The marketing concept: the key to achieving organizational goals is being more effective than
competitors in creating, delivering, and communicating superior customer value to your target
markets.
Why use quantitative marketing models?
Marketers are held accountable
Better business decisions with empirical evidence
Businesses generate and collect data
What is a model?
Simplified representation of the world
Build to help us understand the world and
To make predictions
Identifying competition
Differentiation: what you do to an offering:
o Creating tangible or intangible differences on one or more attributes between a focal
offering and its main competitors
Positioning:
o A set of strategies a firm develops to differentiate its offering in the minds of its
target customers. Successful positioning will result in the offering occupying a
distinct, important, and sustainable position in the minds of the target customers.
Levels of competition: implications for product strategy
Product form Convince customers that the brand is better than others
Product category Convince customers that the product form is best in the category
Generic Convince customers that the product category is the best way to satisfy needs
Budget convince customers that generic benefits are the most appropriate way to spend
their money
Brand switching analysis
Method: Answer the following questions:
o Do you have a smart phone? If yes, which brand?
o If you had to buy a new smartphone, what brand would you buy?
Switching matrix from current to future
o Expected count: (64/180 * 54/180 * 180 = 19.2)
,Lecture 2
Market segmentation is the subdividing of a market into distinct subsets of customers
Segments: Members are different between segments but similar within
STP: segmentation targeting positioning
Desirable criteria for segments:
Measurable: possible to obtain data
Sizeable: substantial
Identifiable: titles for segments
Reachable: accessible
Respond differently: differentiable
Coherent: homogeneous
Stable
Actionable: can we serve them
Segmentation process
1. Articulate a strategic rationale for segmentation
o Why are we segmenting this market?
2. Select a set of (needs-based) segmentation variable most useful for achieving the strategic
goals
3. Select a cluster analysis procedure for (dis)aggregating customers into segments
4. Group customers into a defined number of different segments; describe them with
descriptor data
5. Target the segments that will best serve the firm’s strategy, given its capabilities and the
likely reactions of competitors
Distinguish between bases and descriptors
Bases: characteristics that tell us why segments differ / respond differently
o Needs, preferences, decision processes
Descriptors: characteristics that help us identify and reach segments
, Cluster analysis for segmenting markets
Select variables for analysis: these should be based on their potential for providing
meaningful ways to define the needs of customers
Define an overall measure to assess the similarity of customers on the basis of the needs
variables
Group customers with similar needs
Select the number of segments using numeric and strategic criteria, and your judgement
Profile the needs of the selected segments (using cluster means)
Measures of similarity:
Associations: such as correlations
o Same pattern, go up and down together
Distances: squared differences between variables
o Closest / crossing
Group customers with similar needs
Clustering algorithms (see book for details)
Hierarchical clustering (all customers in their own segment, end with every customer in one
segment, start in wrong segment)
o Single link clustering: closest distance
o Complete link: furthest apart between the 2 clusters
o Average link: TERUGKIJKEN!
o Ward’s method: minimize variance within clusters
Non-hierarchical / K-means clustering
o Specify the number of clusters in advance
o Specify cluster means or random assignment of customers to segments
o Reassignment of customer to closer clusters
o Recalculate centroids (cluster means) and iterate
Choose the number of segments (hierarchical)
Agglomeration coefficient / dendogram (hierarchical clustering)
o Gives some clue about the number of market segments to choose
o Stop when high percentage increase of agglomeration coefficient
Trial and error; interpretation of various solutions
Choose the number of segments (non-hierarchical)
Previous segmentation studies; preliminary knowledge
Trial and error; interpretation of various solutions
Describe market segments
Means per cluster for
Cluster bases
Descriptor variables
Test for significant differences
ANOVA analysis (compare means between all clusters)
T-tests (compare means of 2 clusters)
Cross-tab and Chi-square (actual vs expected frequencies)
competitors in creating, delivering, and communicating superior customer value to your target
markets.
Why use quantitative marketing models?
Marketers are held accountable
Better business decisions with empirical evidence
Businesses generate and collect data
What is a model?
Simplified representation of the world
Build to help us understand the world and
To make predictions
Identifying competition
Differentiation: what you do to an offering:
o Creating tangible or intangible differences on one or more attributes between a focal
offering and its main competitors
Positioning:
o A set of strategies a firm develops to differentiate its offering in the minds of its
target customers. Successful positioning will result in the offering occupying a
distinct, important, and sustainable position in the minds of the target customers.
Levels of competition: implications for product strategy
Product form Convince customers that the brand is better than others
Product category Convince customers that the product form is best in the category
Generic Convince customers that the product category is the best way to satisfy needs
Budget convince customers that generic benefits are the most appropriate way to spend
their money
Brand switching analysis
Method: Answer the following questions:
o Do you have a smart phone? If yes, which brand?
o If you had to buy a new smartphone, what brand would you buy?
Switching matrix from current to future
o Expected count: (64/180 * 54/180 * 180 = 19.2)
,Lecture 2
Market segmentation is the subdividing of a market into distinct subsets of customers
Segments: Members are different between segments but similar within
STP: segmentation targeting positioning
Desirable criteria for segments:
Measurable: possible to obtain data
Sizeable: substantial
Identifiable: titles for segments
Reachable: accessible
Respond differently: differentiable
Coherent: homogeneous
Stable
Actionable: can we serve them
Segmentation process
1. Articulate a strategic rationale for segmentation
o Why are we segmenting this market?
2. Select a set of (needs-based) segmentation variable most useful for achieving the strategic
goals
3. Select a cluster analysis procedure for (dis)aggregating customers into segments
4. Group customers into a defined number of different segments; describe them with
descriptor data
5. Target the segments that will best serve the firm’s strategy, given its capabilities and the
likely reactions of competitors
Distinguish between bases and descriptors
Bases: characteristics that tell us why segments differ / respond differently
o Needs, preferences, decision processes
Descriptors: characteristics that help us identify and reach segments
, Cluster analysis for segmenting markets
Select variables for analysis: these should be based on their potential for providing
meaningful ways to define the needs of customers
Define an overall measure to assess the similarity of customers on the basis of the needs
variables
Group customers with similar needs
Select the number of segments using numeric and strategic criteria, and your judgement
Profile the needs of the selected segments (using cluster means)
Measures of similarity:
Associations: such as correlations
o Same pattern, go up and down together
Distances: squared differences between variables
o Closest / crossing
Group customers with similar needs
Clustering algorithms (see book for details)
Hierarchical clustering (all customers in their own segment, end with every customer in one
segment, start in wrong segment)
o Single link clustering: closest distance
o Complete link: furthest apart between the 2 clusters
o Average link: TERUGKIJKEN!
o Ward’s method: minimize variance within clusters
Non-hierarchical / K-means clustering
o Specify the number of clusters in advance
o Specify cluster means or random assignment of customers to segments
o Reassignment of customer to closer clusters
o Recalculate centroids (cluster means) and iterate
Choose the number of segments (hierarchical)
Agglomeration coefficient / dendogram (hierarchical clustering)
o Gives some clue about the number of market segments to choose
o Stop when high percentage increase of agglomeration coefficient
Trial and error; interpretation of various solutions
Choose the number of segments (non-hierarchical)
Previous segmentation studies; preliminary knowledge
Trial and error; interpretation of various solutions
Describe market segments
Means per cluster for
Cluster bases
Descriptor variables
Test for significant differences
ANOVA analysis (compare means between all clusters)
T-tests (compare means of 2 clusters)
Cross-tab and Chi-square (actual vs expected frequencies)
