People analytics and event history analysis: an a priori preface

I'm writing a book about people analytics and event history analysis. Whenever I start writing something, I write about what I'm going to write and why. Thankfully, when you're writing a book, that exercise isn't a waste of time because it forms the Preface.

So here is my a priori preface before writing a single word in any of its chapters. I hope it gets you excited about the book. It might be a bit too negative and snarky, but that's my personality and I'd feel disingenuous if I wasn't at least a little bit. Acknowledgments and About the Author sections will come after it's done.

Preface to the forthcoming book Often: A book about people analytics and event history analysis

Why I wrote this book

For six years as a data scientist, I’ve witnessed companies repeat the same mistakes when they measure the pace at which they recruit, promote, transfer, and retain workers. No joke: companies spend millions of dollars optimizing bogus people metrics that – turns out – measure fundamentally unknowable, thus meaningless quantities. That’s a travesty of analytics. Considering the “people” part of the term “people analytics”, it’s also a potential tragedy of analytics.

Thankfully, there are better analytical methods called event history analysis that we can use to construct more meaningful metrics and make smarter decisions. I trained in these methods as a National Science Foundation (NSF) Trainee and National Institutes for Child Health and Development (NICHD) Fellow at the Department of Anthropology and Center for Studies in Demography and Ecology (CSDE) at the University of Washington (UW). Since then, I’ve advised three companies in these methods as a staff data scientist. I learn more about event history analysis every day, including while writing this book.

In October 2019, I presented some of the material in this book at the Seattle People Analytics Forum hosted at the Nordstrom headquarters in Downtown Seattle, Washington. I’d been thinking about writing this book for a while. The thoughtful comments and positive response I got at the forum drove me the rest of the way to that decision.

Why read this book?

Read this book if you want to save your company money and increase efficiency by improving your workforce recruiting, development, and retention metrics. This book teaches you analytical methods that address a broad range of business questions with both scientific rigor and boardroom clarity. You’ll learn practical solutions ranging from SQL query templates, to statistical models for prediction and causal inference, to suggestions for workforce-optimizing products driven by event history analysis. You’ll understand both the promise and limitations of these methods.

While applied to the workforce, the methods in this book extend to any research question that involves waiting times, rates, repeated events, or transitions among multiple states. Another book can (and will) focus on customers, vendors, and other populations.

Structure of the book

Chapter 1 reviews the importance of measuring the rates and waiting times associated with people in the workplace, introduces the mistakes that many organizations make when calculating those metrics, and descibes how the book will correct those mistakes. Chapters 2 through 8 correct each mistake in turn, with subsequent chapters building on what came before. Chapter 9 drives home the message that poor decision-making wrought by bad recruiting, development, and retention metrics waste billions of dollars of investors’ money across the private sector alone. Appendices provide mathematical details and supplementary code. There’s an index, of course, and all chapters cite their references.