Introduction to
Marketing Analytics
Introduction to Marketing Analytics
Outline
▪ The new realities of marketing decision
making
▪ Marketing Engineering (ME)
▪ The role of models in ME
▪ Allegro exercise
Introduction to Marketing Analytics
Challenges faced by today’s
marketing decision makers
▪ Global, hypercompetitive business environment.
More demanding customers served by a greater number
of competitors on a global scale.
▪ Exploding volume of data
“We’re drowning in data. What we lack are true insights.”
▪ Need for faster decision making
Information overload and lack of time, yet decisions have
to be made all the time.
▪ Higher standards of accountability
Marketing expenditures have to be justified in the same
way as other investments.
Introduction to Marketing Analytics
Need for better marketing
decision making
▪ Intuitive decision making
□ Instinct, gut feelings, past experience, established
practices;
□ In a world characterized by rapid change, information
overload, greater accountability, etc. intuition is
unlikely to generate superior results;
▪ Data-based and model-supported decision making
□ Marketing Engineering
Introduction to Marketing Analytics
Marketing Engineering (ME)
“a systematic approach to harness data and
knowledge to drive effective marketing decision
making and implementation through a technology-
enabled and model-supported interactive decision
process.” (LRB, p. 2).
Introduction to Marketing Analytics
Skills needed for Marketing Engineering
Introduction to Marketing Analytics
Marketing Engineering
Marketing Environment
Automatic scanning, data entry,
subjective interpretation
Data Database management, e.g..,
selection, sorting, summarization,
report generation
Information
Engineering
Marketing
Mental models,
Decision models
Insights Judgment under uncertainty,
e.g.., modeling, communication,
introspection
Decisions
Financial, human, and other
organizational resources
Implementation
Introduction to Marketing Analytics
Models are the core of ME:
What is a Model?
▪ Models are stylized representations of reality that
structure our thinking about how the world works;
▪ Models indicate which factors should be considered
and which factors can be ignored;
▪ By focusing on the relevant factors and their
interrelationship reality can be simplified;
▪ Models are useful because they facilitate top-down
processing (as opposed to bottom-up processing);
Introduction to Marketing Analytics
Issues in using models
▪ assembling an arsenal of models for a domain of
interest;
▪ retrieving relevant mental models in a given situation;
▪ being aware of the limitations of mental models (they
may overrepresent and underrepresent, or even
malrepresent, things);
“No model is true, but some models are useful.”
Introduction to Marketing Analytics
Types of models
Verbal Box and Arrow Graphical Mathematical
sales
“Sales are a
function of Advertising S=a+ bA+e
advertising”
Sales
advertising
Introduction to Marketing Analytics
Response models in the decision loop
Marketing actions Competitive actions Observations
(inputs) (outputs)
Product design, Awareness
Response
Price, Advertising, Preferences
Model
Selling effort, etc. Sales
Environmental
Conditions
Objectives
Control, Adaptation Evaluation
Introduction to Marketing Analytics
A simple (linear) response model
Sales as a function of advertising spending
13200000
13000000
12800000
12600000
12400000
Sales
12200000
12000000
11800000
11600000
11400000
700000 800000 900000 1000000 1100000 1200000 1300000 1400000 1500000
Advertising spending
Introduction to Marketing Analytics
A simple (linear) response model
Actual and predicted sales as a function of advertising spending
13200000
13000000
12800000
12600000
12400000
Sales
12200000
12000000
11800000
11600000
11400000
700000 800000 900000 1000000 1100000 1200000 1300000 1400000 1500000
Advertising spending
Predicted sales Actual sales
Introduction to Marketing Analytics
A simple (linear) response model
SUMMARY OUTPUT
Regression Statistics
Multiple R 0.95
R Square 0.90
Adjusted R Square 0.88
Standard Error 151514.77
Observations 10
ANOVA
df SS MS F Significance F
Regression 1 1568596189164 1568596189164 68 0
Residual 8 183653810836 22956726354
Total 9 1752250000000
Coefficients Standard Error t Stat P-value Lower 95% Upper 95%
Intercept 9906547.3 292891.1 33.8 0.0 9231139.1 10581955.5
Advertising 2.1 0.3 8.3 0.0 1.5 2.7
Introduction to Marketing Analytics
How to build a model
▪ Specify the model
□ Variables (which ones to include)
□ Relationships, interactions, dynamics (how they are linked)
▪ Calibrate the model
□ Statistical estimation with real (historical) data
□ Experimentation
□ Managerial judgments
▪ Validate the model
□ Face validity (does it make sense?)
□ Global fit (R², model fit)
□ Variable significance (correct signs, t-tests)
▪ Apply the model
□ Ease of use
□ Usefulness
Introduction to Marketing Analytics
Allegro
Resource allocation "smart sheet" exercise
Introduction
Following are two business plan spreadsheets for the same company, Allegro. You have just been assigned to
be the brand manager and must formulate a plan for next year. As you will see, the projected profits for 2006 are
higher than the plan for 2007. To make a good impression, you need to develop a plan that will deliver higher
profits in 2007. Your job is to use each spreadsheet to determine the optimal level of price, advertising, and
selling.
How to use this spreadsheet
Start with the Simple spreadsheet. It is a traditional spreadsheet. Use this as an aid to help you think about
optimal spending levels, and how this might affect market shares and net profits.
Go to the second spreadsheet. It is a "smart" spreadsheet where projected market share is a function of price,
advertising and distribution spending. If you change the planned price, sales (and contribution margin) will
change.
Not surprisingly, as you lower price, volume increases, but the contribution margin decreases. As you increase
advertising or selling effort sales volume increases. These response functions came from managerial judgment,
but could just as easily come from statistical analysis of past data.
Assume these estimates are reasonably good; however, change them if you think it is a good idea.
Bright yellow indicates a cell you should change.
Faded yellow indicates a cell you are allowed to modify. Do this only if you have good reasons to
do so.
Introduction to Marketing Analytics
The Allegro exercise: Simple sheet
Allegro Market Simple Sheet
This year Next year
(current) (projected)
Industry Sales, in Units 2,200,000 2,345,700
Company Market Share 3.0% 3.0%
Price per Unit $250 $250
Variable Cost per Unit $150 $150
Gross Margin per Unit $100 $100
Sales Volume in Units 66,000 69,678
Sales Revenue $16,500,000 $17,419,532
Gross Contribution Margin $6,600,000 $6,967,813
Overhead $3,500,000 $3,900,000
Net Contribution Margin $3,100,000 $3,067,813
Advertising $900,000 $900,000
Sales Force and Distribution $1,000,000 $1,000,000
Marketing Research $100,000 $100,000
Net Profit $1,100,000 $1,067,813
Marketing Mix Decisions
This year Next year
(current) (projected)
Price per Unit $250 $250
Advertising $900,000 $900,000
Sales Force and Distribution $1,000,000 $1,000,000
Introduction to Marketing Analytics
The profit equation
Profit = Revenues − Costs
Sales Volume × Price Variable Costs
Fixed Costs
Industry sales × Market Share
Introduction to Marketing Analytics
Questions
▪ Using the Simple Sheet, what happens when Allegro cuts
advertising and selling effort to 0$ and raises price to
$500/unit? Comment.
▪ Repeat exercise 1 using the Smart Sheet.
▪ Using the Smart Sheet, what is the profit maximizing level of
advertising, selling effort and price? (Hint: Requires Solver)
Would you recommend the firm implement this policy? Why or
why not?
▪ What if the firm's goal was not to maximize profit, but to
maximize market share while maintaining profit at no lower
than last year's level. (Hint: Requires Solver—and be sure to
start with a feasible level of profit). Compare this policy to the
one you found in the previous question.
▪ Comment on the strengths and limitations of a response
function approach (Smart sheet) like this in practice.
Introduction to Marketing Analytics
The Allegro exercise: Simple sheet (Q1)
Allegro Market Simple Sheet
This year Next year
(current) (projected)
Industry Sales, in Units 2,200,000 2,345,700 a
Company Market Share 3.0% 3.0% b
Price per Unit $250 $500 c
Variable Cost per Unit $150 $150 d
Gross Margin per Unit $100 $350 e= c - d
Sales Volume in Units 66,000 69,678 f= a x b
Sales Revenue $16,500,000 $34,839,064 g= c x f
Gross Contribution Margin $6,600,000 $24,387,345 h= e x f
Overhead $3,500,000 $3,900,000 i
Net Contribution Margin $3,100,000 $20,487,345 j= h - i
Advertising $900,000 $0 k
Sales Force and Distribution $1,000,000 $0 l
Marketing Research $100,000 $100,000 m
Net Profit $1,100,000 $20,387,345 n= j - k - l - m
Marketing Mix Decisions
This year Next year
(current) (projected)
Price per Unit $250 $500
Advertising $900,000 $0
Sales Force and Distribution $1,000,000 $0
Introduction to Marketing Analytics
Questions
▪ Using the Simple Sheet, what happens when Allegro cuts
advertising and selling effort to 0$ and raises price to
$500/unit? Comment.
▪ Repeat exercise 1 using the Smart Sheet.
▪ Using the Smart Sheet, what is the profit maximizing level of
advertising, selling effort and price? (Hint: Requires Solver)
Would you recommend the firm implement this policy? Why or
why not?
▪ What if the firm's goal was not to maximize profit, but to
maximize market share while maintaining profit at no lower
than last year's level. (Hint: Requires Solver—and be sure to
start with a feasible level of profit). Compare this policy to the
one you found in the previous question.
▪ Comment on the strengths and limitations of a response
function approach (Smart sheet) like this in practice.
Introduction to Marketing Analytics
The Allegro exercise: Smart sheet (Q2)
Allegro Market Smart Sheet
This year (current) Next year (projected)
Industry Sales, in Units 2,200,000 2,345,700
Company Market Share 3.0% 0.0%
Price per Unit $250 $500
Variable Cost per Unit $150 $150
Gross Margin per Unit $100 $350
Sales Volume in Units 66,000 0
Sales Revenue $16,500,000 $0
Gross Contribution Margin $6,600,000 $0
Overhead $3,500,000 $3,900,000
Net Contribution Margin $3,100,000 -$3,900,000
Advertising $900,000 $0
Sales Force and Distribution $1,000,000 $0
Marketing Research $100,000 $100,000
Net Profit $1,100,000 -$4,000,000
Marketing Mix Decisions
This year (current) Next year (projected)
Price per Unit $250 $500
Advertising $900,000 $0
Sales Force and Distribution $1,000,000 $0
Managerial Judgmental Inputs
What increase in sales will we see with a.....
...1% decrease in price? 2.20%
...1% increase in advertising? 0.2%
...1% increase in sales effort? 0.1%
Introduction to Marketing Analytics
The profit equation
Profit = Revenues − Costs
Sales Volume × Price Variable Costs
Fixed Costs
(Advertising, Distribution)
(Other Fixed Costs)
Industry sales × Market Share
Introduction to Marketing Analytics
Questions
▪ Using the Simple Sheet, what happens when Allegro cuts
advertising and selling effort to 0$ and raises price to
$500/unit? Comment.
▪ Repeat exercise 1 using the Smart Sheet.
▪ Trial and error using the Smart Sheet.
▪ Using the Smart Sheet, what is the profit maximizing level of
advertising, selling effort and price? (Hint: Requires Solver)
Would you recommend the firm implement this policy? Why or
why not?
▪ What if the firm's goal was not to maximize profit, but to
maximize market share while maintaining profit at no lower
than last year's level. (Hint: Requires Solver—and be sure to
start with a feasible level of profit). Compare this policy to the
one you found in the previous question.
▪ Comment on the strengths and limitations of a response
function approach (Smart sheet) like this in practice.
Introduction to Marketing Analytics
Questions
▪ Using the Simple Sheet, what happens when Allegro cuts
advertising and selling effort to 0$ and raises price to
$500/unit? Comment.
▪ Repeat exercise 1 using the Smart Sheet.
▪ Using the Smart Sheet, what is the profit maximizing level of
advertising, selling effort and price? (Hint: Requires Solver)
Would you recommend the firm implement this policy? Why or
why not?
▪ What if the firm's goal was not to maximize profit, but to
maximize market share while maintaining profit at no lower
than last year's level. (Hint: Requires Solver—and be sure to
start with a feasible level of profit). Compare this policy to the
one you found in the previous question.
▪ Comment on the strengths and limitations of a response
function approach (Smart sheet) like this in practice.
Introduction to Marketing Analytics
Using Solver for Optimization
Introduction to Marketing Analytics
The Allegro exercise: Smart sheet (Q3)
Allegro Market Smart Sheet
This year Next year
(current) (projected)
Industry Sales, in Units 2,200,000 2,345,700 a
Company Market Share 3.0% 2.7% b
Price per Unit $250 $275 c
Variable Cost per Unit $150 $150 d
Gross Margin per Unit $100 $125 e= c - d
Sales Volume in Units 66,000 63,571 f= a x b
Sales Revenue $16,500,000 $17,481,989 g= c x f
Gross Contribution Margin $6,600,000 $7,946,358 h= e x f
Overhead $3,500,000 $3,900,000 i
Net Contribution Margin $3,100,000 $4,046,358 j= h - i
Advertising $900,000 $1,589,271 k
Sales Force and Distribution $1,000,000 $1,033,026 l
Marketing Research $100,000 $100,000 m
Net Profit $1,100,000 $1,324,061 n= j - k - l - m
Marketing Mix Decisions
This year Next year
(current) (projected)
Price per Unit $250 $275
Advertising $900,000 $1,589,271
Sales Force and Distribution $1,000,000 $1,033,026
Introduction to Marketing Analytics
Questions
▪ Using the Simple Sheet, what happens when Allegro cuts
advertising and selling effort to 0$ and raises price to
$500/unit? Comment.
▪ Repeat exercise 1 using the Smart Sheet.
▪ Using the Smart Sheet, what is the profit maximizing level of
advertising, selling effort and price? (Hint: Requires Solver)
Would you recommend the firm implement this policy? Why or
why not?
▪ What if the firm's goal was not to maximize profit, but to
maximize market share while maintaining profit at no lower
than last year's level. (Hint: Requires Solver—and be sure to
start with a feasible level of profit). Compare this policy to the
one you found in the previous question.
▪ Comment on the strengths and limitations of a response
function approach (Smart sheet) like this in practice.
Introduction to Marketing Analytics
The Allegro exercise: Smart sheet (Q4)
Allegro Market Smart Sheet
This year (current) Next year (projected)
Industry Sales, in Units 2,200,000 2,345,700
Company Market Share 3.0% 3.8%
Price per Unit $250 $244
Variable Cost per Unit $150 $150
Gross Margin per Unit $100 $94
Sales Volume in Units 66,000 88,467
Sales Revenue $16,500,000 $21,611,347
Gross Contribution Margin $6,600,000 $8,341,344
Overhead $3,500,000 $3,900,000
Net Contribution Margin $3,100,000 $4,441,344
Advertising $900,000 $1,964,533
Sales Force and Distribution $1,000,000 $1,276,811
Marketing Research $100,000 $100,000
Net Profit $1,100,000 $1,100,000
Marketing Mix Decisions
This year (current) Next year (projected)
Price per Unit $250 $244
Advertising $900,000 $1,964,533
Sales Force and Distribution $1,000,000 $1,276,811
Introduction to Marketing Analytics
Questions
▪ Using the Simple Sheet, what happens when Allegro cuts
advertising and selling effort to 0$ and raises price to
$500/unit? Comment.
▪ Repeat exercise 1 using the Smart Sheet.
▪ Using the Smart Sheet, what is the profit maximizing level of
advertising, selling effort and price? (Hint: Requires Solver)
Would you recommend the firm implement this policy? Why or
why not?
▪ What if the firm's goal was not to maximize profit, but to
maximize market share while maintaining profit at no lower
than last year's level. (Hint: Requires Solver—and be sure to
start with a feasible level of profit). Compare this policy to the
one you found in the previous question.
▪ Comment on the strengths and limitations of a response
function approach (Smart sheet) like this in practice.
Introduction to Marketing Analytics
Assignment for next class
▪ Read LRB Chapter 3
▪ Download [Link] or [Link]
▪ Read the Segmentation Tutorial (Enginius)