Fighting Churn with Data: The science and strategy of customer retention

A must read case study for anyone looking to break into Data Science. "Fighting Churn with Data" is full of practical examples from Dr. Gold's career up to his time as the Chief Data Scientist at the subscription services company Zuora. Churn occurs when a paying customer leaves a subscription service. It is a crucial metric for any business with reoccurring revenue. As more and more companies move to the subscription economy, this is an important business model for Data Scientists to understand. Therefore, this book represents an excellent practice project for a budding Data Scientist or a skilled practitioner looking to better understand this niche area. In the book and associated Twitch streamed videos, Dr. Gold provides guidance on the SQL and Python code required to conduct churn analysis. In both resources, Dr. Gold really delves into the process of feature engineering (i.e. finding and/or generating predictive features from a mess of raw data). For the project, Dr. Gold created a realistic simulated social network dataset so that data practitioners can follow his analysis through hands-on coding. TL;DR: This is a solid and practical guide for all Data Scientists, as well as anyone looking to improve customer retention. The book teaches how to transform raw data into measurable behavioral indicators, calculate customer lifetime value, and use demographics to improve churn predictions.

This is a must-read book for anyone who wants to improve or start on analyzing churn. I'm an engineer and the book provided me with just what I needed - a solid foundation and discussion of techniques, algorithms and code samples for analyzing subscription data and patterns. All the examples and instructions in the book are well written, clearly annotated and easy to understand. I really appreciated the many war stories and examples of how to avoid pitfalls and incorrect assumptions.

Even though Carl Gold didn't write FIGHT CHURN WITH DATA for strategists like me (it's full of code and data science) I'm glad I got my hands on it. Chapter 1 is a goldmine (datamine) of frameworks, definitions and cultural guidance.As Tien Tzuo says in the foreword, Chapter 1 should be mandatory reading for anyone who's interested in running a successful subscription-based businesses.If you're a data scientist involved in fighting the good fight against churn, this book is for you.

Fighting churn is hard (really hard!) and the author does a wonderful job equipping readers with the tools to understand and fight churn in a rigorous and data-driven way. A comprehensive set of topics are covered with practical uses that will likely be valuable to anyone interested in boosting their company's retention. Highly recommend!

Carl Gold is one of those rare figures that can be said to have "written the book on churn analysis" both figuratively and literally. True Story.This book is, by all measures, a veritable jewel in the body of literature for a field that has been long suffering from lazy exposition, unoriginal approaches, shallow techniques, and bloated texts.Does the author resort to using the same tired benchmark datasets to carry out his analyses? Nope. He not only provides you with a completely novel, realistic, simulated social network dataset, but he also provides all of the code, schemas, wrappers, etc... so you can create these on your own to follow along, or adapt them to your own production use-cases. This alone is worth many times the price of the text.Chapter One alone is by far the best opening chapter in any data science text available to date, and trust me, I read a lot of them... If you are not convinced by the time you are done with this chapter, then by all means, stop here and go no further.The techniques go from the very basic to the advanced: data analysis, metric creation/feature engineering, loading matrices, dataset creation, variable grouping, cohort analysis, forecasting, and more.So while you think you're just reading a book on "churn analysis", you're getting much more. It is essentially, a textbook on data science techniques and best practices applied to churn events. Substitute churn with some other event, and you can pretty much extend the ideas in here to just about any use-case you'll come across. It's Gold.I can't applaud the author enough for his intellectual generosity in capturing, synthesizing, and sharing his decades of experience in this field for the benefit of the rest of us. I can't wait to see what he comes up with next!Highest Possible Recommendation - Must Read for all data scientists!

This is a solid and practical book, focused on increasing customer retention. Probably there is too much SQL and Python code, that might be hard to read in the Kindle version.The most valuable part for me was the overall framework - how to approach the churn problem, what metrics do you need, when to use the advanced analytics, and when a simple analysis can suffice. The book highlights the importance of explainability, and the need to communicate the results to business people in plain English. I am a data analyst moving from healthcare to B2C, and the book was an invaluable resource for me. The author is generous in sharing his wisdom and knowledge and providing pragmatic, actionable advice.