Transformational Marketing Mix Optimization Using a Virtual Marketplace

TRANSFORMATIONAL MARKETING MIX

OPTIMIZATION USING A VIRTUAL

MARKETPLACE

Jeffrey Maloy, Senior Vice President and Chief Marketing Officer

Tuesday, June 18, 2013

2

WELCOME!

2 weeks, 5 webcasts, improved marketing effectiveness

1. Transformational Marketing Mix Optimization Using a Virtual Marketplace Date: Tuesday, June 18 Time: 1 pm EDT Presenter: Jeffrey Maloy, SVP and CMO

2. Using a Virtual Marketplace to Evaluate Your Marketing Strategy Date: Wednesday, June 19 Time: 1 PM EDT Presenter: Eric Paquette, Senior Vice President

3. Optimizing Your Media Plan for the Bought-Owned-Earned World Date: Tuesday, June 25 Time: 1 pm EDT Presenter: Rolf Olsen, Vice President, Director, Marketing Analytics

4. Leveraging Marketing Investments with Marketing Mix Modeling Date: Wednesday, June 26 Time: 1 pm EDT Presenter: Irina Pessin, Managing Partner, Data2Decisions US

5. Marketing Analytics: 5 Things Every CMO Should Know Date: Thursday, June 27 Time: 1 pm EDT Presenter: Peter Krieg, President and CEO

3

SERIES SCHEDULE

For a PDF of this presentation and our advertorial on using big

data for marketing planning…

Email [email protected]

4

5

WHO AM ?

Jeffrey Maloy Senior Vice President Copernicus Marketing Consulting & Research

Leading our CPG practice Former Director of Marketing, with multiple brand management and consumer insights assignments. Developing relationships with clients such as Kraft, P&G, Nestlé, and Hershey’s. Helping my clients develop, execute and evaluate transformational marketing strategies

THEN: NOW:

6

MARKETING IS BECOMING MORE COMPLEX

7

FROM THIS…

27 4 11 18 25 1 8 15 22 29 5 12 19 26 3 10 17 24 31 7 14 21 28 5 12 19 26 2 9 16 23 30 6 13 20 27 5 12 19 26 2 9 16 23 30 7 14 21 28 4 11 18

$654,000

$654,000

$654,000

$654,000

$11,315,000

$11,315,000

$11,315,000

$11,315,000

$13,657,000

$13,657,000

$13,657,000

$2,962,000

$2,962,000

$2,962,000

$2,962,000

$1,200,000

$550,000

$550,000

$33,760,000

$33,760,000

$765,000

$765,000

$2,970,000

$2,970,000

$9,085,000

$2,325,000

$2,325,000

$1,025,000

$1,025,000

$525,000

$525,000

$225,000

$225,000

$64,920,000$64,920,000

$175,000 $175,000 $175,000

$75,000 $75,000 $75,000

$1,050,000 $775,000 $500,000

$450,000 $375,000 $200,000

$9,085,000

$255,000$350,000 $385,000 $385,000 $385,000 $255,000 $255,000$350,000 $350,000

$33,760,000

$255,000 $255,000 $255,000

$550,000

$5,627,000 $13,624,333 $9,985,667 $4,523,000

135 180 90180 135 90

35/83 43/107 27/5239/113 35/89 27/60

$483,000 $624,000 $305,000$670,000 $528,000 $352,000

285 243 243 243 132 265285 380 190

65/156 62/132 62/131 60/126 45/70 60/14263/155 70/210 54/106

$2,115,000 $1,275,000 $1,265,000 $1,335,000 $492,000 $974,000$2,053,000 $2,802,000 $1,346,000

200 168 166 163 82 163200 266 133

59/104 56/87 56/86 53/81 36/40 50/8257/102 64/135 46/68

$1,810,000 $1,020,000 $1,010,000 $1,080,000 $397,000 $785,000$1,738,000 $2,350,000 $1,125,000

77 50 34

M ay Jun

28/37 21/23 14/16

$312,000 $212,000 $130,000

CostJul Aug Sep Oct Nov Dec Jan Feb M ar Apr

66 HH GRPs/Week GM + DR TV 50/50 :30/:15

95 HH GRPs/week GM + DR TV 100% :30s 81 HH GRPs/week GM 35% :30s/65% :15s + DR TV 100% :30s

$1,200,000

45 HH GRPs/week DRTV Only 100% :30s

~45% reach/month

~55%+ reach/month~50% reach/month + Dog Wet & Treats New Creative Launch (P4CB)

66 HH GRPs/wk GM + DR TV 100% :15

~45% reach/month

… TO THIS

8

Search

Aggregated Content

In-Store

E - Commerce Content Syndication

Pre - Launch

Public Relations

Outreach to Influencers

Sponsorships

Print Live Read Radio

Lifestyle Paid Media

Paid Video (TV, Digital)

Awareness Driving Paid Media

Radio

WOM Live Events

Paid Social

Mobile App Facebook App

Partnerships

YOUR BRAND

HERE

55% of CMO projects

utilize marketing analytics

before a decision is made (Source: The CMO Survey, February 2013)

But…

72% of CEOs say

marketers are always

asking for more money

but can rarely explain how

much incremental

business it will generate (Source: Forbes, 2011)

9

BUT NO “GET OUT OF JAIL FREE” CARD

?

WHO? HOW MUCH?

WHAT? WHERE? WHEN?

“Always On” Planning

Sense and Respond

Highly Targeted

Paid, Owned And Earned Media

Behavioral/ Programatic Buys

Synchronized Swimming

How and Why

Did It Happen

FROM:

Marketing Plan Calendar

Top-Down Allocation

Mass Marketing

Traditional Paid Media

Demographic Buys

Swim Lanes

What Happened

TO:

10

A NEW APPROACH TO MARKETING AND PLANNING IS

NEEDED

“Traditional” Marketing

Evaluation Methods

Regression-based (linear, top-

down) approaches

Historical perspective

Aggregate-level results

Most relevant in cases when:

Less dynamic pace of change

Less competitive environment

Heavier spending on traditional

(paid) media

Less segmented market

More conservative planning

approach (top-down allocation)

11

NEW TOOLS ARE NEEDED TO COMPLEMENT

“TRADITIONAL” TECHNIQUES

Enter: Complexity Science

A study of complex systems.

Simulates how individual parts

(or “agents”) interact with each

other, and the likely outcome

for the system as a whole.

Models dynamic, social

interactions.

What is an Agent:

For our purposes, agents generally represent consumers or groups, and agent relationships represent social interactions.

A discrete entity with its own goals and behaviors.

Mechanisms by which agents interact can be described by behavioral “rules.”

May have the capacity to “learn” (change the rules).

12

A COMPLEX MARKET REQUIRES COMPLEXITY

SCIENCE

Simulation models, such as agent-based modeling is not new.

Until recently, it primarily has been used for simulating:

– Migration of species

– Spread of viruses and diseases

– Supply chain

Now being used in fields in which simulation modeling has never been applied before (markets)

13

TRULY “VIRAL” MARKETING

In the 2000s, some smart man asked this question:

“In this social age, why can’t

ABMs describe the spread of

marketing messages?”

Fortunately there

was no good answer.

14

WHY NOT MARKETING

A simulation approach to marketing mix optimization is especially

relevant when:

Dynamic marketplace

Forwarding-looking perspective

Experimental marketing plans – paid, owned, and earned media

Highly segmented market

Want to understand impact on sales

and brand equity

Test and Learn mentality

Marketing never sleeps

Need to know

the How and Why

behind the What

15

WHY SIMULATION MODELING?

Bought Owned Earned

Marketing Plans

Target Segments…….….

Message Strategy……....

Media Strategy (Paid,

Owned and Earned)….…

Pricing Strategy……….…

Product Strategy…….…..

16

CONNECTS NUANCES OF MARKETING PLANNING AND

EVALUATION

Complex marketplace

Diverse population

Fragmented media

“Always on” devices

“Big Data”

Need for computing power

17

WHY NOW?

18

TURNING 3V’S INTO 3W’S

How to turn “big data” into “smart data”?

Answer: 3W’s:

When

Where

Why

Volume

Velocity

Variety

3 V’s

Economic

Data

Environ-

mental

Data

Media

Costs

POS

Data

Price/

Promotion

Data

Competitive

Data

GPS/RFID

Data

Survey

Data

1. Build your “simulated society”

2. Assign “rules of engagement”

3. Calibrate and validate

4. Explore new frontiers

19

SO HOW DO WE BUILD THIS AMAZING MODEL?

First we build a “virtual” panel of agents:

Each agent is like a mini-database, waiting to be populated with data and

behavioral rules.

Panel must be representative of relevant population on:

Gender

Age

Income

Ethnicity

Corpo-graphics

Geographic region

Category incidence

21

1. BUILD SIMULATED SOCIETY

Establish decision rules for each agent and/or segment of agents:

How do they react when acted upon by:

• Other agents

• Marketing and Media

• External factors

Rules are based upon:

• Marketing sciences

• Secondary data (media consumption)

• Custom research (e.g., segment purchase behaviors)

• Top-down analysis (Marketing Mix Models)

• Advanced analytics

• Understanding of their “journey” and key decision points

22

2. PROGRAM THE AGENTS

23

2. PROGRAM BEHAVIORS ALONG SHOPPER JOURNEY

Once rules are established, we run and refine the model until

results reflect “reality”:

24

3. CALIBRATE AND VALIDATE

Examine model’s results, relative to

expectations

Calibrate agent rules

− Are results logical, intuitive, reasonable?

− Possibly validate the model on a “hold out” sample

Proceed with caution

− As with any good model, results needed to be

combined with judgment

Now, we’re ready to dream. Test those risky plans in a risk-free

simulation.

25

4. EXPLORE NEW FRONTIERS

What if…

Spend more/less?

Shift to more digital tactics?

Target new/different segments?

Reposition brand?

Competitive aggression?

Powered by CCM (agent-based simulation and

other advanced analytics), we analyzed multiple

strategic alternatives:

– Targeting strategies

– Messaging strategies

– Product focus

– Media mix and allocation

26

CASE STUDY

After multiple scenarios, we identified one that delivered…

Challenge

Analysis

Increase sales by +4-5% with a flat marketing budget.

+ >$200 MM

+ >4% incremental revenue lift

27

CASE STUDY

Action

Results

From: To:

GM W 25-44 USH W 18-49

TV Digital, Mobile and Social

Holiday

Message Promotional Message

Assortment

Focus Bed & Bath Focus

1. Simulation models (such as ABMS) can be

an effective tool in the analytic toolbox

2. Complex, dynamic models for complex,

dynamic markets

3. Simulate new, “stretch” scenarios

4. Understand the “how” and “why” behind

the “what”

28

CONCLUSIONS

Questions?

For a PDF of this presentation and our advertorial on using big data for marketing planning…

Email [email protected]

29

USING A VIRTUAL MARKETPLACE TO EVALUATE

YOUR MARKETING STRATEGY

Eric Paquette, Senior Vice President

Wednesday, June 19, 2013

SAVE THE DATE (TOMORROW):

Jeffrey Maloy, Senior Vice President

(203) 831-2375

Jeff.Maloy @copernicusmarketing.com