depend upon the decision environment in which we operate.[iii] Good decisions are possible in deterministic decision environments in which the answer can be determined from the data at hand. Unfortunately, most business decisions are taken in stochastic environments in which the right decision cannot be found in the available data — but better decisions are possible in reducing the causes of uncertainty. Finally, great decisions are those in which the context, challenges, and solutions allow us to re-examine the nature of the decision-making process itself. Focusing first on the context in which the problems arise, before looking at the data, is one of the major takeaways of decision science.
If measurement is about reducing uncertainty, management is about elucidating complexity.[iv] Our business challenges arise from our illusions of well-structured organizations and organized markets and underpinned by ideal simplifications of how business “should” be run. Organizations and markets today are interdependent open and interdependent, consumer and business logics are more often than not self-generating. “Emergence” describes the resulting behavior that defies time tested business practices, and premises the difficulties of predictive and prescriptive analytics. If data science’s role is to reduce risk, uncertainty, and ambiguity, we must take a close look at how consumers, employees, and management apprehend the problems before measuring their motivations, objectives, and actions. The bottom line, before zapping back to work, and the data behind your screen, take a good look at the world around you.
Looking to enhance your data science skills? In our Summer School in Bayonne, as well as in our Master Classes in Europe, we put analytics to work for you and for your organization. The Institute focuses on five applications of data science for managers: digital economics, data-driven decision making, machine learning, community management, and visual communications. Improving managerial decision making can make difference in your future work and career.
Lee Schlenker is a Professor at ESC Pau, and a Principal in the Business Analytics Institute http://baieurope.com. His LinkedIn profile can be viewed at www.linkedin.com/in/leeschlenker. You can follow us on Twitter at https://twitter.com/DSign4Analytics
[i] Marr, B. (2014), Big Data: the 5v’s everyone should know, LinkedIn
[i] Snowdon, D. and Boone, M. (2007), A Leader’s Framework for Decision-Making, HBR
[ii] Schlenker, L. (2017), Measure the Quality of Your Decisions and Not Just Your Data, Medium
[iii] Straub, R. (2013), Why Managers Haven’t Embraced Complexity, HBR