Class Notes (1,100,000)
CA (620,000)
Ryerson (30,000)
GEO (1,000)
GEO 301 (20)
Lecture 9

GEO 301 Lecture Notes - Lecture 9: Census Tract, Market Segmentation, Experian

Course Code
GEO 301
Joseph Aversa

This preview shows half of the first page. to view the full 3 pages of the document.
L9: Geodemographics and Geodemographic Profiles
A. Market segmentation
B. Providers of commercial geodemographics cluster systems
C. Geodemographics methodology
D. Critiques of geodemographics
A. Market Segmentation
Basic Assumptions
1. Tobler’s First Law of Geography “things are similar, but close things are more similar than distant things”
2. “you are where you live”
- similar people with similar tastes cluster in neighbourhoods
- where you live determines whether or not you are likely to be a good market prospect
3. clustering
- people living in neighbourhoods share many demographic and other characteristics which vary from neighbourhood to neighbourhood
- many different variables can be entered into a complex statistical program in SPSS or SAS which looks for areas or neighbourhoods where
the different variables tend to have similar values
- the program then joins these similar areas into a single cluster
- different types of populations who share similar values are lumped together into their own clusters
- the clusters are meant to be as different from each other as possible to designate different types of neighbourhoods
What is the purpose of segmentation?
- identify areas with clusters of customers which are:
o large enough for marketing purposes
o homogeneous, targetable and reachable populations
- used in many business strategies
o customer profiling, cross-selling, site selection, strategic planning, media buying
Types of Market Segmentation
Types of market segmentation cluster systems:
A. Simple demographics
- Simple demographics clusters people by a number of demographic variables available from the census and other sources such as age, sex,
language and ethnicity, occupation, educational attainment, etc.
o may or may not include a geographic component
- Eg. to market financial services to university graduates, use alumni association lists and market to every grad, or only those living in the
Toronto area
B. Special interest demographics
- Special interest demographics clusters people according to demographic, socioeconomic and other variables from the census and other sources
which are thought to have a direct relationship with the behaviours you are trying to examine
- eg. a demographic segmentation meant to predict health outcomes will focus on those variable which are known to influence the health status
of population groups: age, sex, family structure, housing, income, occupation, employment, social stability, ethnicity, etc.
- eg. a special interest demographic segmentation system for an automobile retailer would have a different set of variables, or at least assign
different weights to the common variables, than one created for a pharmacy chain retailer because the factors that are most significant in one
market may be less important in another
C. Market demographics
- Market demographics adds market survey data to census-based demographic data
- may be based on actual buyer behaviours obtained from credit card and affinity card data through data mining
- may include psychographic data that links attitudes and dispositions built on perceptions of what you want to be to actual consumer behaviour
rather than demographics
D. Geodemographics
- Geodemographics adds geography to these variables in recognition that people’s behaviour both influences and is influenced by where they live
- Not just ethnic and income-based clustering which is obvious, but people with similar attitudes to property, status and achievement, etc.
- neighbourhood is meaningful
Geodemographic Clusters
- cluster small areas (neighbourhoods) where residents have similar demographic and other characteristics
- all areas in a single cluster are as homogeneous as possible
- clusters are as dissimilar as possible from each other
- the smaller the areas, the more refined the segmentation
- neighbourhood clusters are not contiguous
- The smaller the neighbourhood, the more refined the market segmentation
- if areas as large as CTs are used, the clusters may not be homogeneous
- Toronto census tracts have an average population of 4 000 to 6 000
- census data is also compiled in much smaller areas called Dissemination Areas (DAs) which average about 650 people each, though 1 in 8 has a
population greater than 1000
- people in an area that has a population of 650 are much more likely to be similar than people living in areas whose population is 8-10X greater
- in the United States, the census publishes even smaller groupings of people by block, so their market segmentation can be at an even finer scale
than Canadian ones
Development of Geodemographics
Early 1900s
- census tract level data became available in the US (1910)
- 1920 urban sociologists at University of Chicago
- studies of US cities based on ecological principles
You're Reading a Preview

Unlock to view full version