PwC presentation at the Chief Analytics Officer, Fall 2016

Don’t Be Data Rich & Decision Poor Insights from PwC’s Big DecisionsTM Research CAO Forum Fall - NYC October 2016

PwC

Decision making models…..

2

“Guitar groups are on the way out.”

Dick Rowe, Decca Records executive, 1962

I’m bringing you into the decision making process

Ruggles, here – flip this coin!

Problem Solving / Decision Making

PwC 3

Think of a bad decision your company has made?

Think of a good decision your company has made?

What was the difference?

PwC 4

Mind vs. Machine

10,000 Brains Anywhere, Anytime

To Trust or Not to Trust

Data Ecosystems

4 V’s of Data

Show Me a Picture, Please

What will you do differently?

PwC‘s Global Data and Analytics Survey 2016: Big Decisions™

PwC

PwC’s Global Data and Analytics Survey 2016: Big DecisionsTM

5

Why

• Strategic decisions create value for an organisation.

• Decision-makers are now face-to-face with an opportunity to learn from massive amounts of data.

• How can we apply data analytics to create greater value?

What

• What types of decisions will you need to make between now and 2020?

• What types of data and analytics do these decisions require?

• What is the role of machines in decision making?

• What’s your ambition for improving your company’s decision speed and sophistication to make these decisions?

Who

• 2,100+ senior decision-makers

• 50+ countries

• 15 industries


PwC

0% 5% 10% 15% 20% 25% 30% 35%

Developing or launching new products or services

Entering new markets with existing products or services

Developing Partnerships

Investment in IT

Change to business operations

Corporate restructuring or outsourcing

Entering a new industry or starting a new business

Shrinking existing business

Other Decision

Which one of the following best describes this key strategic decision?

Global

The leading “big decision” across Global Markets is “developing or launching new products” followed by “entering markets” and “investment in IT ”, all projected to increase shareholder value

Most Important Strategic Decisions & Impact

Across all strategic decision types, on average 90% of

leadership thinks their strategic

decision will increase shareholder value, with the majority estimating 5-50%

increase and 1/3rd estimating 50-200% increase


*n = total # of the top key coming strategic decisions

6

PwC

Organizations are in different stages in their approach to using data and analytics to support strategic decision making

Evolving capabilities, finding their way…..

Reconciling how to integrate “gut based” approach and avoid bias…..

Hampered by structure…..

Somewhat detailed incorporating

lengthy periods for reflection and refinement.

Hierarchical validation within a

fragmented decision making structure

We are growing our use of complex data sets and relying more and

more on external market data to make

decisions.

Generally analytics are rarely relied upon…from a business perspective data does not drive our decisions.

We make decisions and then find supporting data to justify them.

Fragmented & ad hoc

Especially

manual data

processing (low speed and

small amount of

data that we can

handle at on-time)

…improving...data is

becoming more and more key in decision making

It is patchy. There is still a noticeable

reliance on gut based on

what has been experienced in the past.

Continuing to evolve; have recently implemented big

data effort/strategy to enhance use

The use of comprehensive analytics to inform pro-

active decision making

7


PwC

3 9 %

5 3%

8 %

Highly data-driven Somewhat data-driven

Rarely data-driven

Most companies are not “highly data driven” and rely on descriptive and diagnostic analytics the most

Global

Which of the following best describes decision-making in your organization?

Majority Aren’t Highly Data Driven.. …Or Using Predictive or Prescriptive

Similar pattern across industries

8


*n = # of type of data-driven organization *n = # of type of data-driven organization by type of analytical technique applied

0%

5%

10%

15%

20%

25%

30%

35%

Descriptive (What hashappened?)

Diagnostic (Why did ithappen?)

Predictive (Whatwill/could happen?)

Prescriptive (Whatshould happen and how?)

The use of analytics in your organization is mostly…

Highly data-driven Somewhat data-driven Rarely data driven

Global

PwC

The new order will change the balance of algorithms and human judgment used in decision making and make “unknown” risks “known”

9

Reliance on Judgment vs. Machine Analysis by Risk Profile (n= # of Decisions)

• Complement human judgment with machine algorithms (i.e. AI)

• Continuously improve algorithms

Strike the right balance of mind & machine….

• Know something your competitors don’t

• Be the first to react to emerging, latent demand

• Migrate from “beta” to “alpha”

Address risks by making them known….


Opportunities

Min

d &

& M

ac

hin

e

– T

he

rig

ht

ba

lan

ce

Known Manageable…...….RISK…….….Unknown, Uncertain Ma

chin

e A

lgo

rith

ms.

...A

NA

LY

SIS

…..

Hu

ma

n J

ud

gem

ent

Make Unknown Risks Known

PwC

The satellite selected has a spectral resolution of a 31cm per pixel, the highest commercially available, for the analysis

Collect data in novel ways…

Perform market sizing analysis in emerging markets…

Use Satellite Imagery to Size Markets


10

Finding the right mix of “mind and machine”…..

PwC

Finding the right mix of “mind and machine”…..

Use Drone Imagery to Assess Capital Projects

Identify likely safety and code violations

Reduce schedule overruns…


Key: Concrete Background Steel Reinforcement Wooden Boards 11

PwC

Use of human judgment and machine algorithms varies by industry across decisions that involve know and unknown risks

X-A

xis

(E

ac

h G

ra

ph

)

M

ac

hin

e A

lgo

rit

hm

s v

s.

Hu

ma

n J

ud

gm

en

t

Y-Axis (Each Graph) Known vs. Unknown Risks

Health Services Pharma & Life Sciences Technology Communications Entertainment & Media

Retail & Consumer Energy, Mining, Utilities Industrial Products Insurance Banking & Cap Markets

Reliance on Judgment vs. Machine Analysis by Decision Risk (n= # of Decisions)

12


PwC

Companies are at different levels of maturity in decision making “speed” and “sophistication” to create value……

Speed

• Time to answer question

• Time to decide action

• Time to implement / measure

Sophistication

•Analytics maturity

•Data breadth & depth

•Decision approach


Sophistication

A ccelerated A gility

Ma st er the Ch ess Mov es

In t elligence in t he

Mom en t

Cov er the Ba sics

Low High

Lo

w

Hig

h

Sp

ee

d

Increasing sophistication should simplify, not increase complexity

Speed is as much about structure as it is about

data & analytics

PwC’s Decision Sophistication & Speed Matrix (n=# of decisions)

13

PwC

Increase “speed” and “sophistication”…..


Simulate Adoption of Autonomous Personal Mobility Solutions More Speed…..

More Sophistication…..

• Quickly analyze and adapt go-to-market approaches based on in market feedback

• Simulate a million ‘consumer’ agents and their purchase choices based on causal reasoning

• Run over 200K + go-to- market scenarios to prescribe the right city, pricing, and # of vehicles

Modeling demand for vehicle miles travelled

Simulating demand, charging and utilization by geography Driverless & Electric Vehicles

14

PwC

Increase “speed” and “sophistication”…..


Simulate Adoption of Autonomous Personal Mobility Solutions More Speed…..


• Faster tracking of frequency of movement and use of space

• Complete dangerous inspections

• 2-D images are converted to 3-D digital models

• Automate inspection and visual analysis with deep learning models

Simulating demand, charging and utilization by geography

Identify & analyze physical objects to deliver new insights

15

PwC

Increase “speed” and “sophistication”


Machine Learning/NLP: Modeling Willingness to Pay

More Speed…..


• Reduce time of market research

• Implement targeted outbound campaign messaging

• Leverage Word2Vec NLP techniques to go beyond “positive / negative sentiment”

• Design more targeted price points

16

PwC

Ambition is high to improve decision speed and sophistication Orange shows today; blue shows where companies want to be by 2020

17

Global


Lo

w

Hig

h

Sp

ee

d

Low High

Sophistication

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0

United States

PwC 18

Each decision type requires a focused approach for improvement Focus may require improving speed, sophistication or both.

Developing/Launching New Products


Entering New Markets

Improve Operations Investment in IT

PwC

Organizations view leadership courage, budgetary constraints, and resource availability as barriers data driven decision making…

19

Barriers to Decisions

The C-Suite is marginally more

confident in leadership courage,

with it’s top two concerns being #1

Budgetary considerations

and #2 Available resource/manpo

wer


0% 2% 4% 6% 8% 10% 12% 14% 16% 18% 20%

Leadership courage

Budgetary considerations

Availability of resource/manpower

Operational capacity

Policy regulations

Issues with implementation

Poor market response

Ability to analyse data

Data limitations

The Decision will likely be limited by…

Global - Total*n = top decision by top limitation

PwC 20


Improved decision making with Data & Analytics requires overcoming common decision traps

Anchoring Trap Overconfidence Trap

Status-Quo/Sunk Cost Trap

Confirming-Evidence Trap

(Confirmation Bias)

Framing Trap Availability Bias

(Rush to Solve)

Disproportionate weight to first information received

Overestimate judgment and predictions; remember success, forget errors

Perpetuates the current state or past decisions; risk-averse mindset

Seek supporting information; avoid contradictory information

How a problem is framed influences the decisions made

Rely on information that is most readily available

Show options & present range of facts

Use gaming Simulate and quantify risk of status quo

Leverage benchmarks

Use different framings (competitor, customer, employee)

Create Comprehensive Decision Support Systems

De

cis

ion

T

ra

p

De

scr

ipti

on

M

ax

imiz

e D

&A

Im

pa

ct

PwC 21


Effectively making decisions with D&A requires tailoring the approach and benefits to the decision makers style

Controller Skeptic Follower Charismatic Thinker

The Skeptic (Larry Ellison, Steve Case)

• Decisions made on gut feeling

• Challenges every data point Applying D&A • Co-present with

trusted advisor • Emphasize credibility of

D&A data sources • Arguments grounded

in reality • Presentation capitulates to

skeptic leaders’ ego

The Charismatic (Richard Branson, Marc Benioff)

• Easily enthralled, but uses balanced approach

• Emphasize bottom line results

The Controller (Martha Stewart, Ross Perot)

• Unemotional and analytical

• Only implements own ideas

The Follower (Peter Coors, Carly Fiorina)

• Relies on others’ past decisions to make current choices

• Late adopter

The Thinker (Bill Gates, Michael Dell)

• Toughest to persuade • Risk-averse • Attention to detail

PwC 22

Key findings from Big Decisions survey


More organizations are taking a data-driven approach to making strategic decisions.

Are you?

Data-driven organizations are using machines to de-risk their decisions.

Executives have great ambition to increase decision speed and sophistication, but everyone expects to fall short of their ambition.

What’s your expectation?

Organizations face many limitations in their decision making, however data and the ability to analyze data are the least of their concerns.

PwC

This publication has been prepared for general guidance on matters of interest only, and does not constitute professional adv ice. You should not act upon the information contained in

this publication w ithout obtaining specif ic professional advice. No representation or w arranty (express or implied) is given as to the accuracy or completeness of the information

contained in this publication, and, to the extent permitted by law , Pricew aterhouseCoopers LLP, its members, employees and agents do not accept or assume any liability,

responsibility or duty of care for any consequences of you or anyone else acting, or refraining to act, in reliance on the information contained in this publication or for any decision

based on it.

© 2016 Pricew aterhouseCoopers LLP. All rights reserved. In this document, “Pw C” refers to Pricew aterhouseCoopers LLP w hich is a member f irm of Pricew aterhouseCoopers

International Limited, each member f irm of w hich is a separate legal entity.

Thank you For more information visit, www.pwc.com/bigdecisions Continue the conversation with us online, follow: PwC Advisory Services, @PwCAdvisory Paul Blase, Global and US Data and Analytics Consulting Leader, @paulblase

23

http://www.pwc.com/bigdecisions

http://www.twitter.com/pwcadvisory

https://twitter.com/paulblase