33,99 €
Mehr erfahren.
- Herausgeber: John Wiley & Sons
- Kategorie: Wissenschaft und neue Technologien
- Serie: SAS Institute Inc
- Sprache: Englisch
An evidence-based organizational framework for exceptional analytics team results The Analytics Lifecycle Toolkit provides managers with a practical manual for integrating data management and analytic technologies into their organization. Author Gregory Nelson has encountered hundreds of unique perspectives on analytics optimization from across industries; over the years, successful strategies have proven to share certain practices, skillsets, expertise, and structural traits. In this book, he details the concepts, people and processes that contribute to exemplary results, and shares an organizational framework for analytics team functions and roles. By merging analytic culture with data and technology strategies, this framework creates understanding for analytics leaders and a toolbox for practitioners. Focused on team effectiveness and the design thinking surrounding product creation, the framework is illustrated by real-world case studies to show how effective analytics team leadership works on the ground. Tools and templates include best practices for process improvement, workforce enablement, and leadership support, while guidance includes both conceptual discussion of the analytics life cycle and detailed process descriptions. Readers will be equipped to: * Master fundamental concepts and practices of the analytics life cycle * Understand the knowledge domains and best practices for each stage * Delve into the details of analytical team processes and process optimization * Utilize a robust toolkit designed to support analytic team effectiveness The analytics life cycle includes a diverse set of considerations involving the people, processes, culture, data, and technology, and managers needing stellar analytics performance must understand their unique role in the process of winnowing the big picture down to meaningful action. The Analytics Lifecycle Toolkit provides expert perspective and much-needed insight to managers, while providing practitioners with a new set of tools for optimizing results.
Sie lesen das E-Book in den Legimi-Apps auf:
Seitenzahl: 515
Veröffentlichungsjahr: 2018
Ähnliche
Table of Contents
Cover
Title Page
Preface
Acknowledgments
REFERENCES
PART I: The Foundation of Analytics
CHAPTER 1: Analytics Overview
FUNDAMENTAL CONCEPTS
ANALYTICS CONCEPTS
THE METHODS OF ANALYTICS
THE GOAL OF ANALYTICS
CHAPTER SUMMARY
REFERENCES
NOTE
CHAPTER 2: The People of Analytics
WHO DOES ANALYTICS?
ROLES PEOPLE PLAY
CRITICAL COMPETENCIES FOR ANALYTICS
CHAPTER SUMMARY
REFERENCES
NOTES
CHAPTER 3: Organizational Context for Analytics
ORGANIZATIONAL STRATEGY AND ANALYTICS ALIGNMENT
ORGANIZATIONAL CULTURE
ORGANIZATIONAL DESIGN FOR ANALYTICS
WHICH ORGANIZATIONAL DESIGN IS BEST?
CHAPTER SUMMARY
REFERENCES
CHAPTER 4: Data Strategy, Platforms, and Architecture
DATA STRATEGY
STRATEGY DEVELOPMENT PROCESS
DEVELOPING A DATA STRATEGY ROADMAP
AN AGILE APPROACH TO DEVELOPING YOUR DATA STRATEGY
DATA STRATEGY SUMMARY
PLATFORMS AND ARCHITECTURE ANALYTICS
ANALYTICS ARCHITECTURE
DATA FOR PURPOSE OR POTENTIAL
CHAPTER SUMMARY
REFERENCES
PART II: Analytics Lifecycle Best Practices
CHAPTER 5: The Analytics Lifecycle Toolkit
ANALYTICS LIFECYCLE BEST PRACTICE AREAS
ANALYTICS AS A PRODUCT OF DATA SCIENCE
GOALS OF ANALYTICS
SCALE AND SCOPE OF ANALYTICS PRODUCTS
HOW THE ANALYTICS LIFECYCLE TOOLKIT IS ORGANIZED
DESIGN THINKING FOR ANALYTICS
CHAPTER SUMMARY
REFERENCES
CHAPTER 6: Problem Framing
PROCESS OVERVIEW
WHY DO THIS?
PROCESS AREAS
CHAPTER SUMMARY
TOOLKIT SUMMARY
REFERENCES
CHAPTER 7: Data Sensemaking
PROCESS OVERVIEW
PROCESS AREAS
MAINTAINING AN ANALYTICS JOURNAL
CHAPTER SUMMARY
TOOLKIT SUMMARY
REFERENCES
CHAPTER 8: Analytics Model Development
PROCESS OVERVIEW
PROCESS AREAS
MAKING COMPARISONS
MEASURING ASSOCIATIONS
MAKING PREDICTIONS
CHAPTER SUMMARY
PROBLEM SUMMARY AND EXERCISES
TOOLKIT SUMMARY
REFERENCES
CHAPTER 9: Results Activation
PROCESS OVERVIEW
SOLUTION EVALUATION
OPERATIONALIZATION
PRESENTATION AND STORYTELLING
CHAPTER SUMMARY
EXERCISE
TOOLKIT SUMMARY
REFERENCES
NOTE
CHAPTER 10: Analytics Product Management
PROCESS OVERVIEW
PROCESS AREAS
CHAPTER SUMMARY
REFERENCES
PART III: Sustaining Analytics Success
CHAPTER 11: Actioning Analytics
THE POWER OF ANALYTICS
EFFICIENT AND EFFECTIVE ANALYTICS PROGRAMS
WHY OPERATIONALIZATION OF ANALYTICS FAILS
CHANGE MANAGEMENT
BEST PRACTICES FOR LEADING CHANGE
TROUBLESHOOTING CHANGE
CHAPTER SUMMARY
REFERENCES
CHAPTER 12: Core Competencies for Analytics Teams
INTRODUCTION
ANALYTICS COMPETENCIES DETAILED
IDEALIZED COMPETENCIES FOR ANALYTICS JOB FAMILIES BY KNOWLEDGE DOMAIN
CHAPTER SUMMARY
REFERENCE
CHAPTER 13: The Future of Analytics
THE ANALYTICS LIFECYCLE AS A FRAMEWORK
THE ROLE OF ANALYTICS IN OUR FUTURE
FINAL THOUGHTS
REFERENCES
About the Author
About the Companion Web Site
Index
End User License Agreement
List of Tables
Chapter 1
Table 1.1 Styles of machine learning
Chapter 3
Table 3.1 Eight Elements of Organization Design
Chapter 4
Table 4.1 Six Stages for Data Strategy Execution
Chapter 5
Table 5.1 Types of analytics “project types”
Table 5.2 Best practice area: Problem framing
Table 5.3 Best practice area: Data sensemaking
Table 5.4 Best practice area: Analytics model development
Table 5.5 Best practice area: Results activation
Table 5.6 Best practice area: Analytics Product Lifecycle management
Chapter 6
Table 6.1 Types of diagrams useful in identifying potential cause and effect relationships
Table 6.2 Design thinking tools useful in analytics problem framing
Table 6.3 Problem Statement Table
Table 6.4 Project prioritization matrix
Chapter 7
Table 7.1 Measures important while examining the distributions of quantitative variables
Chapter 8
Table 8.1 Major issues in achieving analytics maturity
Table 8.2 Model development best process areas, common questions, and common methods
Chapter 10
Table 10.1 Objective and subjective measures for calculating return on investment
Chapter 11
Table 11.1 Differences between effectiveness and efficiency in analytics
Table 11.2 Stakeholder perspectives on analytics effectiveness and efficiency
Table 11.3 Change management best practice areas for analytics
Table 11.4 Examples of analytics impacting change
Table 11.5 Examples of communications goals by change curve
List of Illustrations
Chapter 1
Figure 1.1 BI dashboard
Figure 1.2 The difference between health information management, health IT, and informatics
Figure 1.3 The relationship between statistics and other quantitative disciplines
Figure 1.4 Inductive reasoning compared to deductive reasoning
Figure 1.5 Approaches to forecasting and time series analysis
Figure 1.6 Natural language processing conceptual workflow
Figure 1.7 Techniques found in machines learning
Chapter 2
Figure 2.1 The analytics lifecycle supports the decision lifecycle.
Figure 2.2 Using data to support decisions includes a variety of roles
Figure 2.3 Five job families for analytics
Figure 2.4 Competencies for analytics vary by job family.
Figure 2.5 The knowledge domains for analytics. (The Analytics Competency Model)
Figure 2.6 The elusive data scientist combines competencies in technology, data, business domain, and methods.
Figure 2.7 Key competencies needed across analytics roles
Figure 2.8 Decomposition diagram for why a car won't start
Figure 2.9 Analysis and synthesis are two processes used in problem solving.
Figure 2.10 Analytics innovation planning model
Figure 2.11 Cultural alignment—patient-centered care
Chapter 3
Figure 3.1 Analytics alignment: relationship between strategy, organizational capabilities, resources, and management systems
Figure 3.2 Data pipeline supports the Analytics Lifecycle
Figure 3.3 Culture and competencies of the modern analytics organization
Figure 3.4 Analytics culture and readiness rests on five dimensions
Figure 3.5 Organizational design is a function of structure, people, and processes
Figure 3.6 Galbraith's Star Model for the design of organizations
Figure 3.7 Centralized organizational structure for analytics
Figure 3.8 Decentralized organizational structure for analytics
Figure 3.9 Center of excellence organizational structure for analytics
Figure 3.10 The analytics concierge model
Figure 3.11 Staff configuration in the analytics concierge model
Chapter 4
Figure 4.1 Relationship between corporate strategy and the analytics function
Figure 4.2 Analytics strategy development process
Figure 4.3 Relationship between the data pipeline and the Analytics Lifecycle
Figure 4.4 Analytics maturity assessment pyramid
Figure 4.5 Percent of problems by differing complexities
Figure 4.6 The dimensions of crisis analytics
Chapter 5
Figure 5.1 Analytics Lifecycle
Figure 5.2 The motivation for analytics
Figure 5.3 The five best practice areas in the Analytics Lifecycle Toolkit
Figure 5.4 Five steps in design thinking
Figure 5.5 Adaptation of various toolkits (such as design thinking, product thinking, and agile methods) can enable better analytics solutions
Figure 5.6 Design thinking processes applied to analytics
Chapter 6
Figure 6.1 Problem framing best practice area in context
Figure 6.2 Problem framing relates to the empathy stage in design thinking
Figure 6.3 Affinity diagram
Figure 6.4 Tree diagram map from root cause analysis (Determining why sales are slow)
Figure 6.5 Divergent vs. convergent thinking in problem framing
Figure 6.6 Three-step process for brainstorming in problem framing (prioritizing software features)
Figure 6.7 Belief statements translated to SMART goals
Figure 6.8 Impact and risk matrix
Figure 6.9 Potential study designs used in applied analytics settings
Figure 6.10 Visual presentation of the relative priorities for a project
Chapter 7
Figure 7.1 Hypothesis-based versus data-driven analysis
Figure 7.2 Data sensemaking focuses on describing, exploring, and explaining
Figure 7.3 Example of Markov microsimulation for the comparison of various treatment strategies
Figure 7.4 In analytics, we often augment analytics with data that lives outside of the traditional data pipeline.
Figure 7.5 The analytics sandbox
Figure 7.6 Types of variables
Figure 7.7 Descriptive analysis options
Figure 7.8 Distribution of quantitative variables
Chapter 8
Figure 8.1 Vectors of analytics maturity
Figure 8.2 Regression model for predicting grades based on study time
Figure 8.3 Plotting the regression line can help visualize the relationships between two variables
Figure 8.4 Cross Industry Standard Process for Data Mining (CRISP-DM)
Figure 8.5 Statistical modeling cultures: Data modeling culture vs. algorithmic modeling culture
Figure 8.6 Five questions for selecting a statistical model
Figure 8.7 Steps used in the evaluation of a hypothesis
Figure 8.8 Example scatter plot to visualize the relationships between two quantitative variables
Figure 8.9 Example of correlation matrix heat map showing the strength of the relationships between variables (Hemedinger, 2013)
Figure 8.10 Example of scatter plot matrix showing the relationships between variables
Figure 8.11 2 × 2 matrix to help classify the type of question being asked
Figure 8.12 2 × 2 matrix outlining common statistical tests for differences and associations
Figure 8.13 2-×-2 matrix of common statistical tests for prediction
Figure 8.14 Scatter plot revealing patterns in the data
Figure 8.15 A depiction of where feature engineering factors into the analytics process
Figure 8.16 Pattern detection process flow for supervised and unsupervised model development
Chapter 9
Figure 9.1 Results activation in the context of design thinking
Figure 9.2 Planning for telling your data story
Figure 9.3 Example of audience transformation plan
Chapter 10
Figure 10.1 Model assembly approach illustrated
Figure 10.2 Well-rounded analytics product manager maintains balance in business, analytics, and product thinking.
Figure 10.3 Product thinking explained
Figure 10.4 Create-measure-learn feedback loop is at the core to the lean startup model and essential to building an MVP for analytics.
Figure 10.5 Visible progress is tracked and shared.
Figure 10.6 Seattle Children's analytics home page
Figure 10.7 Testimonial from satisfied customer at Seattle Children's
Figure 10.8 Measuring return on investment for analytics can come in many forms.
Figure 10.9 Analytics Lifecycle is iterative.
Figure 10.10 Analytics product prioritization matrix
Figure 10.11 Analytics products must balance the user experience with the value.
Figure 10.12 Structure-process-outcome model for analytics quality
Figure 10.13 Example measures for quality using structure-process-output model
Figure 10.14 Align talent development to analytics aspirations.
Chapter 11
Figure 11.1 2 × 2 Effectiveness and efficiency matrix
Figure 11.2 The Analytics Lifecycle
Figure 11.3 Common challenges in analytics
Figure 11.4 Dimensions of change
Figure 11.5 Emotional reactions to change
Figure 11.6 Change commitment curve
Figure 11.7 Agility in change can occur by focusing on those with the most influence (positive and negative).
Chapter 12
Figure 12.1 The point at which there is a shift from toolset to mindset
Figure 12.2 Analytics career lattice
Chapter 13
Figure 13.1 The linkage between analytics strategy and execution
Guide
Cover
Table of Contents
Begin Reading
Pages
C1
ii
iii
iv
v
vi
viii
xi
xii
xiii
xiv
xv
xvi
1
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
127
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
349
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
437
438
439
440
441
442
443
444
445
446
447
448
E1
Wiley & SAS Business Series
The Wiley & SAS Business Series presents books that help senior-level managers with their critical management decisions.
Titles in the Wiley & SAS Business Series include:
The Analytic Hospitality Executive
by Kelly A. McGuire
The Analytics Lifecycle Toolkit: A Practical Guide for an Effective Analytics Capability
by Gregory S. Nelson
Analytics: The Agile Way
by Phil Simon
Analytics in a Big Data World: The Essential Guide to Data Science and Its Applications by
Bart Baesens
Bank Fraud: Using Technology to Combat Losses
by Revathi Subramanian
Big Data Analytics: Turning Big Data into Big Money
by Frank Ohlhorst
Big Data, Big Innovation: Enabling Competitive Differentiation through Business Analytics
by Evan Stubbs
Business Analytics for Customer Intelligence
by Gert Laursen
Business Intelligence Applied: Implementing an Effective Information and Communications Technology Infrastructure
by Michael Gendron
Business Intelligence and the Cloud: Strategic Implementation Guide
by Michael S. Gendron
Business Transformation: A Roadmap for Maximizing Organizational Insights
by Aiman Zeid
Connecting Organizational Silos: Taking Knowledge Flow Management to the Next Level with Social Media
by Frank Leistner
Data-Driven Healthcare: How Analytics and BI Are Transforming the Industry
by Laura Madsen
Delivering Business Analytics: Practical Guidelines for Best Practice
by Evan Stubbs
Demand-Driven Forecasting: A Structured Approach to Forecasting, Second Edition
by Charles Chase
Demand-Driven Inventory Optimization and Replenishment: Creating a More Efficient Supply Chain
by Robert A. Davis
Developing Human Capital: Using Analytics to Plan and Optimize Your Learning and Development Investments
by Gene Pease, Barbara Beresford, and Lew Walker
The Executive's Guide to Enterprise Social Media Strategy: How Social Networks Are Radically Transforming Your Business
by David Thomas and Mike Barlow
Economic and Business Forecasting: Analyzing and Interpreting Econometric Results
by John Silvia, Azhar Iqbal, Kaylyn Swankoski, Sarah Watt, and Sam Bullard
Economic Modeling in the Post Great Recession Era: Incomplete Data, Imperfect Markets
by John Silvia, Azhar Iqbal, and Sarah Watt House
Enhance Oil & Gas Exploration with Data Driven Geophysical and Petrophysical Models
by Keith Holdaway and Duncan Irving
Foreign Currency Financial Reporting from Euros to Yen to Yuan: A Guide to Fundamental Concepts and Practical Applications
by Robert Rowan
Harness Oil and Gas Big Data with Analytics: Optimize Exploration and Production with Data Driven Models
by Keith Holdaway
Health Analytics: Gaining the Insights to Transform Health Care
by Jason Burke
Heuristics in Analytics: A Practical Perspective of What Influences Our Analytical World
by Carlos Andre Reis Pinheiro and Fiona McNeill
Human Capital Analytics: How to Harness the Potential of Your Organization's Greatest Asset
by Gene Pease, Boyce Byerly, and Jac Fitz-enz
Implement, Improve, and Expand Your Statewide Longitudinal Data System: Creating a Culture of Data in Education
by Jamie McQuiggan and Armistead Sapp
Intelligent Credit Scoring: Building and Implementing Better Credit Risk Scorecards, Second Edition
by Naeem Siddiqi
JMP Connections
by John Wubbel
Killer Analytics: Top 20 Metrics Missing from Your Balance Sheet
by Mark Brown
Machine Learning for Marketers: Hold the Math
by Jim Sterne
On-Camera Coach: Tools and Techniques for Business Professionals in a Video-Driven World
by Karin Reed
A Practical Guide to Analytics for Governments: Using Big Data for Good
by Marie Lowman
Predictive Analytics for Human Resources
by Jac Fitz-enz and John Mattox II
Predictive Business Analytics: Forward-Looking Capabilities to Improve Business Performance
by Lawrence Maisel and Gary Cokins
Profit Driven Business Analytics: A Practitioner's Guide to Transforming Big Data into Added Value
by Wouter Verbeke, Cristian Bravo, and Bart Baesens
Retail Analytics: The Secret Weapon
by Emmett Cox
Social Network Analysis in Telecommunications
by Carlos Andre Reis Pinheiro
Statistical Thinking: Improving Business Performance, Second Edition
by Roger W. Hoerl and Ronald D. Snee
Strategies in Biomedical Data Science: Driving Force for Innovation
by Jay Etchings
Style & Statistics: The Art of Retail Analytics
by Brittany Bullard
Taming the Big Data Tidal Wave: Finding Opportunities in Huge Data Streams with Advanced Analytics
by Bill Franks
Too Big to Ignore: The Business Case for Big Data
by Phil Simon
Using Big Data Analytics: Turning Big Data into Big Money
by Jared Dean
The Value of Business Analytics: Identifying the Path to Profitability
by Evan Stubbs
The Visual Organization: Data Visualization, Big Data, and the Quest for Better Decisions
by Phil Simon
Win with Advanced Business Analytics: Creating Business Value from Your Data
by Jean Paul Isson and Jesse Harriott
For more information on any of the above titles, please visit www.wiley.com.
The Analytics Lifecycle Toolkit
A Practical Guide for an Effective Analytics Capability
Gregory S. Nelson
Copyright © 2018 by John Wiley & Sons, Inc. All rights reserved.
Published by John Wiley & Sons, Inc., Hoboken, New Jersey.
Published simultaneously in Canada.
No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, scanning, or otherwise, except as permitted under Section 107 or 108 of the 1976 United States Copyright Act, without either the prior written permission of the Publisher, or authorization through payment of the appropriate per-copy fee to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, (978) 750-8400, fax (978) 646-8600, or on the Web at www.copyright.com. Requests to the Publisher for permission should be addressed to the Permissions Department, John Wiley & Sons, Inc., 111 River Street, Hoboken, NJ 07030, (201) 748-6011, fax (201) 748-6008, or online at www.wiley.com/go/permissions.
Limit of Liability/Disclaimer of Warranty: While the publisher and author have used their best efforts in preparing this book, they make no representations or warranties with respect to the accuracy or completeness of the contents of this book and specifically disclaim any implied warranties of merchantability or fitness for a particular purpose. No warranty may be created or extended by sales representatives or written sales materials. The advice and strategies contained herein may not be suitable for your situation. You should consult with a professional where appropriate. Neither the publisher nor author shall be liable for any loss of profit or any other commercial damages, including but not limited to special, incidental, consequential, or other damages.
For general information on our other products and services or for technical support, please contact our Customer Care Department within the United States at (800) 762-2974, outside the United States at (317) 572-3993, or fax (317) 572-4002.
Wiley publishes in a variety of print and electronic formats and by print-on-demand. Some material included with standard print versions of this book may not be included in e-books or in print-on-demand. If this book refers to media such as a CD or DVD that is not included in the version you purchased, you may download this material at http://booksupport.wiley.com. For more information about Wiley products, visit www.wiley.com.
Library of Congress Cataloging-in-Publication Data is Available:
ISBN 978-1-119- 42506-9 (Hardcover)
ISBN 978-1-119-42509-0 (ePDF)
ISBN 978-1-119-42510-6 (ePub)
Cover Design: Wiley
Cover Image: © mattjeacock/Getty Images
To Nick and MaryLu, for showing me what it means to be a part of something bigger than yourself.
Preface
The modern enterprise is often characterized as “data rich, but information poor.” This challenge is exacerbated by the pure volume and variety of data generated at the point of interaction (e.g., customers, patients, suppliers) and careening outward. Whether you are preparing, analyzing, presenting, or consuming data, having a strong foundation in data and analytics is paramount for conveying ideas effectively.
In this book, I translate the world of big data, data science, and analytics into a practical, comprehensive guide where you can explore the art and science of analytics best practices through a proven framework for managing analytics teams and processes.
The focus of the book is on creating effective and efficient analytics organizations and processes in order to strengthen the role of data and analytics in producing organizational success.
When I started thinking about writing about this specific topic, it was primarily in response to the lack of information about “the people and process” side of analytics. That is, for over a decade, authors have written about the concept of analytics, its importance in business, and specific implementations of technologies such as Python, R, or SAS, among others. However, those resources generally do not address the tactics of analytics model development or business case development, nor do they address the impact of analytics on operational processes.
The issues that organizations have grappled with over the past 10 years since Tom Davenport and Jeanne Harris published their seminal work Competing on Analytics (Davenport & Harris, 2007) have shifted from “What problems can we solve with analytics?” to “How do we find, nurture, and retain analytics professionals?” This shift from the “what” to the “how” supports the basic premise of this book. I also think the timing is right for the book, as entire industries are transforming themselves with the use of data and analytics. While many organizations have solved the barriers of effectively using analytics in everyday operations as well as strategic decision making, other industries are just now getting on the “analytics bandwagon,” and they see the promise of analytics without a clear roadmap for getting there. For the former, the challenge is one of effectiveness and improved efficiencies. For the latter, the real struggle can often be with creating an organizational culture—or mindset—for analytics, justifying the development of an analytics capability, and organizing for success.
My personal inspiration for this book came from the works of Ralph Kimball. I remember reading his first edition of the Data Warehouse Toolkit (Kimball, 1996) and thinking to myself, “This makes sense.” It was so very different from the conceptual treatments often found in business and technology books, in that Kimball gave us the language, tools, and processes to actually do data warehousing. He provided a solid overview of the areas relevant to someone who was either familiar with or completely new to data warehousing, along with a framework for the data warehousing lifecycle and key process areas. I hope that you will find that The Analytics Lifecycle Toolkit lives up to this inspiration and that it provides a comprehensive and practical guide to the Analytics Lifecycle with focus on creating an effective analytics capability for your organization.
This book differs from other “how-to” books in that it is not designed as a cookbook of analytics models, but rather, is a primer on the best practices and processes used in analytics. It is intended for:
Organizational leaders and analytics executives
who need to understand what it means to build and maintain an analytics capability and culture, including those in newly minted chief analytics officer or chief data officer positions.
Analytics teams
on the front lines of designing, developing, and delivering analytics as a service or as a product. This group includes analytics product managers, team leads, analysts, project managers, statisticians, scientists, engineers, data scientists, and the “quants” who build analytics models.
Aspiring data champions
, those who use data or consume analytics products in their role as fact-based, problem solvers. The data champion is anyone who wishes to use data to improve performance, support a decision, or change the trajectory of some business process.
This book is organized in three sections:
The Foundation of Analytics:
Starts by outlining what analytics is and how it can be applied to a number of problems in the organization. The focus shifts to analytics as an organizational capability, outlining a different perspective on how analytics can serve the organization's purpose, and how analytics (and data) strategy informs what we do and how we deliver those capabilities. Then this section will address how to deliver analytics capabilities through resources—that is, people, processes, technology, and data.
Analytics Lifecycle Best Practices:
Introduces analytics products and how to support the design, development, and delivery of analytics products and/or services. The lifecycle is then broken down into five best practice areas with specific processes that support analytics product development.
Sustaining Analytics Success:
Rounds out the discussion of how to ensure that analytics products have the greatest impact on the organization and sustain improvements. The discussion includes how to measure effectiveness and efficiency for analytics programs and apply lessons learned from other disciplines such as behavioral economics, social psychology, and change management.
In the first chapter, you will see that the language of analytics can be confusing and even down right daunting. Terms like the science of, the discipline of, and the best practice of generally refer to the usual manner in which analytics are conceptualized.
However, terms like method, methodology, or approach typically mean the processes used in common practice.
One of my goals in writing The Analytics Lifecycle Toolkit is to assume nothing and to clarify things along the way. To that end, I will do my best to make analytics accessible by providing explicit examples and using precise language wherever possible.
You've made it this far, so perhaps you agree that this topic is interesting and worth the price of admission. But if you need 10 more reasons, here they are:
This book:
Offers a practical guide to understanding the complete analytics lifecycle and how to translate that into organizational design and efficient processes.
Provides a framework for building an analytics team in the organization, including functions and team design.
Explores the people and process side of analytics with a focus on analytics team effectiveness and design thinking around the creation of analytics products.
Discusses the analytics job families and roles needed for a successful analytics program.
Includes case studies from real-world experiences.
Bridges concepts appropriate to an analytics culture such as data-centrism and numeracy with data and technology strategies.
Creates understanding and awareness for analytics leaders and a toolbox for practitioners.
Provides access to a library of tools and templates that include areas of best practice that support leadership, process improvement, and workforce enablement.
Begins with fundamentals of the analytics lifecycle, discusses the knowledge domains and best practice areas, and then details the analytics team processes.
Was written by someone who does analytics for a living and has seen hundreds of unique customer perspectives and applications across multiple industries.
Hopefully, this book will provide some useful guidance for those just starting their analytics journey and some tips for those more experienced. Happy trails!
Acknowledgments
This work would not have been possible without the support of my colleagues and clients who gave me the space to write. I am especially indebted to Monica Horvath, PhD, for picking up the pieces I dropped along the way. Not only did she provide scrutiny during technical review of this book, but was my sounding board and co-conspirator for the past several years at ThotWave as we helped clients improve the “people and process side of analytics.” Much of the content around organizational design and our analytics competency model was rooted in these efforts.
I am grateful to all of those with whom I have had the pleasure to work during this project. I learn from each of my clients at ThotWave and my professional colleagues throughout the industry as they continue to teach me a great deal about the real-world implications of analytics and the real struggles that organizations have.
I am indebted to those who agreed to review drafts of this book. In particular, I want to thank Anne Milley from JMP Software; Marc Vaglio-Luarin, analytics product manager from Qlik Software; Linda Burtch, founder of Burtch Works; Mark Tabladillo, lead data scientist from Microsoft; Randy Betancourt from Accenture; Robert Gladden, chief analytics officer at Highmark Health; Mary Beth Ainsworth, product marketing at SAS for artificial intelligence and text analytics; and Teddy Benson from the Walt Disney Company. Your contributions to this work made it a better product.
I would especially like to thank my personal copyeditor, MaryLu Giver. Despite the massive amount of red ink, she was encouraging, thorough, and incredibly kind. In addition, thanks goes to the editorial team at Wiley and, in particular, Julie Kerr, who made the process of publishing a book easy and allowed me to focus on the writing.
Nobody has been more important to me in the pursuit of this project than the members of my family. I would like to thank my family, whose love and guidance are with me in whatever I pursue. They are the ultimate role models. Most importantly, I wish to thank my loving and supportive wife, Susan, who makes me a better person, and my daughter and grandson, who give me hope.
REFERENCES
Davenport, T. H., & Harris, J. G. (2007).
Competing on analytics: the new science of winning
. Boston: Harvard Business School Press.
Kimball, R. (1996).
The data warehouse toolkit: practical techniques for building dimensional data warehouses
. New York: John Wiley & Sons.
PART IThe Foundation of Analytics
CHAPTER 1Analytics Overview
…what enables the wise commander to strike and conquer, and achieve things beyond the reach of ordinary men, is foreknowledge. Now, this foreknowledge cannot be elicited from spirits…
The Art of War, Sun Tzu (as seen in Giles, 1994)
FUNDAMENTAL CONCEPTS
Peter Drucker first spoke of the “knowledge economy” in his book The Age of Discontinuity (Drucker, 1969). The knowledge economy refers to the use of knowledge “to generate tangible and intangible value.” Nearly 50 years later, organizations have virtually transformed themselves to meet this challenge, and data and analytics have become central to that transformation.
In this chapter, we highlight the “fundamentals” of analytics by hopefully creating a level playing field for those interested in the moving from the concept of analytics to the practice of analytics. The fundamentals include defining both data and analytics using terms that I hope resonate. In addition, I think it is important to consider analytics in the wider context of how it is used and the value derived from these efforts. Finally, in this chapter, I relate analytics to other widely used terms as a way to find both common ground and differentiation with often-confused terminologies.
Data
Data permeates just about every part of our lives, from the digital footprint we leave with our cell phones, to health records, purchase history, and utilization of resources such as energy. While not impossible, it would require dedication and uncanny persistence to live “off-the-grid” in this digital world. Beyond the pure generation of data, we are also voracious consumers of data, reviewing our online spending habits, monitoring our fitness regimes, and reviewing those frequent flyer points for that Caribbean vacation.
But what is data? At its most general form, data is simply information that has been stored for later use. Earliest forms of recording information might have been notches on bones (Sack, 2012). Fast forward to the 1950s, and people recorded digital information on Mylar strips (magnetic tape), then punch cards, and later disks. Modern data processing is relatively young but has set the foundation for how we think about the collection, storage, management, and use of information.
Until recently, we cataloged information that wasn't necessarily computable (e.g., videos, images); but through massive technological change, the class of “unstorable” data is quickly vanishing. Stored information, or data, is simply a model of the real world encoded in a manner that is usable, or for our purposes “computable” (Wolfram, 2010).
The fact that data is a persistent record or “model” of something that happened in the real world is an important distinction in analytics. George Box, a statistician considered by many as “one of the greatest statistical minds of the 20th century” (Champkin, 2013) was often quoted as saying: “All models are wrong, but some are useful.” All too often, we find something in the data that doesn't make sense or is just plain wrong. Remember that data has been translated from the real, physical world into something that represents the real world—George's “model.” Just as the mechanical speedometer is a standard for measuring speed (and a pretty good proxy for measuring velocity), the model is really measuring tire rotation, not speed. (For those interested in a late-night distraction, I refer you to Woodford's 2016 article “Speedometers” that explains how speedometers work.) In sum, data is stored information and serves as the foundation for all of analytics. In visual analytics, for example, we make sense out of the data using visualization techniques that enable us to perform analytical reasoning through interactive, visual interfaces.
Analytics
Analytics may be one of the most overused yet least understood terms used in business. For some, it relates to the technologies used to “beat data into submission,” or it is simply an extension of business intelligence and data warehousing. And yet for others, it relates to the statistical, mathematical, or quantitative methods used in the development of models.
According to Merriam-Webster (Merriam-Webster, 2017), analytics is “the method of logical analysis.” Dictionary.com (dictionary.com, 2017) defines analytics as “the science of logical analysis.” Unfortunately, both definitions use the root word of analysis in the definition, which seems a bit like cheating.
The origin of the word analysis goes all the way back to the 1580s, where the term is rooted in Medieval Latin (analticus) and Greek (analtikós), and means to break up or to loosen. Throughout this book, I frame analytics as a structured approach to data-driven problem solving—one that helps us break up problems through careful consideration of the facts.
What Is Analytics?
There has been much debate over the definition of analytics (Rose, 2016). While the purpose of this book is not to redefine or challenge anyone's definition, for the current discussion I define analytics as:
a comprehensive, data-driven strategy for problem solving
I intentionally resist using a definition that views analytics as a “process,” a “science,” or a “discipline.” Instead, I cast analytics as a comprehensive strategy, and as you will see in Part II of this book, it encompasses best practice areas that contain processes, along with roles and deliverables.
Analytics uses logic, inductive and deductive reasoning, critical thinking, and quantitative methods—along with data—to examine phenomena and determine its essential features. Analytics is rooted in the scientific method (Shuttleworth, 2009), including problem identification and understanding, theory generation, hypothesis testing, and the communication of results.
Inductive Reasoning
Inductive reasoning refers to the idea that accumulated evidence is used to support a conclusion but with some level of uncertainty. That is, there is a chance (probability) that the final conclusions may differ from the underling premises. With inductive reasoning, we make broad generalizations from specific observations or data.
Deductive Reasoning
Deductive reasoning on the other hand makes an assertion about some general case and then seeks to prove or disprove it with data (using statistical inference or experimentation). We propose a theory about the way the world works and then test our hypotheses.
We will explore this in more detail later in this chapter.
Analytics can be used to solve big hairy problems such as those faced by UPS that helped them “save more than 1.5 million gallons of fuel and reduced carbon dioxide emissions by 14,000 metric tons” (Schlangenstein, 2013) as well operational problems like optimizing the scheduling of operating rooms for Cleveland Clinic (Schouten, 2013). With success stories like those, it is no wonder that analytics is an attractive bedfellow with technology vendors (hardware and software) and other various proponents. Of course, the danger in the overuse of analytics can be seen in the pairing of the term with other words such as:
Big data analytics
Prescriptive analytics
Business analytics
Operational analytics
Advanced analytics
Real-time analytics
Edge or ambient analytics
While these pairings offer distinctive qualifiers on the type and context to which analytics is applied, it often creates confusion, especially in C-suites, where technology vendors offer the latest analytics solutions to solve their every pain. My perspective (and one that is shared with lots of other like-minded, rational beings) is that analytics is not a technology but that technology serves as an enabler.
Analytics is also often referred to as “any solution that supports the identification of meaningful patterns and relationships among data.” Analytics is used to parse large or small, simple or complex, structured and unstructured, quantitative or qualitative data for the express purposes of understanding, predicting, and optimizing. Advanced analytics is a subset of analytics that uses highly developed and computationally sophisticated techniques with the intent of supporting the fact-based decision process—usually in an automated or semi-automated manner.
Advanced analytics typically incorporates techniques such as data mining, econometrics, forecasting, optimization, predictive modeling, simulations, statistics, and text mining.
How Analytics Differs from Other Concepts
Vincent Granville, who operates Data Science Central, a social network for data scientists, compared 16 analytics disciplines to data science (Granville, 2014). Without repeating those here (but definitely worth the read!), it is useful to highlight the differences between analytics and similar concepts as a way to clarify the meaning of analytics. Here, analytics will be described as it relates to concepts and methods:
Concepts
Business intelligence and reporting
Big data
Data science
Edge (and ambient) analytics
Informatics
Artificial intelligence and cognitive computing
Methods
Applied statistics and mathematics
Forecasting and time series
NLP, text mining, and text analytics
Machine learning and data mining
To start with, it is important to distinguish between concepts and methods.
Concepts
Concepts are generalized constructs that help us understand what something is or how it works.
Methods
Methods, in this context, are the specific techniques or approaches that are used to implement an analytic solution.
Another way to think about this is that methods describe approaches to different types of problems. For example, we might consider something as an optimization problem or a forecasting problem, whereas big data is a mental model that helps us understand the complexity of modern data challenges. Similarly, as we will see later in this chapter, machine learning can be thought of as simply the current state of artificial intelligence—the latter being the concept and the former being the method.
ANALYTICS CONCEPTS
An analytics concept can be thought of as an abstract idea or a general notion. We differentiate concepts from implementation to highlight the fact that the idea necessarily can take on multiple forms when implemented. For example, the concept of artificial intelligence can be seen in self-driving cars, chatbots, or recommendation engines. The specific implementations are essentially the current state implementation of the concept.
In the following section, I outline my interpretation of what I see as the fundamental definition of business intelligence, reporting, big data, data science, edge analytics, informatics, and the world of artificial intelligence and cognitive computing.
Business Intelligence and Reporting
There is little consensus as to how analytics and business intelligence differ. Some categorize analytics as a subset of business intelligence, while others position analytics in an entirely different box. In a paper I wrote in 2010 (Nelson, 2010), I defined business intelligence (BI) as “a management strategy used to create a more structured and effective approach to decision making…BI includes those common elements of reporting, querying, Online Analytical Processing (OLAP), dashboards, scorecards and even analytics. The umbrella term ‘BI’ also can refer to the processes of acquiring, cleansing, integrating, and storing data.”
There are those who would classify the difference between analytics and business intelligence as differences between (1) the complexity of the quantitative methods used (i.e., algorithms, mathematics, statistics) and (2) whether the focus of the results is historical or future-oriented. That is to say, business intelligence is focused on the presentation of historical results using relatively simple math, while analytics is thought of as much more computationally sophisticated and capable of predicting outcomes of interest, determining causal relationships, identifying optimal solutions, and sometimes also used to prescribe actions to take.
The limit of most business intelligence applications lies not in the constraints of technology but, rather, in the depth of analysis and the true insights created that inform action. Telling me, for example, that something happened doesn't help me understand what to do to change the future—often, that is left for offline analysis. The promise of analytics is that it creates actionable insights about what happened (and where, why, and under what conditions), what is likely to occur in the future, and then what can be done to influence and optimize that future.
Note that the BI dashboard depicted in Figure 1.1 relays facts about the past such as sales, call volumes, products, and accounts, making it easy to get a quick snapshot of the current state of the organization's sales position or activities.
Figure 1.1 BI dashboard
Source: © QlikTech International AB. Reprinted with permission.
Business intelligence and its little sister, “reporting,” are the techniques used to display information about a phenomenon, usually at the tail end of the data pipeline where visual access to data and results are surfaced. Analytics, on the other hand, goes beyond description to actually understand the phenomenon to predict, optimize, and prescribe appropriate actions.
Business intelligence has traditionally suffered from two shortcomings. These shortcomings are related to the fact that BI typically (1) focuses on creating awareness of the facts of the past in that it measures and monitors rather predicting and optimizing; and (2) it is often not quantitatively sophisticated enough to build accurate insights that could be used to influence meaningful change (although the right report or visualization can influence change).
In cases where business intelligence is properly coupled with in-depth “analysis” rather than the mere awareness of facts, it gets closer to analytics. But it often lacks the advanced statistical and mathematical sophistication or “learning” seen in advanced analytics solutions.
To that end, I view analytics as a natural evolution of the concepts contained within the general framework of business intelligence. It places more emphasis on the full pipeline of activities necessary to create insights that fuel action. Analytics is more than just the predefined visual elements used in self-service dashboards or reporting interfaces.
Big Data
Big data is a way to describe the cacophony of information that organizations must deal with in their efforts to turn data into insights. The phrase big data was first used by Michael Cox and David Ellsworth in 1997 (Cox, 1997) who referred to the “problem” as follows:
Visualization provides an interesting challenge for computer systems: data sets are generally quite large, taxing the capacities of main memory, local disk, and even remote disk. We call this the problem of big data. When data sets do not fit in main memory (in core), or when they do not fit even on local disk, the most common solution is to acquire more resources.
Think of big data as a concept that highlights the challenge of utilizing traditional methods of data analysis because of the size and complexity of that data. We contrast big data with traditional “small” data by its volume (how much data we have), velocity (how fast the data is coming at us), and variety (numbers, text, images, video).1
If big data is a concept used to describe the complexity of today's information, analytics is used to help us analyze that complexity in proactive (predictive and prescriptive) ways versus reactive ways (i.e., the realm of business intelligence).
Data Science
It would seem that defining big data was a cakewalk as compared with data science as such little consistency can be found in the term. There is a lot of debate about what it means and whether it is different at all from analytics. Even those who would attempt a definition might do so by discussing the people (data scientists), the skills they need to have, the roles they play, the tools and technologies used, where they work, and their educational backgrounds. But this doesn't give one a meaningful definition.
Rather than describing data science by the people or the types of problems they address, it might be more accurate to define it as follows:
Data Science is the scientific discipline of using quantitative methods from fields like statistics and mathematics along with modern technologies to develop algorithms designed to discover patterns, predict outcomes, and find optimal solutions to complex problems.
The difference between data science and analytics is that data science can help support or even automate the analysis of data, but analytics is a human-centered strategy that takes advantage of a variety of tools, including those found in data science, to understand the true nature of the phenomenon.
Data science is perhaps the broadest of these concepts in that it relates to the entirety of the science and practice of dealing with “data.” I think data science is analytics engineered by computer scientists. In practice, however, data science tends to focus on macro, generalized problems, whereas analytics tends to address particular challenges within an industry or problem space. In Chapter 10, I extend this by defining the relationship between data science and analytics by referring to data science as an enabler of analytics.
Edge (and Ambient) Analytics
Analytics is a predominant activity for most modern organizations that see it as their directive to democratize data through data-driven, human-centered processes. Edge analytics refers to distributed analytics where the analytics are built into the machinery and systems where information is generated or collected as part of the “unconscious” activities of an organization.
Edge analytics is often associated with smart devices where the analytical computation is performed on data at the point of collection (e.g., equipment, sensor, network switch, or other device). Rather than relying on traditional data-pipeline methods where data is collected, transmitted, cleansed, integrated, and warehoused, analytics are embedded within the device or nearby.
DEMOCRATIZE DATA
The democratization of data refers to “freeing” data so that everyone that can and should have access to data is given the tools and the rights to explore the data and these are not limited to the privileged few.
As an example, consider the fact that traditional credit card fraud detection relies on a machine (e.g., card reader) and a connection to an authorization “broker” to validate the transaction by sending a request and very quickly (hundredths of a millisecond) applying an algorithm to authorize or flag the operation and the device receives the authorization. In edge analytics, the algorithm would run on the instrument itself (think smart chip reader with embedded analytics).
Edge analytics is often linked with the Internet of Things (IoT), and a recent IDC FutureScape for IoT report found that “by 2018, 40% of IoT data will be stored, processed, analyzed and acted upon at the edge of the network where it is created” (Marr, 2016). As IoT evolves, we will likely see more attention paid to the Analytics of Things (AoT), which refers to the opportunity of analytics to bring unique value to IoT data.
Ambient analytics is a related term whose name implies that “analytics are everywhere.” Just as the lighting or acoustics of a room often go unnoticed but set the stage for mood, ambient analytics will support and influence the context of how we work and play. We are seeing ambient intelligence play out in everyday scenarios, such as detecting glucose levels and administering insulin. Similarly, home automation devices can detect when you are nearing your home and adjust the temperature and turn on lighting. Ambient analytics goes beyond simple decision rules and utilizes algorithms to decide on the appropriate course of action.
There is little doubt that edge and ambient analytics will continue to challenge the traditional human-centered processes for operationalizing (e.g., understanding, deciding, and acting) analytics.
Informatics
Informatics is a discipline that lies at the intersection of information technology (IT) and information management. In practice, informatics relates to the technologies used to process data for storage and retrieval. In essence, informatics deals with the realm of how information is managed and refers to the ecosystem of data and systems that support process workflows rather than the analysis of the data found therein.
Often used in information sciences and used heavily in healthcare and research, health informatics is a specialization that sits between health IT and health information management and links information technology, communications, and health care to improve the quality and safety of patient care. It lies at the heart of where people, information, and technology intersect.
Health policy refers to decisions, plans, and actions that are undertaken to achieve specific health-care goals within a society. Because health policy makers want to see health care become more affordable, safer, and of higher quality, information technology and health informatics are often the means prescribed to do this. In fact, one of the biggest mandates is to position data resources as to enable a 360-degree view of every patient, and only data sharing can accomplish this (see Figure 1.2).
Figure 1.2 The difference between health information management, health IT, and informatics
Analytics integrates with all of these concepts and relies on the underlying data, supporting technologies, and information management processes.
Artificial Intelligence and Cognitive Computing
Artificial intelligence (AI) is “the science of making computers do things that require intelligence when done by humans” (Copeland, 2000).
The difference between artificial intelligence (AI) and machine learning is that AI refers to the broad concept of using computers to perform the “intelligence” work of discovering patterns, whereas machine learning is a part of AI that relates to the notion that computers can learn from data.
Machine learning is a subset of artificial intelligence that can learn from and make predictions based on data. Rather than following a particular set of rules or instructions, an algorithm is trained to spot patterns in large amounts of data.
Artificial intelligence (and machine learning) can be used in the Analytics Lifecycle to support discovery (e.g., how data is structured, what patterns exist). The application of artificial intelligence found in analytics usually comes in the form of machine learning (as seen above) or cognitive computing.
Cognitive computing is a unique case that combines artificial intelligence and machine-learning algorithms in an approach that attempts to reproduce (or mimic) the behavior of the human brain (Feldman, 2017).
Cognitive computing systems are designed to solve problems the way people solve problems, by thinking, reasoning, and remembering. This approach gives cognitive computing systems an advantage that allows them to “learn and adapt as new data arrives” and to “explore and uncover things you would never know to ask about.” (Saffron Technologies, 2017). The advantage of cognitive computing is that once it learns—unlike humans—it never forgets.
In the battle of man vs. algorithm, unfortunately, man often loses. The promise of Artificial Intelligence is just that. So if we're going to be smart humans, we must learn to be humble in situations where our intuitive judgment simply is not as good as a set of simple rules.
Farnham Street Blog (Parrish, 2017, Do Algorithms Beat Us at Complex Decision Making?)
In slightly pejorative terms, artificial intelligence acts on behalf of a human, whereas cognitive computing provides information to help people decide.
Learn More
To learn more about the difference between AI and cognitive computing, please review the referenced article by Steve Hoffenberg (Hoffenberg, 2016).
THE METHODS OF ANALYTICS
In the prior section, we discussed analytics and some of the related concepts such as big data and data science. We now turn our attention to the practical methods used in analytics, including the tools in the analytics toolbox.
Specifically, in this section, I will outline the methods found in statistics, time series analysis, natural language processing, machine learning, and operations research.
Applied Statistics and Mathematics
Like many of the concepts that have already been discussed, there is a wide disparity about how people define statistics and how it differs from mathematics in general. Some would argue that statistics is a branch of mathematics (Merriam-Webster, 2017b), and others (like John Tukey (Brillinger, 2002)) suggest that it is a science. Most would agree that like physics, statistics uses mathematics but is not math (Milley, 2012).
For present purpose, statistics deals with the collection, organization, analysis, interpretation, and presentation of data. Using that broad definition, it sounds an awful lot like analytics. However, analytics and data science both use the quantitative underpinnings of statistics but their focus is wider than that of traditional statistics. While there are dozens of perspectives about the conceptual relationship between statistics and other disciplines (Taylor, 2016), I have represented what I see as the relationships among those concepts discussed here in Figure 1.3.
Figure 1.3 The relationship between statistics and other quantitative disciplines
Mathematics has a certain absolute and determinable quality about it, and the way that math is taught (at least in US schools) imbues a deterministic way of viewing the quantitative world around them. That is, we are taught to believe that all facts and events can be explained. Statistics, on the other hand, views quantitative data as probabilistic or stochastic. That is, facts may lead to conclusions that may be generally true (beyond simple randomness), but it must be acknowledged that there is some random probability distribution or pattern that cannot be predicted precisely.
Learn More
To learn more about the history of statistics and how it transformed science, please see David Salsburg's book The Lady Tasting Tea (Salsburg, 2002).
As shown in Figure 1.4, mathematical thinking is deductive (i.e., it infers a particular instance by applying a general law or principle) whereas statistical reasoning is inductive (i.e., it infers general laws from specific instances).
Figure 1.4 Inductive reasoning compared to deductive reasoning
This difference is important in the context of analytics in that we apply both inductive and deductive reasoning to analytics problem solving. Thus, the application of both mathematics and statistics to analytics is appropriate and necessary. If analytics is a comprehensive strategy, then statistics and mathematics are tools in our proverbial analytics toolbox that help deliver on that strategy.
Linear programming, for example, can be used to support special types of problems in analytics that are loosely defined as an optimization problem. For example, The Walt Disney Company uses linear, nonlinear, mixed integer, and dynamic programming in its data science work to support the optimization of restaurant seating, reduce wait times for park rides, and schedule staff (i.e., Cast Members). Note that I do not call out operations research, mathematical optimization, decision sciences, or actuarial sciences separately for the purposes of this discussion, as my perspective is that they serve as tools in our proverbial analytics toolkit—just as critical thinking and problem solving.
LINEAR PROGRAMMING
Linear programming is a mathematical method for problem solving where the output is a function of a linear model. For example, we might want to optimize emergency department throughput by looking at several factors, including surgical complexity, number of staff required, and potential complications, for example.
Forecasting and Time Series
In discussing the methods that support analytics, forecasting and time series methods are grouped together, not because they are the same thing but, rather, because they both fall into the same class of problem—the process of characterizing and predicting time-ordered data based on historical information.
Forecasting and time series refer to methods for analyzing time-sequenced data to extract meaningful characteristics from the data. Most often, forecasts are seen as trends represented as a visual display of historical data values, with some providing future predictions. Time series analysis is different than forecasting. You need time series data to make forecasts, but not all time series analysis is done to make a forecast. For example, time series analysis can be used to find patterns or similar features in multiple time series, or to perform statistical process control. Similarly, seasonality can be used to identify patterns.
Time series analysis utilizes a variety of approaches, including both quantitative and qualitative methods. The objective of time series analysis is to discover a pattern in the historical data (or time series) and then extrapolate the trend into the future. In Figure 1.5, note that there are generally four types of time series approaches.
Figure 1.5 Approaches to forecasting and time series analysis
