Market Models Chapter 1
1.2 Verhouten Case p. 2-4
This company sells chocolate bars. The product is often an impulse purchase and influenced by sales
promotions. As a result of which the promotional sales spikes are large. This has consequences for
production and logistics. Management at Verhouten develops a sales forecasting model for
chocolate bars. This so that the production a logistic decisions can be based upon them. A market
research company has been asked to perform the building model. This will be at chain level in which
the chocolate brands: Verhouten, Droste, Baronie and Delicata are joining the research. The
variables are as follows:
- Sales
- Price
- Feature
- Display
- Combined Feature and Display
- Weekly temperature
1.3 Typologies of Maketing Models p. 4-15
Models are classified based on decision models vs. advance marketing knowledge, degree of
explicitness and intended use and level of demand.
1.3.2 Decision Models vs. Advance Marketing Knowledge p.4-7
Decision models:
- Built for primary purpose to support decision making of a marketing manager
- Target audience: Mostly practitioners
- Lead to case-specific insights
Advance Marketing Knowledge:
- Models that aim to advance general marketing knowledge.
- Target audience: Mostly scientists
- Goal is to generalizable insights in marketing phenomena.
New knowledge is acquired when generalizable phenomena are found. This approach is more long-
term oriented and enhances the specificity of a problem.
Ehrenberg (1972) builds model that advance our knowledge. He distinguishes two traditions
Til and EtT. The steps in Til are: (1) construct a theoretical model; (2): test it on a set of data. The
steps for EtT are: (1) establish an empirical pattern; (2) develop a theoretical model.
ET looks if the pattern hols in many different situations, if this is the case, there are generalizable
findings. This can be taken into account in decision models.
Generalizable knowledge can be generated in several ways:
1. Regularities: in customer behaviour (i.e. it revealed that market shares of brands related
positively to the number of households purchasing the brand).
2. Derives from studies that cover Many circumstances. Often panel data is used for this goal.
3. Meta analysis generate generalizable knowledge. It is the statistical analysis of results from
several individual studies for the purpose of generalizing the individual findings. The benefit
of Meta-analysis is that it delivers generalized estimates of various elasticities, quantitative
characteristics of buyer behaviour and an assessment of moderators.
Empirical generalizations are for example that the price elasticity is on average -2.62 (Bijmolt et al.,
2005). However it doesn’t always result in numbers. Kaul en Wittink (1995) describe the relation
between advertising and price sensitivity as follows:
- Increase of price advertising -> higher price sensitivity of consumers
- Use price advertising -> lower prices
- Increase non-pricing advertising -> lower price sensitivity of consumers
1
1.2 Verhouten Case p. 2-4
This company sells chocolate bars. The product is often an impulse purchase and influenced by sales
promotions. As a result of which the promotional sales spikes are large. This has consequences for
production and logistics. Management at Verhouten develops a sales forecasting model for
chocolate bars. This so that the production a logistic decisions can be based upon them. A market
research company has been asked to perform the building model. This will be at chain level in which
the chocolate brands: Verhouten, Droste, Baronie and Delicata are joining the research. The
variables are as follows:
- Sales
- Price
- Feature
- Display
- Combined Feature and Display
- Weekly temperature
1.3 Typologies of Maketing Models p. 4-15
Models are classified based on decision models vs. advance marketing knowledge, degree of
explicitness and intended use and level of demand.
1.3.2 Decision Models vs. Advance Marketing Knowledge p.4-7
Decision models:
- Built for primary purpose to support decision making of a marketing manager
- Target audience: Mostly practitioners
- Lead to case-specific insights
Advance Marketing Knowledge:
- Models that aim to advance general marketing knowledge.
- Target audience: Mostly scientists
- Goal is to generalizable insights in marketing phenomena.
New knowledge is acquired when generalizable phenomena are found. This approach is more long-
term oriented and enhances the specificity of a problem.
Ehrenberg (1972) builds model that advance our knowledge. He distinguishes two traditions
Til and EtT. The steps in Til are: (1) construct a theoretical model; (2): test it on a set of data. The
steps for EtT are: (1) establish an empirical pattern; (2) develop a theoretical model.
ET looks if the pattern hols in many different situations, if this is the case, there are generalizable
findings. This can be taken into account in decision models.
Generalizable knowledge can be generated in several ways:
1. Regularities: in customer behaviour (i.e. it revealed that market shares of brands related
positively to the number of households purchasing the brand).
2. Derives from studies that cover Many circumstances. Often panel data is used for this goal.
3. Meta analysis generate generalizable knowledge. It is the statistical analysis of results from
several individual studies for the purpose of generalizing the individual findings. The benefit
of Meta-analysis is that it delivers generalized estimates of various elasticities, quantitative
characteristics of buyer behaviour and an assessment of moderators.
Empirical generalizations are for example that the price elasticity is on average -2.62 (Bijmolt et al.,
2005). However it doesn’t always result in numbers. Kaul en Wittink (1995) describe the relation
between advertising and price sensitivity as follows:
- Increase of price advertising -> higher price sensitivity of consumers
- Use price advertising -> lower prices
- Increase non-pricing advertising -> lower price sensitivity of consumers
1
