PDF Fighting Churn with Data 1st Edition Carl Gold download
PDF Fighting Churn with Data 1st Edition Carl Gold download
PDF Fighting Churn with Data 1st Edition Carl Gold download
com
https://textbookfull.com/product/fighting-churn-
with-data-1st-edition-carl-gold/
https://textbookfull.com/product/data-analysis-with-small-samples-and-
non-normal-data-nonparametrics-and-other-strategies-carl-f-siebert/
textbookfull.com
https://textbookfull.com/product/information-security-science-
measuring-the-vulnerability-to-data-compromises-1st-edition-carl-
young/
textbookfull.com
https://textbookfull.com/product/outofshapeworthlessloser-1st-edition-
gracie-gold/
textbookfull.com
https://textbookfull.com/product/voices-of-komagata-maru-imperial-
surveillance-and-workers-from-punjab-in-bengal-1st-edition-suchetana-
chattopadhyay/
textbookfull.com
Chinese FDI in the EU and the US: Simple Rules for
Turbulent Times Tim Wenniges
https://textbookfull.com/product/chinese-fdi-in-the-eu-and-the-us-
simple-rules-for-turbulent-times-tim-wenniges/
textbookfull.com
https://textbookfull.com/product/interventional-radiology-for-medical-
students-1st-edition-hong-kuan-kok/
textbookfull.com
https://textbookfull.com/product/marijuana-harvest-how-to-maximize-
quality-and-yield-in-your-cannabis-garden-ed-rosenthal/
textbookfull.com
https://textbookfull.com/product/data-management-at-scale-best-
practices-for-enterprise-architecture-1st-edition-piethein-strengholt/
textbookfull.com
Carl Gold
Foreword by Tien Tzuo
MANNING
Churn-reducing strategy Core concepts/customer metrics Book chapters
1. Product improvement • Cohorts of metrics based on product- 3, 5, and 7
• Make more of the best use events identify engaging and
features. disengaging product features.
• Make the best features easy • Identify optimal feature-use ratios.
to find.
3. Pricing and packaging • Unit cost and unit value metrics 6 and 7
• Differentiate pricing to identify customers getting high/low
provide value without value on the product.
discounting.
• Correlations show relationships between
• Monetize valued groups the use of different features/content.
of product features.
CARL GOLD
FOREWORD BY TIEN TZUO
MANNING
SHELTER ISLAND
For online information and ordering of this and other Manning books, please visit
www.manning.com. The publisher offers discounts on this book when ordered in quantity.
For more information, please contact
Special Sales Department
Manning Publications Co.
20 Baldwin Road
PO Box 761
Shelter Island, NY 11964
Email: [email protected]
Many of the designations used by manufacturers and sellers to distinguish their products are
claimed as trademarks. Where those designations appear in the book, and Manning Publications
was aware of a trademark claim, the designations have been printed in initial caps or all caps.
Recognizing the importance of preserving what has been written, it is Manning’s policy to have
the books we publish printed on acid-free paper, and we exert our best efforts to that end.
Recognizing also our responsibility to conserve the resources of our planet, Manning books
are printed on paper that is at least 15 percent recycled and processed without the use of
elemental chlorine.
ISBN: 9781617296529
Printed in the United States of America
brief contents
PART 1 BUILDING YOUR ARSENAL ..............................................1
1 ■ The world of churn 3
2 ■ Measuring churn 39
3 ■ Measuring customers 80
4 ■ Observing renewal and churn 134
v
contents
foreword xv
preface xvii
acknowledgments xix
about this book xxi
about the author xxv
about the cover illustration xxvi
vii
viii CONTENTS
2 Measuring churn
2.1
39
Definition of the churn rate 43
Calculating the churn rate and retention rate 44
The relationship between churn rate and retention rate 45
2.2 Subscription databases 46
2.3 Basic churn calculation: Net retention 48
Net retention calculation 48 ■ SQL net retention calculation 50
Interpreting net retention 53
2.4 Standard account-based churn 56
Standard churn rate definition 56 Outer joins for churn
■
calculation with SQL 66 MRR churn vs. account churn vs. net
■
(retention) churn 68
2.7 Churn rate measurement conversion 70
Survivor analysis (advanced) 70 Churn rate conversions 72
■
3 Measuring customers
3.1 From events to metrics 83
80
metric case study with two metrics 277 Percentage of total metrics
■
with extreme outliers (fat tails) 285 Measuring the time since the
■
8 Forecasting churn
8.1
317
Forecasting churn with a model 318
Probability forecasts with a model 318 Engagement and ■
calculation 325
8.2 Reviewing data preparation 326
8.3 Fitting a churn model 329
Results of logistic regression 330 Logistic regression code 332
■
groups 428
10.4 Churn analysis for date- and numeric-based
demographics 430
10.5 Churn forecasting with demographic data 432
Converting text fields to dummy variables 432 Forecasting churn
■
index 467
foreword
This book is a rarity. Although it’s intended primarily for technically oriented people
with some familiarity with coding and data, it also happens to be lucid, compelling,
and occasionally even (gasp!) funny. The first chapter in particular should be manda-
tory reading for anyone who’s interested in running a successful subscription-based
business. Buy a copy for your boss.
It’s exciting to think about all the different companies that will benefit from the
sharp analysis in these pages. Data folks from all sectors of the global economy, from
streaming-media services to industrial manufacturers, will be paying close attention to
Carl’s book. Today, the whole world runs “as a service”: transportation, education,
media, health care, software, retail, manufacturing, you name it.
All these new digital services are generating vast amounts of data, resulting in a
huge signal-to-noise challenge, which is why this book is so important. I study this
topic for a living, and no one has written such a practical and authoritative guide to
effectively filtering through all that information to reduce churn and keep subscribers
happy. When it comes to running a subscription business, churn rates are a matter of
life and death!
Thousands of entrepreneurs are already deeply familiar with Carl Gold’s work. He
is the author of the Subscription Economy Index, a biannual benchmark study that
reflects the growth metrics of hundreds of subscription companies spread across a
variety of industries. As Zuora’s chief data scientist, Carl works with the most timely
and accurate dataset in the subscription economy. He’s a big part of why Zuora is not
only a successful software company but also a respected thought leader.
xv
xvi FOREWORD
If you’re reading this book, you will soon have the ability to make immediate and
material contributions to the success of your company. But as Carl discusses exten-
sively throughout the book, it’s not enough to do the analysis; you also need to be able
to communicate your results to the business at large.
So by all means, use this book to learn how to conduct the proper analysis, but also
use it to learn how to share, execute, and basically excel at your job. There are exam-
ples and case studies and tips and benchmarks galore. How lucky are we? We get to
work in the early days of the subscription economy, and we get to read the first land-
mark book on churn.
xvii
xviii PREFACE
sequence, and some of these tasks are nontrivial. Also, there is a need to “rinse and
repeat” the process multiple times. That kind of workflow is well suited to short pro-
grams but difficult in spreadsheets and graphical tools.
Because the book is written for a data person, it does not go into details on the
churn-reducing actions that products and services can take. So this book does not
contain details on how to do things like run email and call campaigns, create churn-
save playbooks, and design pricing and packaging. Instead, this book is strategic in
that it teaches a data-driven approach to devising your battle plan against churn: pick-
ing which churn-reducing activities to pursue, which customers to target, and what
kinds of results to expect. That said, I will introduce various churn-reducing tactics at
a high level as is necessary to understand the context for using the data.
Visit https://textbookfull.com
now to explore a rich
collection of eBooks, textbook
and enjoy exciting offers!
acknowledgments
There are many people without whom it would not have been possible for me to cre-
ate this book for you.
Starting at the beginning, I thank Ben Rigby for bringing me to my first churn case
study and everyone who worked at Sparked (Chris Purvis, Chris Mielke, Cody Chapman,
Collin Wu, David Nevin, Jamie Doornboss, Jeff Nickerson, Jordan Snodgrass, Joseph
Pigato, Mark Nelson, Morag Scrimgoeur, Rabih Saliba, and Val Ornay) and all the cus-
tomers of Retention Radar. Next, I have Tien Tzuo and Marc Aronson to thank for
bringing me to Zuora, and thanks to Tom Krackeler, Karl Goldstein, and everyone
from Frontleaf (Amanda Olsen, Greg McGuire, Marcelo Rinesi, and Rachel English)
for welcoming me to their team. Continuing in chronological order, I also thank
everyone who worked on or with the Zuora Insights team (Azucena Araujo, Caleb
Saunders, Gail Jimenez, Jessica Hartog, Kevin Postlewaite, Kevin Suer, Matt Darrow,
Michael Lawson, Patrick Kelly, Pushkala Pattabhiraman, Shalaka Sindkar, and Steve
Lotito), the data scientist on my team who worked on churn (Dashiell Stander), and
all the Zuora Insights customers. All these people were part of the projects on which I
learned what I now know about churn; in that way, they made it possible for me to
write this book for you. And I want to thank everyone at Zuora who either helped pro-
mote or edit the book: Amy Konary, Gabe Weisert, Helena Zhang, Jayne Gonzalez,
Kasey Radley, Lauren Glish, Peishan Li, and Sierra Dowling.
Next comes my publisher, Manning, where I thank my first acquisitions editor,
Stephen Soehnlen, for bringing me on board; my main development editor, Toni
Arritola, and my temporary DE, Becky Whitney, for patiently teaching me how to write
xix
xx ACKNOWLEDGMENTS
a Manning-style book; and my second AE, Michael Stephens, for getting the book
across the finish line. I also thank my technical and code editors—Mike Shepard,
Charles Feduke, and Al Krinkler—and everyone who commented on the liveBook
forum during the early access period. My thanks also go to Deirdre Hiam, my project
editor; Pamela Hunt, my proofreader; and Frances Buran, Tiffany Taylor, and Keir
Simpson, my copyeditors. I would also like to thank all the reviewers: Aditya Kaushik,
Al Krinker, Alex Saez, Amlan Chatterjee, Burhan Ul Haq, Emanuele Piccinelli,
George Thomas, Graham Wheeler, Jasmine Alkin, Julien Pohie, Kelum Senanayake,
Lalit Narayana Surampudi, Malgorzata Rodacka, Michael Jensen, Milorad Imbra,
Nahid Alam, Obiamaka Agbaneje, Prabhuti Prakash, Raushan Jha, Simone Sguazza,
Stefano Ongarello, Stijn Vanderlooy, Tiklu Ganguly, Vaughn DiMarco, and Vijay
Kodam. Your suggestions helped make this book better.
Special thanks go to the three companies that allowed me to present a selection of
their case study data to bring the material in the book to life: Matt Baker and everyone
at Broadly; Yan Kong and everyone at Klipfolio; and Jonathan Moody, Tyler Cooper,
and everyone at Versature.
Finally, I thank my wife, Anna, and children, Clive and Skylar, for their support and
patience during a challenging but fruitful time.
about this book
This book was written to enable anyone with a little background in coding and data to
make a game-changing analysis of customer churn for an online product or service.
And if you are experienced in programming and data analysis, the book contains tips
and tricks for churn and customer engagement that you won’t find anywhere else.
xxi
xxii ABOUT THIS BOOK
book explains the data typically available for analyzing churn, the practices used to
turn that data into actionable intelligence, and the most typical findings. One empha-
sis of the book is how to communicate data results to businesspeople; consequently, all
the important takeaways are explained in plain English (no jargon!). So if you care about
churn but aren’t a coder, you should skim the book for the takeaways (clearly labeled)
and skip the coding and math. Then share the book with one of your developers to
get help putting the concepts into action.
Part 2, which contains the core techniques in the book, is devoted to understanding
how customer behavior relates to churn and retention and using that knowledge to
drive churn-reducing strategies:
■ Chapter 5 teaches a form of cohort analysis, which is the primary method for
understanding and explaining the relationship between behaviors and churn.
Chapter 5 also includes many case study examples, and the code is in Python.
■ Chapter 6 looks at how to deal with data that is big in an undesirable way: most
company datasets have closely related measurements of the same underlying
behavior. How you deal with this somewhat-redundant information is important.
■ Chapter 7 returns to the subject of metric creation and uses the information
from chapters 5 and 6 to design advanced metrics, which help explain complex
customer behaviors such as price sensitivity and efficiency.
Part 3 covers forecasting with regression and machine learning. When it comes to
reducing churn, forecasting is less important than having a good set of metrics, but it
can still be useful, and some special techniques are needed to get it right:
■ Chapter 8 teaches how to forecast customer churn probabilities with a regres-
sion and how to interpret the results of those forecasts, including calculating
customer lifetime value.
■ Chapter 9 is about machine learning and measuring and optimizing the accu-
racy of churn forecasts.
■ Chapter 10 covers analyzing demographic or firmographic data in the context
of churn and finding lookalikes for your best customers.
Most readers should start at the beginning and read parts 1 and 2. If, after learning
and applying those techniques, you need to make forecasts or find lookalike custom-
ers, continue to part 3. If you are already using advanced analytics, you may be able to
skip part 1 and start in part 2 and/or 3. For purposes of this book, being advanced in
analytics means that you already have a good set of customer metrics and can identify
and measure churned customers. Otherwise, start with part 1.
Author: J. U. Giesy
Language: English