Cluster Analysis for Market Segmentation

Customer/ Market segmentation: dividing a market into distinct groups of

buyers who have different, needs, characteristics or behaviour and who might
require separate products or marketing programmes/strategies/mixes.

A market segment is a group of consumers who respond in a similar way to a

given set of marketing efforts.
Customer Segmentation is the division of potential customers in a given
market into discrete groups. That division is based on customers having
similar enough:

1. Needs, so that a single whole product can satisfy them.

2. Buying characteristics, responses to messaging, marketing channels,

and sales channels, so that a single go-to-market approach can be used
to sell to them competitively and economically.

“One size fits all” usually doesn’t work (all potential customers are
not created equal
There are different ways to segment a market:

1.Geographic segmentation: dividing a market into different

geographical units, such as nations, states, regions, counties, cities or
even neighbourhoods.

2.Demographic segmentation: dividing the market into different

segments based on variables such as age, gender, family size, family life
cycle, income, occupation education, religion, race, generation and
nationality. 
 Age and life-cycle segmentation is dividing a market into different
age and life-cycle groups. 
 Gender segmentation means dividing a market based on gender,
while income segmentation divides a market based on income
levels.
1.Psychographic segmentation: dividing a market into different segments
based on social class, lifestyle or personality characteristics.

2.Behavioural segmentation: dividing a market into segments based on

consumer knowledge, attitudes, uses or responses to a product. This can
be done via 

 Occasion segmentation: dividing the market according to occasions

when buyers get the idea to buy, actually making their purchase or use
the purchased items. 

 Benefit segmentation: dividing the market according to the benefits

that customers seek from the product.

 Markets can also be segmented based on user states, usage rate and
loyalty status.
5
Marketers often use multiple segmentation bases to identify a well-defined target group.

A segmentation base is defined as: a set of variables or characteristics used to assign

potential customers to homogenous groups.
Segmentation bases

General Product-specific

Observable features Behavioral characteristics

of the physical and (user status, loyalty
Observable social environment status, usage rate)
(esp. demographics) Usage situations

Values, lifestyles and Awareness

Latent psychographics, Product attributes and
personality variables benefits
Willingness to buy
 For segmentation to be effective, market segments must be measurable, accessible,
substantial, differentiable and actionable.

 Business markets can be segmented with the same variables, but also with additional ones,
such as customer operating characteristics, purchasing approaches and situational factors.
International markets can be segmented using a combination of variables. 

 Intermarket segmentation (cross-market segmentation): forming segments of consumers

who have similar needs and buying behaviour even though they are located in different
countries

8
Segmentation in the real world

 In practice, we have
 Many potential customers
 Many segmentation variables
What to do?

 Basic question: How can objects (customers, brands, stores, etc.) be

grouped such that objects within the same cluster are similar and
objects in different clusters are dissimilar?

Cluster analysis to the rescue!

Often the marketer needs to categorize objects into groups (or clusters) so
that the objects in each group are similar, and the objects in each group are
substantially different from the objects in the other groups.

Here are some examples:

 When a company test markets a new cosmetic, it may want to group cities of a
country into groups that are similar on demographic attributes such gender
distribution, median age, unemployment rate, and median income level.

 A marketing analyst at Coca-Cola wants to segment the soft drink market based
on consumer preferences for price sensitivity, preference of diet versus regular
soda, and preference of Coke versus Pepsi.

 Microsoft might cluster its corporate customers based on the price a given
customer is willing to pay for a product. For example, there might be a cluster of
construction companies that are willing to pay a lot for Microsoft Project but not
so much for Power Point.
Cluster Analysis
 It is a class of techniques used to classify cases into groups that are
relatively homogeneous within themselves and heterogeneous
between each other, on the basis of a defined set of variables.

 Cluster Analysis is useful to identify market segments, competitors in

market structure analysis, matched cities in test market etc.

 General types of similarity measure are available:

 •Distance measures
 •Correlation measures
 •Agreement or matching-type measure

11
Cluster Analysis

12
Cluster Analysis

13
Cluster Analysis
 The goal of Clustering: Identifying the set of objects with similar
characteristics. We want that:
 (1) The objects in the same group are more
 similar to each other
 (2) Than to those in other groups

For what purpose?

  Identify underlying structures in the data
  Summarize behaviors or characteristics
 Assign new individuals to groups
  Identify totally atypical objects

14
Cluster Analysis in view of Market Segmentation
 • Market segmentation. E.g. clustering of consumers according to their
attribute preferences

 • Understanding buyers behaviours.

 Consumers with similar behaviours/characteristics are clustered

 • Identifying new product opportunities.

 Clusters of similar brands/products can help identifying competitors /
market opportunities

 • Reducing data. E.g. in preference mapping

Cluster analysis

 In segmentation, the objects of interest are customers and similarity

is assessed in terms of relevant segmentation variables;
 Issues in cluster analysis:
 How is similarity measured?
 How are clusters formed?
 How many clusters should be distinguished?
 How should the clusters be interpreted?
17
Cluster Analysis for Market Segmentation using K-Means

 k-means cluster analysis is an algorithm that groups similar objects

into groups called clusters.

 The endpoint of cluster analysis is a set of clusters, where each

cluster is distinct from each other cluster, and the objects within
each cluster are broadly similar to each other.
 K-means clustering is a simple unsupervised learning algorithm that is
used to solve clustering problems.
 It follows a simple procedure of classifying a given data set into a
number of clusters, defined by the letter "k," which is fixed
beforehand.
 The clusters are then positioned as points and all observations or data
points are associated with the nearest cluster, computed, adjusted and
then the process starts over using the new adjustments until a desired
result is reached.

The algorithm:
1.K points are placed into the object data space representing the initial
group of centroids.
2.Each object or data point is assigned into the closest k.
3.After all objects are assigned, the positions of the k centroids are
recalculated.
4.Steps 2 and 3 are repeated until the positions of the centroids no longer
move
The required data for k-means cluster analysis
 k-means cluster analysis is performed on a table of raw data, where
each row represents an object and the columns represent quantitative
characteristics of the objects.

 These quantitative characteristics are called clustering variables. For

example, in the table below there are 18 objects, and there are two
clustering variables, x, and y. In a real-world application, there will
typically be many more objects and more variables.

 For example, in market segmentation, where k-means is used to find

groups of consumers with similar needs, each object is a person and
each variable is commonly a rating of how important various things
are to consumers (e.g., quality, price, customer service,
convenience).
Step 1: Specify the number of clusters (k). The first step in k-means is to specify the
number of clusters, which is referred to as k. Traditionally researchers will conduct k-means
multiple times, exploring different numbers of clusters (e.g., from 2 through 10).

Step 2: Allocate objects to clusters. The most straightforward approach is to randomly

assign objects to clusters, but there are many other approaches (e.g., using hierarchical
clustering).

Step 3: Compute cluster means. For each cluster, the average value is computed for each
of the variables.

Step 4: Allocate each observation to the closest cluster center.

Step 5: Repeat steps 3 and 4 until the solution converges. (i.e., reallocating observations
and updating means cannot improve the solution). When you have more data, more iterations
are typically required (i.e., steps 3 and 4 are repeated until no respondents change clusters).
Steps:
•Step 1: Choose the number of clusters k.>>>
•Step 2: Make an initial selection of k centroids

•Step 3: Assign each data element to its nearest centroid (in this way k clusters are formed
one for each centroid, where each cluster consists of all the data elements assigned to that
centroid)

•Step 4: For each cluster make a new selection of its centroid

•Step 5: Go back to step 3, repeating the process until the centroids don’t change (or some
other convergence criterion is met)
There are various choices available for each step in the process.
An alternative version of the algorithm is as follows:
•Step 1: Choose the number of clusters k

•Step 2: Make an initial assignment of the data elements to the k clusters

•Step 3: For each cluster select its centroid

•Step 4: Based on centroids make a new assignment of data elements to

the k clusters

•Step 5: Go back to step 3, repeating the process until the centroids

don’t change (or some other convergence criterion is met)
The algorithm works as follow:

•Step 1: Choose groups in the feature plan randomly

•Step 2: Minimize the distance between the cluster center and the different
observations (centroid). It results in groups with observations
•Step 3: Shift the initial centroid to the mean of the coordinates within a
group.
•Step 4: Minimize the distance according to the new centroids. New boundaries
are created. Thus, observations will move from one group to another
•Repeat until no observation changes groups
NB:
Suggested approach
1. First perform a hierarchical method to define the number of clusters

2. Then use the k-means procedure to actually form the clusters