Transcript Targeting Research

Targeting Research: Segmentation


Birds of a feather flock together, i.e. people
with similar characteristics tend to exhibit
similar behaviors
Characteristics on which to segment include:
–
–
–
–
Demographics
Geographics
Psychographics
Attitudes and behaviors relevant to a brand of
product category
Demographic Segmentation

Age: only useful when appropriate ranges are used
to define age group segments







Too broad (18-45) blurs differences between segments
Too narrow (18-20) results in too many small segments
Gender
Household characteristics
Life stage: combination of all of the above (see page
447)
Race and Ethnicity
Social Class: comprised of education, occupation,
and income
Geographic Segmentation

Region
–
–

Population Size
–


East, North, West, South
Pacific, Mountain, West North Central, West South Central,
East North Central, East South Central, South Atlantic,
Middle Atlantic, New England
MSA, PMSA, CMSA
Population Density
Climate
Psychographic Segmentation

Attitudes and Values
–
–

Motivations
–

Different attitudinal segments typically require
different advertising messages
Different attitudinal segments often have different
media habits, requiring different media plans
Reasons for purchase, i.e. pregnancy tests
Lifestyle
Category and Brand Related Attitudes
and Behaviors

Product Usage
–
–
–
–


Brand usage
Product usage
Quantity of consumption
Situation of consumption
Brand Loyalty
Benefit
Performing Segmentation Research
(figure 19.2, p. 456)

Determine basis for segmentation and descriptive variables
–
–




Identify range of segmentation and descriptive variables
Collect data
Form groups using cluster analysis
Describe each cluster or segment using descriptive variables
–



Segmentation variables are the variables used to create subgroups, or
market segments
Descriptive variables are used to describe differences between the
subgroups
Radar charts are very useful for this
Name and describe each segment
Relate segments to particular brand behaviors
Select a segment to target
Determine the Basis for Segmentation

Select segmentation variables
–
–
–
Relevant to product category
Related to category, product, or brand related
perceptions or behaviors
Have a high likelihood of being causal factors of
category , product or brand choice and usage
Identify Descriptive Variables



Equivalent of dependent variables
Used to further describe or identify segment
members
Necessary to provide an understanding of
the internal characteristics of each segment
Identify the Range of Segmentation
and Descriptive Variables

There are no “Standard” categories or
ranges for segmentation variables
–


Specific to the researchers informational needs
Language use
Use to guide content and form attitude or
other scales
Sample and Survey the Population

Random or stratified random sampling

Mail or telephone surveys

Typically likert scale questions
For groups using cluster analysis



Respondents with similar patterns of
response to segmentation variables are
placed in the same group
Frequencies (cross tabs) on segmentation
variables used to summarize differences
between groups
Use a radar chart to present a better
descriptor of segments
Example Radar Chart from Table 19.5
data (partial)
Survival
40
Nurturance
30
Housing
20
10
Belonging
0
Status
Functionality
Entertainment
Security
Final Steps




Describe each sentence in narrative, without
using the numbers
Describe segments further based on
descriptive variables
Relate Segments to Brand specific behaviors
Select a segment
Cluster Analysis in SPSS

Identify segmentation variables
–



Be sure to tell SPSS to save cluster assignments as a
variable
SPSS automatically cross-tabulates data on each
segmentation variable for each cluster
Identify defining characteristics for each cluster, now
called a segment
Compare meaningful descriptive variables across
segments
Group Practice

Page 479, Exercise #3

Each group pick three product categories
For each, determine the most important
segmentation type
Identify at least four segmentation and descriptive
measures
Craft the specific items to measure those variables



Q-Methodology



Method of targeting using a large set of
statements (q-sample) that respondents are
asked to rank based on their agreement with
the statements
Factor analysis is used to identify groups
Typically results in groups based on
perceptions and attitudes
Q-Methodology Process







Develop, Edit, and Select Q-sort statements
Select type of sort
Determine Sample Characteristics
Create Instructions
Administer and Record the Sort
Analyze data
Apply findings to decision
Selection of statements






Prior projects
Prior survey research
Responses to open ended, in depth interview
questions
Volunteered comments in focus groups
Responses to projective stimuli such as
sentence completion and picture projection
Brainstorming or idea-generation sessions
Statement Selection

Range of statements
–
–



Must be unbiased, i.e. same number of positive,
negative and neutral statements
If more than one concept, equal number of
statements for each concept
Item appropriateness
Item Clarity
Item independence
Types of Q-sorts

Forced
–

Unforced
–

A predetermined number of cards are to be
placed on each point of the scale
Any number of statements may be placed on any
point of the scale
Little agreement about which type is better,
so use whichever you prefer
Select the sample

Due to the difficult nature of the task,
samples tend to be small, yet representative
of the population
–
–
–
Number of popluation variables that are important
Number of required replications (between 3-5)
See table 20.1, page 493
Create Instructions, Administer Sort

Establish and communicate sorting criteria
–

“sort the statements to reflect how much you think
each is ‘most like my point of view’ or ‘most unlike
my point of view’.”
Record placement of each statement using
numerical scale position
–
Depending on the number of statements, you may
have anywhere from 7 to as many as 11 points on
the scale
Analysis of Q-sort data




Variables are individuals, cases are statements
Factor analysis then groups the individuals
Examine attitudinal characteristics by comparing
mean scores of individual statements across factors
(groups)
Once group attitudes are identified, demographics,
brand related behaviors or other salient
characteristics of individuals in each group are
compared
Group Practice


Page 501-502
Application exercises 1, 2, and 4