270 The Marketing Book
Figure 10.4 GIS data fusion
(a)
targeted with different treatments, according to
whatever gender, age, marital status or geode-
mographic characteristic might underpin the
‘creative’ (Figure 10.6).
This sort of software incorporates data
mining techniques, as shown, but it is worth
explaining these a little further. Data mining, in
principle, is a ‘process of extracting hidden or
previously unknown, comprehensible and
actionable information from large databases’
(Antoniou, 1997). From this there are two
approaches that data mining can adopt. The
first is verification driven: ‘extracting information
to validate a hypothesis postulated by a user’.
Many organizations will have their own seg-
mentation model. For example, supermarkets,
in analysing their loyalty card data, might use a
model that incorporates: Stage in family life cycle (this is a very popular
segmenting base for data-driven marketers).
Vegetarianism.
Interest in alcohol.
Baby products.
Orientation to predominantly own labels or
manufacturer brands, etc.