Market segmentation - ideals - University of Illinois at Urbana ...
Market segmentation - ideals - University of Illinois at Urbana ...
Market segmentation - ideals - University of Illinois at Urbana ...
You also want an ePaper? Increase the reach of your titles
YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.
-7-<br />
segments. Thus the marketer will weigh the pr<strong>of</strong>it potential<br />
against the large fixed cost associ<strong>at</strong>ed with a mass marketing<br />
str<strong>at</strong>egy or the accumul<strong>at</strong>ion <strong>of</strong> multiple fixed costs associ<strong>at</strong>ed<br />
with many marketing mixes differing<br />
in media schedules.<br />
4. there is uo longer aiiy need (or relevance) to specify target<br />
markets versus non- tar gets.<br />
Traditional non-target groups<br />
essentially reflect a low pr<strong>of</strong>it potential and thus receive<br />
little weight in the determin<strong>at</strong>ion <strong>of</strong> the appropri<strong>at</strong>e marketing<br />
nixes.<br />
Segment Form<strong>at</strong>ion:<br />
A Disaggreg<strong>at</strong>ive Process<br />
Segment form<strong>at</strong>ion can easily be accomplished using cluster analysis<br />
or a similar procedure. It should be apparent from our previous discussion<br />
th<strong>at</strong> the number <strong>of</strong> segments formed need not cause concern; some<br />
segments will subsequently be aggreg<strong>at</strong>ed by procedures th<strong>at</strong> deal with<br />
the<br />
segment-marketing mix pr<strong>of</strong>it m<strong>at</strong>rix.<br />
It is important, however, th<strong>at</strong> segments be formed on the basis <strong>of</strong><br />
determinant variables .<br />
By determinant variables we mean those measures<br />
tn<strong>at</strong> lead to some way (analytically or subjectively) <strong>of</strong> deriving the<br />
response <strong>of</strong> the segment (in terms cf u and F) to the various r.arketing<br />
mixes. Thus variables th<strong>at</strong> include <strong>at</strong>titudinal measures, conjoint<br />
analysis measures, media exposure, shopping habits, consumption, etc.<br />
will be more determinant than demographic measures.<br />
The disaggreg<strong>at</strong>ion process should proceed as far as the l<strong>at</strong>a will<br />
allot., itcause large number <strong>of</strong> segments can result if the J<strong>at</strong>a s<strong>at</strong> is<br />
large (in order to achieve high intra-segraent homogeneity), the number