Data is the new oil 2 Hitachi Ltd

Data is the new oil. 2 © Hitachi, Ltd. 2018. All rights reserved.

“Data is the New Oil…” Understanding the “Economics of Oil” requires understanding the difference in value between “Raw” Oil and “Refined” Oil Raw WTI Crude Oil VP MRX 02 Racing Fuel $62 / Barrel $125 / 5 -Gallon = $1, 050 / Barrel Refined high-octane racing fuel 17 x more valuable than raw WTI crude oil * But how much more valuable is that barrel of high-octane fuel if that barrel never depletes, never wears out and can be used across unlimited use cases? Source: Bill Schmarzo “Big Data MBA” Course Curriculum 3 * as of 04/04/2019

Driving Business Strategies with Data Science Bill Schmarzo Hitachi Vantara CTO, Io. T and Analytics University of San Francisco, Executive Fellow Honorary Professor, National University of Ireland-Galway Twitter: @Schmarzo 4 © Hitachi Vantara Corporation 2018. All Rights Reserved © Hitachi Vantara Corporation 2019. All Rights Reserved.

Big Data Business Model Maturity Index How Effective is Your Organization at Leveraging Data and Analytics to Power your Business Models? DIGITAL TRANSFORMATION Key Business Processes Big Data Economics INSIGHTS MONETIZATION BUSINESS OPTIMIZATION BUSINESS MONITORING BUSINESS INSIGHTS Prescriptive Recommendations Bill Schmarzo “Big Data MBA” Curriculum

Data Science Value Engineering Framework Business Initiative Stakeholders Decisions Analytics Data Architecture & Technology Source: Bill Schmarzo “Big Data MBA” Course Curriculum “Economics of Data™“ Playing Cards

Monetizing Big Data and IOT 7 CONFIDENTIAL – For use by Hitachi and Disney employees under NDA only. © Hitachi, Ltd. 2018. All Rights Reserved © Hitachi, Ltd. 2018. All rights reserved.

Every Two Days We Create As Much Information As We Did From The Dawn Of Civilization Until 2003 - Eric Schmidt, Google CEO A Billion People Visit Facebook Every Day - Business Insider By 2018, Leading Enterprises Will Support 1, 000– 10, 000 TIMES More Customer Touch Points - IDC 8 8

“Internet of Things” (IOT) Causing Data Explosion 30 Billion 44 Zettabytes CONNECTED DEVICES OF DATA 7 Billion CONNECTED PEOPLE Source: Gartner Group, 2014 Source: Bill Schmarzo “Big Data MBA” Course Curriculum 9 Millions OF NEW BUSINESSES

Personal Experiences Changing Corporate Expectations Recommends movies, restaurants, friends, spouses, books, routes, etc. Source: Bill Schmarzo “Big Data MBA” Course Curriculum 10

Big Data Isn’t About Big…it’s About Small INCLINATIONS AFFINITIES PROPENSITIES PASSIONS AFFILIATIONS INTERESTS TASTES BIASES PREFERENCES ASSOCIATIONS BEHAVIORS TENDENCIES Source: Bill Schmarzo “Big Data MBA” Course 11 Source: Bill. Curriculum Schmarzo “Big Data MBA” Course Curriculum 11

Big Data isn’t about Big…it’s about Small AFFINITIES PROPENSITIES AFFILIATIONS INCLINATIONS PREFERENCES BIASES TRENDS PATTERNS TENDENCIES 12 Source: Bill Schmarzo “Big Data MBA” Course Curriculum BEHAVIORS © Hitachi, Ltd. 2018. All rights reserved.

Basic Data Science Concepts 13 CONFIDENTIAL – For use by Hitachi and Disney employees under NDA only. © Hitachi, Ltd. 2018. All Rights Reserved © Hitachi, Ltd. 2018. All rights reserved.

What is Data Science? Data Science: Identifying variables and metrics that might be better predictors of performance 2000 2001 14 Source: Bill Schmarzo “Big Data MBA” Course Curriculum 2002 2003 2004 2005 2013 14

Analytics Monetization Curve Descriptive Questions Predictive Analytics Prescriptive Actions What were revenues and profits last year? What will revenues & profits likely be next year…? Plant X and Y crops across N acres How much fertilizer did I use last planting season? How much fertilizer will I likely need next planting season…? Pre-order X amount of fertilizer at 5% discount (What happened? ) How much downtime did I have last month due to unplanned equipment maintenance? How many workers did I use last month? (What is likely to happen? ) When will my equipment likely need maintenance next month…? How many workers will I likely need next month and when will I need them…? Source: Bill Schmarzo “Big Data MBA” Course Curriculum (What should we do? ) Service your harvester and tractor #2 in January Hire X number of workers for Y days “Economics of Data™“ Playing Cards

The Art of “Thinking Like A Data Scientist” 1 Identify Business Initiative 5 Identify Data Sources 2 Identify Stakeholders 6 Group Metrics Into Scores Source: Bill Schmarzo “Big Data MBA” Course Curriculum 3 Identify Analytic Entities 7 4 Identify / Prioritize Identify Recommendations Use Cases 8 Map Scores to Recommendations “Economics of Data™“ Playing Cards

Data Science Collaborative Engagement Process Supports rapid exploration, rapid testing, continuous learning “Scientific Method” Step 1: Define Hypothesis (Decision) to test or Prediction to make Step 2: Gather data…and more data (Data Lake: SQL + No. SQL) Historical Weather Forecast CDC Google Trends Physician Notes Local Events Kronos Epic Lawson REPEAT Step 3: Prepare data; Build schema (schema-on-query) Step 4: Visualize the data (Tableau, Pentaho, ggplot 2, …) Step 5: Build analytic models (Tensor. Flow, Python, Jupyter…) Step 6: Evaluate model “goodness of fit” (coefficients, confidence levels) Source: “Scientific Method: Embrace the Art of Failure”, University of San Francisco School of Management Big Data MBA Source: Bill Schmarzo “Big Data MBA” Course Curriculum “Economics of Data™“ Playing Cards

Saddle Point Optimization Challenges Optimization paths (A and B) may top out sooner with lower “Saddle Points” than optimal path (C). Data scientist must be prepared to jettison current path to find more predictive path. C B A Source: Bill Schmarzo “Big Data MBA” Course Curriculum “Economics of Data™“ Playing Cards

Hypothesis: Increase Same Store Sales by X% Completed by: Date: Schmarzo 10/27 (3) Business Value (1) Hypothesis (11) Impediments • Increased top line revenue • Better (faster) customer experience • Fresher inventory • Increase overall profits • Increased asset utilization Increase Same Store Sales by 7. 1% over the next 12 months (2) KPI’s Average Sales per Visit, Store Traffic, Sales per • Lack of quality data • Lack of analytic skills to create predictions • Store/Field Management buy-in • Modern technology architecture • Financing/budget (6) Decisions (5) Entities (12) Risks • Staffing • Local Events Sponsorship • Promotions & Types • Corporate Catering • Loyalty Program Employee, Line Wait Time, % Abandonment, Mobile Orders, Positive Social Media Mentions, Table Turns • Non-corporate Catering • Inventory Management • Suppliers • Customer Satisfaction • New Product Intros (7) Predictions • Demand (Traffic) Profitability • Poor execution affects customer satisfaction • Increased demand stresses employee satisfaction • Weather disrupts local events • Increased demand impacts product quality • Suppliers can’t keep up with increased demand (10) Recommendations • Recommend Staffing • Recommend Local • Recommend Inventory • Recommend Catering Targets Events Promotions • Recommend Store Hours (13) Financial Assessment Avg Sales / Store • Store Operations • Corporate Marketing • Field Marketing • Procurement • Finance (8) Data • Staff Availability • Abandonment • Mobile Orders • Weather Forecast • Promotional Response • Basket Size • New Product Demand (4) Stakeholders • Stores • Customers • Suppliers • Store Managers • Competitors 01 Iteration: (9) Variables (Dimensions) Customer Sat 4 2 3 Product Quality Brand Building Employees 2 2 3 Source: Bill Schmarzo “Big Data MBA” Course Curriculum • • Store location Store size Store open date Local demographics Local house values Local economics Products sold Time of day • Day of week • • (weekends) Holidays Seasonality Weather conditions Traffic patterns Miles from high school Miles from mall Local sporting events (14) Impediments Assessment Data Analytic Skills Store Mgmt 3 4 1 Technology Financing TBD 3 2 0 Hypothesis Development Canvas (v 2. 0)

2 1 Design Thinking: Humanizing Machine and Deep Learning 21 © Hitachi Vantara Corporation 2019. All Rights Reserved

Design and Data Science “ Data Science uncovers and codifies the trends, patterns and relationships buried in data Design Thinking uncovers and codifies the trends, patterns and relationships buried in people ” 22 Source: Bill Schmarzo “Big Data MBA” Course Curriculum

Why Design Thinking and Data Science? Converting Hunches, Heuristics and Rules of Thumbs into Math Mystery Heuristics Algorithm 23 Source: Bill Schmarzo “Big Data MBA” Course Curriculum “The secret to moving organizations from mystery and heuristics to algorithms lies in the ability of Design Thinking to uncover and codify the deep knowledge and insights of the organization’s subject matter experts” - Bill Schmarzo

Design thinking involves observation to discover unmet needs within the context and constraints of a particular situation. It frames the opportunity and scope of innovation, generating creative ideas, testing and refining solutions. It creates a repeatable and scalable process for innovation. Design Thinking Learn about audience for whom you are designing. Who is my user? What matters to this person? Create POV based on user needs and insights. What are their needs? Ideate Empathize designs products this way… Brainstorm as many creative solutions as possible. Wild ideas encouraged! Define Model one or more of your ideas to show to others. How can I show my idea? Remember: A prototype is just a rough draft. Prototype Test refines, tunes and predicts this way… Machine Learning Machine learning is a method of data analysis that automates analytical model building. Using algorithms that iteratively learn from data, machine learning allows computers to find hidden insights without being explicitly programmed where to look. Source: John Morley & Bill Schmarzo Share prototype for feedback. What worked? What didn’t? Analyze Break down needs into each of base parts (decisions) Ideate Synthesize Combine separate elements in order to create a new ‘whole’ Brainstorm to process the product requirement into: -Feature maps -Classes -Metrics -Adaptive needs Humancentered Tuning Tune the model to boost accuracy. Avoid ‘overfitting’ Validate Meet inference performance metrics. Avoid ’overfitting’ Machinecentered

Exploiting the Economic Value of Data University of San Francisco Economic Value of Data Research Project Bill Schmarzo Hitachi Vantara CTO, Io. T and Analytics University of San Francisco, Executive Fellow Honorary Professor, National University of Ireland-Galway Twitter: @Schmarzo 25 CONFIDENTIAL – For use by Hitachi and Disney employees under NDA only. © Hitachi, Ltd. 2018. All Rights Reserved © Hitachi, Ltd. 2018. All rights reserved.

Economic Value of Data Research • Data is an asset that never depletes, never wears out, and can be used across unlimited use cases at near zero marginal cost • Economic Multiplier Effect: an increase in spending produces an increase in value greater than the initial amount Customer point of sales data Sales Marketing Call Center Product Dev Promotional effectiveness Customer acquisition Customer retention New product intro +2. 5% +2. 0% +3. 5% +2. 6% Source: Bill Schmarzo “Big Data MBA” Course Curriculum “Economics of Data™“ Playing Cards

Intellectual Capital “Rubik’s Cube” Challenges • How does the organization determine the economic value of its data in order to drive prioritization and investment decisions? • How does the organization avoid data silos and shadow IT spend that thwarts potential value of data? • How does the organization avoid the disillusionment of “orphaned analytics”? How does one leverage an asset that appreciates (not depreciates) with usage, and can be used simultaneously across multiple business processes? Source: Bill Schmarzo “Big Data MBA” Course Curriculum 28

Intellectual Capital “Rubik’s Cube” Solution USE CASES Clusters of decisions around common subject area in support of organization’s key business initiatives DATA Detailed historical transactions coupled with internal unstructured and publiclyavailable data sources Source: Bill Schmarzo “Big Data MBA” Course Curriculum 29 ANALYTICS Data transformed into actionable analytic insights (scores, rules, propensities, segments, recommendations)

3 1 Digital Transformation and the “Economies of Learning” 31 © Hitachi Vantara Corporation 2019. All Rights Reserved

“Economies of Learning” versus “Economies of Scale” • “Economies of Scale” have given large enterprises unsurmountable market advantages through exploitation of mass production, distribution and marketing. • In knowledge-based industries, “Economies of Learning” more powerful than the “Economies of Scale” SMART CITY SMART FACTORY SMART AIRPORT SMART HOSPITAL • Today, every industry is becoming a knowledge-based industry Source: Bill Schmarzo “Big Data MBA” Course Curriculum “Economics of Data™“ Playing Cards

Schmarzo Economic Digital Asset Valuation Theorem The more the data and analytics get used, the more accurate, more complete, more robust, more predictive and consequently more valuable they become Hi Effect #3: Economic Value Accelerates Value ($$$) • Refining Analytic Module predictive effectiveness ripples thru previous use cases that use that Analytic Module Effect #2: Economic Value Grows • Cumulative financial and operational value grows use case by use case Effect #1: Marginal Costs Flatten • Reusing “curated” data and analytic modules reduces marginal costs for new use case (no data silos or orphaned analytics) Lo Number of Use Cases Source: Bill Schmarzo “Big Data MBA” Course Curriculum “Economics of Data™“ Playing Cards

3 Horizons of Digital Transformation How do organizations exploit new digital technologies, real-time data, advanced analytics and communication channels to digitally transform their business and operational models? Aspirational Agriculture Company Horizon 3: Autonomous Farming Exploit analytics-infused automation technologies and deep customer, product and operational insights to create Autonomous Farming-as-a-Service business model Horizon 2: Digital Farms Create Digital Farms that self-monitor, self-diagnose and self-heal by exploiting equipment, farming usage patterns, soil, crop, pesticide/herbicide, weather, commodities pricing and operational insights Horizon 1: Farming Excellence Today Optimize farming operational management via yield optimization, waste and energy reduction, predictive maintenance, resource scheduling, machinery utilization, inventory optimization, asset lifecycle management etc. Source: Bill Schmarzo “Big Data MBA” Course Curriculum

Data is not the new oil. Data is the new sun. 35 Source: Bill Schmarzo “Big Data MBA” Course Curriculum University of San Francisco Economic Value of Data Research Project Data is unlike any other corporate asset It never depletes It never wears out Same data set can be used across infinite use cases © Hitachi Vantara Corporation 2019. All Rights Reserved.

Bill Schmarzo BILL SCHMARZO Hitachi Vantara CTO, Io. T and Analytics University San Francisco School of Management, Executive Fellow Honorary Professor, National University of Ireland-Galway Top-ranking Blogs v To Achieve Big Data’s Potential, Get It into the Boardroom v Big Data Business Model Maturity Index v 6 Laws of Digital Transformation v History Lesson on Economic-Driven Business Transformation v User Experience: The New King of the Business v IOT: Transitioning from Connected to “Smart” v Learning How to “Think Like a Data Scientist” Contact Information Bill. Schmarzo@Hitachi. Vantara. com Find me on Twitter: @schmarzo Source: Bill Schmarzo “Big Data MBA” Course Curriculum 36