Market Models (20/21)
1. MARKETING MODEL BUILDING
Marketing models will create generalizable insights in marketing phenomena, whereas decision
models generate case-specific insights. Ehrenberg distinguished two kinds of research traditions in
marketing:
1. Theoretical-in-Isolation (TiL); construct a theoretical model test it on data
2. Empirical-then-Theoretical (EtT); establish empirical pattern develop theoretical model
Generalizable knowledge about market phenomena can be generated in several ways
1. Find regularities in customer behaviour
2. Studies that cover many circumstances and relatively long time periods. Often use panel data.
3. Meta analysis refers to the statistical analysis of results from several individual studies.
What are the basic elements of every model?
Inputs Process Outputs
Inputs: marketing variables
Process: consumers, competitors and government might react to the inputs
Outputs: sales, profit, market share
Goal = understand the reality
What = model is a stylized representation of reality
Needed = model should contain the most important elements (but is never complete)
When = design, optimize, evaluate; use models to investigate several questions in DOE cycle
Types of models
1. Implicit models: intuition and experiences are the basis for a solution
2. Verbal models: state in words the important elements surrounding a problem
3. Iconic models: sketches, prototypes, virtual reality
4. Analog models: focus on key elements, not details, flow charts, not resemble reality
5. Symbolic models
Represent ideas using coding (e.g. numbers, formulas, words, music notes)
Symbolic models are abstract representations of reality
6. Formalized models
Logical flow model
o Extension of the verbal model by the use of a diagram
o Known as graphical or a conceptual model
Formalized mathematical model
o Specifying relations between some explanatory variables and some effect
(criterion) variable(s)
7. Numerically specified models
Various components and their interrelations are quantified
Most appropriate representations of real-world systems
Can help to examine the consequences of alternative courses of actions
Numerically specified model gives management the opportunity to explore the
consequences of a myriad of actions.
But: costly!
,
,Why models?
Forces one to abstract away from details
Puts focus on the key elements only
Less ad-hoc than one-time analyses
Include more than one factor simultaneously
Can be used for very simple analyses and very complex analyses
Can be used to describe / predict / prescribe
o Model is not limited to observed cases; generalizable insights in marketing phenomena
Direct benefits for marketing manager
“Better” decision making: contributing to the fulfillment of the company’s goals
1. Model can help in the allocation process of a promotion budget. For example, a firm might
overspend on advertising, a model might help to adjust the spending level which will result in
more profit.
2. Marketing managers are often faced with the question whether they should increase
advertising or decrease their pricing. A model may incorporate the empirical generalization
that the optimal price depends on advertising expenditures
3. Systematizing information on past data into a model may result in a pricing strategy that will
lead to an increase in expected profit. Improve future choices.
Indirect benefits
Indirect benefits = those that are not related directly to the reasons for which the model was built
1. Model forces to explicate how the market works. This explication will often lead to an
improved understanding of the role of advertising and how advertising effectiveness might
depend on a variety of other marketing and environmental conditions.
2. Models may work as problem-finding instrument. Problems may emerge after a model has
been developed if the model outcomes are contrary to expectations. Managers may identify
problems by discovering differences between their perception of the environment and a model
of that environment.
3. Model can be instrumental in improving the process by which decision makers deal with
existing information.
4. Models can help managers decide what information should be collected. Thus models may
lead to improved data collection. And guide research by identifying areas in which
information is lacking.
5. Model allows management to pinpoint changes in the environment faster than is possible
otherwise.
6. A model may result in beneficial reallocation of management time.
Why not always models?
Time consuming (e.g. learn the data, think about the model, estimate & validate the model)
Data availability (e.g. not enough data results in an unsuccessful model)
Costly
When do you use models?
In the cycle of design, optimize and evaluation, you might use models everywhere.
Design: what combination of product is an important driver of customer value?
Optimize: whom to target with what proposition?
Evaluate: what media channel was more effective than others?
Use models to investigate several questions raised in the D.O.E cycle.
, Four steps of marketing model building
Step 1: Specification
come up with formula that captures the idea about
marketing phenomena
1.1.) Specifying variables to be included in the model and
making a distinction between those to be explained (DV)
and those providing explanation (IVs)
1.2.) Specifying the functional relationship between
variables. For example, the effects of the explanatory
variables can be linear or non-linear, immediate and/or
lagged, additive or multiplicative.
!! Data is not the basis for specifying the model, theory is
mostly used to specify the model. A model can be specified without looking at the data at all.
Step 2: Estimation
obtaining estimates for the parameters specified in step 1.
Apart form data collection issues, we need to identify techniques to be applied for extracting
estimates of the model parameters from the data collected. Data type & scale of dependent
variable impact the technique for data estimation.
Step 3: Validation
checking the assumptions that we assume while creating the model, are there reliable
outcomes? Assessing the quality or the success of the model.
3.1.) degree to which the results are in accordance with theoretical expectations (face
validity)
3.2.) degree to which the results satisfy statistical criteria or tests (statistical validity)
3.3) degree to which the result is relevant to the original purpose
3.4) check the assumptions that underlie the estimations
Step 4: Use
Other steps in marketing model building include:
1. Opportunity identification:
evaluate whether a model can improve decision making. Ideally the model builder and
manager work together to define the problem, to agree on an approach and to determine that
the expected benefits exceed the costs of model building.
2. Model Purpose:
define the intended use (e.g. to obtain accurate sales forecasts), define the level of demand
3. Model Scope:
the model-building effort might solely focus on a single decision variable. Or a broader set of
decisions. If the effect of promotion influences both sales and therefore adjustments in
production / distribution, then we need a model with a broader scope.
4. Data Availability:
the increasing availability of large amounts of data is one reason for building a model.
8. Cost-Benefit Consideration:
cost and benefits should be known with a fair amount of precision. Re-examine the cost-
benefit trade-off.
10. Updating:
model could require modifications as the manager develops a better understanding of the
marketplace.
1. MARKETING MODEL BUILDING
Marketing models will create generalizable insights in marketing phenomena, whereas decision
models generate case-specific insights. Ehrenberg distinguished two kinds of research traditions in
marketing:
1. Theoretical-in-Isolation (TiL); construct a theoretical model test it on data
2. Empirical-then-Theoretical (EtT); establish empirical pattern develop theoretical model
Generalizable knowledge about market phenomena can be generated in several ways
1. Find regularities in customer behaviour
2. Studies that cover many circumstances and relatively long time periods. Often use panel data.
3. Meta analysis refers to the statistical analysis of results from several individual studies.
What are the basic elements of every model?
Inputs Process Outputs
Inputs: marketing variables
Process: consumers, competitors and government might react to the inputs
Outputs: sales, profit, market share
Goal = understand the reality
What = model is a stylized representation of reality
Needed = model should contain the most important elements (but is never complete)
When = design, optimize, evaluate; use models to investigate several questions in DOE cycle
Types of models
1. Implicit models: intuition and experiences are the basis for a solution
2. Verbal models: state in words the important elements surrounding a problem
3. Iconic models: sketches, prototypes, virtual reality
4. Analog models: focus on key elements, not details, flow charts, not resemble reality
5. Symbolic models
Represent ideas using coding (e.g. numbers, formulas, words, music notes)
Symbolic models are abstract representations of reality
6. Formalized models
Logical flow model
o Extension of the verbal model by the use of a diagram
o Known as graphical or a conceptual model
Formalized mathematical model
o Specifying relations between some explanatory variables and some effect
(criterion) variable(s)
7. Numerically specified models
Various components and their interrelations are quantified
Most appropriate representations of real-world systems
Can help to examine the consequences of alternative courses of actions
Numerically specified model gives management the opportunity to explore the
consequences of a myriad of actions.
But: costly!
,
,Why models?
Forces one to abstract away from details
Puts focus on the key elements only
Less ad-hoc than one-time analyses
Include more than one factor simultaneously
Can be used for very simple analyses and very complex analyses
Can be used to describe / predict / prescribe
o Model is not limited to observed cases; generalizable insights in marketing phenomena
Direct benefits for marketing manager
“Better” decision making: contributing to the fulfillment of the company’s goals
1. Model can help in the allocation process of a promotion budget. For example, a firm might
overspend on advertising, a model might help to adjust the spending level which will result in
more profit.
2. Marketing managers are often faced with the question whether they should increase
advertising or decrease their pricing. A model may incorporate the empirical generalization
that the optimal price depends on advertising expenditures
3. Systematizing information on past data into a model may result in a pricing strategy that will
lead to an increase in expected profit. Improve future choices.
Indirect benefits
Indirect benefits = those that are not related directly to the reasons for which the model was built
1. Model forces to explicate how the market works. This explication will often lead to an
improved understanding of the role of advertising and how advertising effectiveness might
depend on a variety of other marketing and environmental conditions.
2. Models may work as problem-finding instrument. Problems may emerge after a model has
been developed if the model outcomes are contrary to expectations. Managers may identify
problems by discovering differences between their perception of the environment and a model
of that environment.
3. Model can be instrumental in improving the process by which decision makers deal with
existing information.
4. Models can help managers decide what information should be collected. Thus models may
lead to improved data collection. And guide research by identifying areas in which
information is lacking.
5. Model allows management to pinpoint changes in the environment faster than is possible
otherwise.
6. A model may result in beneficial reallocation of management time.
Why not always models?
Time consuming (e.g. learn the data, think about the model, estimate & validate the model)
Data availability (e.g. not enough data results in an unsuccessful model)
Costly
When do you use models?
In the cycle of design, optimize and evaluation, you might use models everywhere.
Design: what combination of product is an important driver of customer value?
Optimize: whom to target with what proposition?
Evaluate: what media channel was more effective than others?
Use models to investigate several questions raised in the D.O.E cycle.
, Four steps of marketing model building
Step 1: Specification
come up with formula that captures the idea about
marketing phenomena
1.1.) Specifying variables to be included in the model and
making a distinction between those to be explained (DV)
and those providing explanation (IVs)
1.2.) Specifying the functional relationship between
variables. For example, the effects of the explanatory
variables can be linear or non-linear, immediate and/or
lagged, additive or multiplicative.
!! Data is not the basis for specifying the model, theory is
mostly used to specify the model. A model can be specified without looking at the data at all.
Step 2: Estimation
obtaining estimates for the parameters specified in step 1.
Apart form data collection issues, we need to identify techniques to be applied for extracting
estimates of the model parameters from the data collected. Data type & scale of dependent
variable impact the technique for data estimation.
Step 3: Validation
checking the assumptions that we assume while creating the model, are there reliable
outcomes? Assessing the quality or the success of the model.
3.1.) degree to which the results are in accordance with theoretical expectations (face
validity)
3.2.) degree to which the results satisfy statistical criteria or tests (statistical validity)
3.3) degree to which the result is relevant to the original purpose
3.4) check the assumptions that underlie the estimations
Step 4: Use
Other steps in marketing model building include:
1. Opportunity identification:
evaluate whether a model can improve decision making. Ideally the model builder and
manager work together to define the problem, to agree on an approach and to determine that
the expected benefits exceed the costs of model building.
2. Model Purpose:
define the intended use (e.g. to obtain accurate sales forecasts), define the level of demand
3. Model Scope:
the model-building effort might solely focus on a single decision variable. Or a broader set of
decisions. If the effect of promotion influences both sales and therefore adjustments in
production / distribution, then we need a model with a broader scope.
4. Data Availability:
the increasing availability of large amounts of data is one reason for building a model.
8. Cost-Benefit Consideration:
cost and benefits should be known with a fair amount of precision. Re-examine the cost-
benefit trade-off.
10. Updating:
model could require modifications as the manager develops a better understanding of the
marketplace.
