Geography Reference
In-Depth Information
Box 42.1
Finding customers for a large cinema chain
Imagine you are a marketing analyst for a large cinema
chain. From various surveys of your customers, you are
able to build up profiles of their individual and/or
household characteristics. Most new cinemas are now
built out of town, with large car-parking facilities, multi-
screen complexes and in association with many other
leisure facilities such as ten-pin bowling and fast food
restaurants. Who do you think is most likely to visit such
cinemas? Using ACORN (see Table 42.2), Thompson
(1997) identified the following target groups:
•
wealthy achievers
•
affluent urbanites
•
prosperous professionals
•
better-off executives
•
comfortable middle ages
•
skilled workers.
Thez next question is where are these groups
located? The purchase of ACORN would allow the
analyst to search through the database and highlight
all postal districts in the UK that have one of the
above customer groups associated with it. In Leeds,
this would produce the following list: LS6, LS7, LS8,
LS15, LS16, LS17, LS18, LS19, LS20, LS21, LS22,
LS23, LS29.
Thus the analyst could now either (1) target these
areas of Leeds through localised advertising (say local
newspapers) or through direct mail shots to addresses in
these areas; (2) if the total numbers were large enough,
maybe encourage the company to locate somewhere
within such neighbourhoods.
outlet sales. The most common method is the so-
called 'fair-share' approach (Beaumont 1991b).
Hence, if there are three other competing stores
in the buffered catchment area of the new store,
then the new store may be expected to obtain 25
per cent of the revenue generated in that
catchment area. This simple fair-share allocation
could be weighted by store size or by retail brand
to increase realism. The alternative is to assume
that the consumer will simply travel to the
nearest store within the catchment area
(dominant
store analysis:
see Ireland 1994).
Many retailers now use geodemographics and
GIS for marketing and site appraisal. In the UK,
these include most of the major grocery retailers
as well as large international groups such as Marks
& Spencer and Kingfisher. A good example of a
linked GIS/geodemographic system in the UK is
CACI's 'Insite System', which has been targeted
specifically at retail businesses wishing to match
catchment area profiles (based on census and
geodemographics) with those obtained from their
customer databases. CACI is working with a
number of UK high street retailers, including
Norweb, Britdoc, Budgens, Woolworth and
Yorkshire Building Society (CACI 1993). Such
bespoke geodemographic systems are also
increasingly available within general GIS packages
for other regions of Europe (Hinton and Wheeler
1992; Reynolds 1993). Indeed, 'Experian'
(formerly CCN) has recently launched a pan-
European version of its popular MOSAIC system
(Webber 1993; Birkin 1995).
Although the GIS/geodemographics approach
is popular with retailers, there are two principal
drawbacks in relation to estimating store
turnovers. First, there is the problem of how to
define the catchment area and, second, how to
adequately treat the competition. As we saw above,
the former is normally represented by distance or
drive time bands, and it is often assumed that the
store will capture trade uniformly in all directions.
Even when drive time bands are drawn in relation
to transport networks (Reynolds 1991), there is
still the assumption of equal drawing power in all
directions. Second, the treatment of the
competition is wholly inadequate. As suggested
above, the method most often used is 'fair share',
with the potential revenue of the catchment area
simply being divided between all retailers on some
ad hoc
basis (type of retailer, amount of floorspace,
etc.). This does not allow for the complex set of
real interactions between residential areas and
retail locations, which are distorted in the real
world by intervening opportunities. As Elliot
(1991) acknowledges, the presence of competing
centres will restrict the catchment boundary of a
new store in some directions. Her response is to
'override the drive time where it seems
appropriate' (p. 171). Such subjectivity is the
