Optimizing Data-to-Learning-to-Action

201 Pages·2018·3.529 MB·English

by  Steven Flinn

Optimizing
Data-to-Learning-
to-Action
The Modern Approach to Continuous
Performance Improvement for
Businesses
—
Steven Flinn
OPTIMIZING
DATA-TO-LEARNING-
TO-ACTION
THE MODERN APPROACH
TO CONTINUOUS PERFORMANCE
IMPROVEMENT FOR BUSINESSES
Steven Flinn
Optimizing Data-to-Learning-to-Action: The Modern Approach to Continuous
Performance Improvement for Businesses
Steven Flinn
Brenham, Texas, USA
ISBN-13 (pbk): 978-1-4842-3530-0 ISBN-13 (electronic): 978-1-4842-3531-7
https://doi.org/10.1007/978-1-4842-3531-7
Library of Congress Control Number: 2018939149
Copyright © 2018 by Steven Flinn
This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part
of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations,
recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission
or information storage and retrieval, electronic adaptation, computer software, or by similar or
dissimilar methodology now known or hereafter developed.
Trademarked names, logos, and images may appear in this book. Rather than use a trademark
symbol with every occurrence of a trademarked name, logo, or image we use the names, logos, and
images only in an editorial fashion and to the benefit of the trademark owner, with no intention of
infringement of the trademark.
The use in this publication of trade names, trademarks, service marks, and similar terms, even if
they are not identified as such, is not to be taken as an expression of opinion as to whether or not
they are subject to proprietary rights.
While the advice and information in this book are believed to be true and accurate at the
date of publication, neither the authors nor the editors nor the publisher can accept any legal
responsibility for any errors or omissions that may be made. The publisher makes no warranty,
express or implied, with respect to the material contained herein.
Managing Director, Apress Media LLC: Welmoed Spahr
Acquisitions Editor: Susan McDermott
Development Editor: Laura Berendson
Coordinating Editor: Rita Fernando
Cover designed by eStudioCalamar
Cover image designed by Freepik (www.freepik.com)
Distributed to the book trade worldwide by Springer Science+Business Media New York, 233
Spring Street, 6th Floor, New York, NY 10013. Phone 1-800-SPRINGER, fax [phone],
email [e-mail], or visit www.springeronline.com. Apress Media, LLC is a
California LLC and the sole member (owner) is Springer Science + Business Media Finance Inc
(SSBM Finance Inc). SSBM Finance Inc is a Delaware corporation.
For information on translations, please email [e-mail], or visit http://www.apress.
com/rights-permissions.
Apress titles may be purchased in bulk for academic, corporate, or promotional use. eBook versions
and licenses are also available for most titles. For more information, reference our Print and eBook
Bulk Sales web page at http://www.apress.com/bulk-sales.
Any source code or other supplementary material referenced by the author in this book is available
to readers on GitHub via the book’s product page, located at www.apress.com/9781484235300.
For more detailed information, please visit http://www.apress.com/source-code.
Printed on acid-free paper
To my mother and father
Contents
About the Author                                            vii
About the Technical Reviewer                                   ix
Acknowledgments                                             xi
Introduction                                                 xiii
Chapter 1: Case for Action                                    1
Chapter 2: Roots of a New Approach                          17
Chapter 3: Data-to-Learning-to-Action                         29
Chapter 4: Tech Stuff and Where It Fits                        49
Chapter 5: Reversing the Flow: Decision-to-Data                 61
Chapter 6: Quantifying the Value                              79
Chapter 7: Total Value                                      107
Chapter 8: Optimizing Learning Throughput                   121
Chapter 9: Patterns of Learning Constraints and Solutions       135
Chapter 10: Organizing for Data-to-Learning-to-Action Success    161
Chapter 11: Conclusion                                      177
Index                                                      187
About the Author
Steven Flinn is founder and CEO of
ManyWorlds, Inc., which is a pioneer of machine
learning–based solutions for enterprises, the
market-leading provider of visual UX software
for collaborative systems, and a provider of
related advisory services to leading organiza-
tions around the world. Mr. Flinn has extensive
consulting experience at the intersection of
strategy, decision science, and technology with
Global 1000 enterprises, as well as with selected
high-impact startups. He has been awarded over
40 patents in the field of machine learning and
its applications and is the author of The Learning
Layer (Palgrave Macmillan, 2010), which pre-
dicted, and established the imperative for, apply-
ing machine learning–based capabilities in the enterprise, an imperative that
is now widely accepted and a reality. Prior to ManyWorlds, he was a chief
information officer and vice president of strategy at Royal Dutch Shell. His
education includes graduate degrees from Northwestern University’s Kellogg
School of Business and Stanford University’s School of Engineering.
About the Technical
Reviewer
Sébastien Caron is the founder of Conova
Solutions, a management consultancy and stra-
tegic advisory firm that helps organizations
leveraging the digital workplace to attract talent
and build in their DNA the capabilities required
to address increasing complexity and innovate
continuously. His expertise covers approaches
and models such as co-innovation, internal lean
startup, social business, Agile, Cynefin model,
SECI model, capability model, maturity model,
and design thinking. He has worked with organi-
zations such as CGI, Loto-Québec, and recently
with the National Bank of Canada to help
them design and implement an insight-driven
organization (enterprise knowledge graph,
machine learning, analytics, personalized recommendations, and virtual assis-
tant). As a PhD researcher in cognitive and computer science, his expertise
has been recognized by many awards: winner of the Cognitive Informatics
Symposium contest in Montreal (2011 & 2004), winner of the International
Scientific Competition of Ile de France Regional Council (2005), winner of
the International Competition for Researchers in Information Science &
Computing Technology (2005), and more. He likes doing robot fights with his
kids, playing soccer during the summer, and killing zombies in virtual reality
games with his friends.
Acknowledgments
This book synthesizes concepts that I have in recent years developed and put
into practice that are, in turn, based upon experiences and learnings spanning
a good number of years before. For example, I am grateful for obtaining an
excellent grounding in mathematics and economics as an undergraduate at
Binghamton University. I was fortunate to have benefitted from Dr. Ronald
Howard’s pioneering work in the field of decision analysis during my graduate
studies in Stanford University’s Engineering-Economic Systems Department.
I was also fortunate to attend Northwestern University’s Kellogg School of
Management and benefit from, among other things, a deep dive into strategy
and finance. And at Royal Dutch Shell I had the opportunity to experience
the excitement and challenges of leading information technology at the global-
enterprise level. Finally, at my current company, ManyWorlds, I have been able
to pioneer machine-learning applications for enterprises, as well as to advise,
and learn from, a remarkable mix of leading businesses on a variety of subjects
at the intersection of strategy, decision science, and information technology.
It seems as if this unique melange of opportunities and experiences was
necessary to make the method outlined in this book fully come together. I very
much doubt it could have happened otherwise.
In making the book a published reality, I’d like to thank the wonderful editors
at Apress, particularly Susan McDermott and Rita Fernando, for believing in
the work and keeping everything on track. Thanks to Naomi Moneypenny for
pioneering with me at ManyWorlds many of the concepts that are detailed in
the book and providing early guidance on the manuscript. And a very special
thanks to Sébastien Caron, who agreed to be the technical reviewer for the
book (without knowing exactly what he was getting into!), helping me to see
some of my blind spots and making the book better than it would otherwise
be. Of course, any deficiencies of the book remain solely my responsibility.
And finally, thanks to my family for being understanding about the commitment
(which is always greater than expected!) that writing a book requires.
Introduction
We live in interesting times. On the one hand, technology-based advances are
occurring at a bewildering pace. But on the other hand, a careful examination
of macro-level data suggests that U.S. business performance continues to flag
compared to historical results. And it seems that our current approaches to
improving business performance remain stuck in the last century. If the tech-
nology references are stripped out, one would be hard-pressed to distinguish
the popularly prescribed methods in today's management publications from
those of a decade or two ago. Really, not much seems to be new in that regard
since at least the 1990s.
This book aims to challenge this business-performance somnolence by putting
forward a new approach to performance improvement that is relevant for
today’s dynamic environment and promises to be robust enough to continue
to stay relevant. The design goals of this new approach are as follows:
• Provide a clear and effective method for continuous per-
formance improvement for today’s organizations
• Demonstrate how to systematically get the most out of
people, process, and technology investments, and do so
on a continuing basis
• Be widely applicable to different business sectors and
business functions
So, What Is It?
This new approach begins with a recognition that the key to improved busi-
ness performance in today’s world is improved decisions. That really has always
been the case, but this reality has often been skirted past or treated in a
tangential way by the continuous stream of popular management memes.
And the data certainly backs up the intuition that if you improve decisions
you improve financial performance. Bain & Company, for example, found that
“decision effectiveness and financial results correlated at a 95% confidence
level or higher for every country, industry, and company size.”1
1Blenko, Marcia, Michael Mankins, and Paul Rogers, “The Decision-Driven Organization”
Harvard Business Review, June 2010. https://hbr.org/2010/06/the-decision-driven-
organization
xiv Introduction
And not all decisions are created equal—it is the decisions that are aligned
with the most important drivers of value for an organization that really make
a difference, and those drivers may not always be obvious. It always has been
the case, although often not explicitly recognized, that the bigger impact with
respect to the long-term value of, for example, a manufacturing facility, is not
necessarily the way it is operated at any given time. Rather, it is more likely
the decisions on whether it should be built at all, how it should be configured,
where it should be located, what logistics should be developed to support it,
and so on, that will dictate the enduring performance legacy.
The field of decision analysis, a prescriptive discipline for improved decision
making that has evolved over the past half century, can classically be applied
to help with big, one-off, decisions such as making an acquisition or whether
and where to build a plant. But it has been somewhat awkward to apply full-
blown decision analyses to those myriad recurring decisions that are at the
heart of what organizations do on a day-to-day basis. The approach presented
here uniquely adapts key aspects of decision analysis and integrates them
with techniques from other fields of management science so that they can be
beneficially applied to all types of decisions that are made in organizations, and
at all levels of the organization.
By putting decisions front and center, we can more readily dispense with the
fuzziness in thinking that often comes along for the ride with the viral popular-
ity of the latest technology and management memes. For example, take that
field of exploding popularity, data science. The question to consider for any
set of “big data” or associated data analysis is, “What decisions can potentially
be improved, and what is the financial upside for that improvement?” Those
new machine learning–based capabilities that are being breathlessly touted?
Same question. A new system to better facilitate collaboration? Same ques-
tion. A new marketing process? Same question. You get the picture. It seems
so simple, and yet these questions are often not asked, and if they are, they are
not answered in any rigorous way.
The second foundational element this new approach emphasizes is learning, in
its most robust meaning and application. It is subordinate to decisions in that
learning—at least in the business sense, as perhaps opposed to our everyday
life—only has value insofar that it has the potential to change a decision that would
otherwise occur, a decision that in turn would change an action that would
otherwise occur. And the corollary is that data, and its more filtered and
organized derivatives, information and knowledge, only have value insofar as
they enhance learning. This is an important concept: there is no intrinsic value
to data (or information or knowledge) unless it serves as a basis for enhanced
learning or has the potential to do so. Classic decision analysis has a concept
of “value of information”—and we will discuss and apply the concept in this
book—but the label is somewhat of a misnomer. What that notion really con-
notes is value of learning, and data, information, and knowledge only have value
to the extent that they are derived from this value of learning.