Walter DeGrange - Field Guide to Compelling Analytics

We wish to personally thank the following people for their contribution in terms of inspiration, shared knowledge, and other help in creating this book:

All the analytics professionals with whom we have collaborated and learned over the years.

Natalie Scala and James Howard for allowing us to write this book in their analytics series.

Norm Reitter and Jennifer Ferrat for their help in selecting the perfect title.

Al Roth for his passion for the Kidney Exchange program that provided many of the stories to demonstrate the concepts in the book.

Walter DeGrange is the Director of Analytics Capabilities for CANA. He has extensive experience in implementing analytical models in both the Department of Defense and commercial areas. Prior to CANA, Walter served 21 years in the US Navy as a Supply Corps Officer. He was the Director of Operations Research at several military commands as well as a Military Assistant Professor on faculty at the Naval Postgraduate School in the Operations Research Department.

Walter is also very active in analytics education. He is an adjunct faculty member at the University of Arkansas with both the Master of Science Operations Management and Engineering Management programs. He is an MBA Executive Advisor at the NC State University Poole School of Management. Walter serves as the Military Operations Research Society (MORS) Course Director for the Critical Skills for Analytics Professionals Certificate Program and he teaches the Analytics Capability Evaluation (ACE) Coaching Course for INFORMS.

Lucia Darrow is a data visualization expert with experience implementing analytical models in a variety of environments, including healthcare, manufacturing, defense, and finance. An industrial engineer by training, Lucia also has extensive experience with Lean manufacturing and the modeling of complex systems. In a professional role as a learning and development lead and as a community events organizer, she has facilitated many successful networking events, tutorials, and classes in a wide range of analytics topics. Lucia is actively involved in the analytics community through RLadies and recently co-organized the first Vancouver, BC Datajam.

She holds a Master of Science in Industrial Engineering from Oregon State University and a Bachelor of Science in Mathematics from Dickinson College. Lucia is currently Content Marketing Data Analyst with RepRisk AG, an environmental, social, and corporate governance (ESG) data science company based in Zurich, Switzerland.

DOI: 10.1201/9781003204190-1

A NALYTICS HAS influenced the world in many ways over the past half century. With advancements in computers, software, and algorithms, analytics will continue to grow and transform the world of the future in ways we can only dream of today. The impact has been felt in many areas including business, medicine, sports, defense, transportation, and education. With the increased use of analytics, it is important to understand how to design, develop, communicate, and represent analytical models and solutions. While the application of analytics grows, the fact remains that the mathematical principles that analytics is based on, are only understood by a small percentage of the population known as Analytics Professionals (APs). The APs must be able to convince the larger population that these algorithms work and guide their application.

An AP is someone who uses data and math to improve decisions. In this book, we are including all of the following titles into the broader description of AP: Data Scientist, Operations Research Analyst, Data Engineer, Data Architect, Data Storyteller, Machine Learning (ML) Engineer, and Artificial Intelligence (AI) Engineer. This definition encapsulates any field that uses data and advanced math to yield insight. We do not refer to any specific job title due to the fact that these roles change over time and that sometimes the meaning of the titles vary from one organization to another. New titles are continuously created and the connection between all of these roles and what they entail evolve over time.

Could many of these lessons and techniques that we cover in this book be used by a spreadsheet analyst? Absolutely. However, we will focus on use cases where the mathematical techniques are more complicated in nature, such as mathematical optimization or data science.

This book is designed for the individual AP. Most other books dealing with analytics tend to be focused either on the analytics team or the organization that is using analytics. When trying to apply the insights provided by these books at either the team or organizational levels, the lessons are difficult to scale down to a single person. For example, the requirement for senior leadership at an organization to support the implementation of a major analytics project is a common recommendation. Who should be in charge of accomplishing this and how can an individual AP make a difference? Another recommendation that is commonly highlighted is an increase in the level of communication. When implementing this insight for an entire team, then what team member(s) need to sharpen their communication skills?

This book focuses on the contribution of the individual AP to the entire project. In analytics, as with the rest of life, a person can only control their own actions. Focusing on the individual AP is a unique perspective. While steps taken by one person can increase the probability of success for any given analytics project, these actions alone cannot guarantee success. As such, we recommend pairing the lessons from this book with approaches from the team and organizational level to help shape an analytics culture.

The book is organized using the equation in to illustrate the individual AP's ability to convince another person or group of people.

Convince them equation.

The other person or group may range from fellow analytics team members to leadership within their own organization to world leaders. Using the equation allows us to emphasize the point that the variables on the left-hand side are different for every audience. No two people or group of people have the exact same experience, education, and relationship with the AP. Likewise, no two APs have identical backgrounds and strengths. Therefore, factors on both sides of the equation are changing over time and vary from individual to individual.

Compelling analytics answers the question so what?. It presents focused, engaging results that show the impact potential of an analysis, while maintaining truthfulness and transparencywithout skipping over or distorting limitations to the analysis. This integrated approach to analytics not only brings stakeholders into the driver's seat of an analysis, but also makes them invested in where it can lead.

The focus of this book will be largely on analytics in a business context, which aims to help companies understand their operations and perform better. However, many insights on analytics implementation and the case studies woven throughout the text are applicable in other contexts such as academic research.

Any complex process is built by a series of steps. These steps are dependent on each other for overall success. Analytics is a process that takes data as an input, uses math modeling as a tool to transform the data, and outputs information that can be used in other formats such as data visualization or decision support. When you break any process down to component actions, there is an inherent risk of over-emphasis on the success of one of the actions. In the case of analytics, these actions include such things as communication and data visualization.