Business-to-business marketing: organizational buying behaviour, relationships and networks 153
Complex models
In the late 1960s, several more comprehensive
models were postulated, incorporating a large
number of variables. For example, the decision
process model (Wind and Robinson, 1968)
depicts decisions as occurring over a consider-
able period involving segmented stages, i.e.
‘problem alternatives’ and ‘selection’.
Thecompact model(Robinsonet al., 1967)
attempts to establish general rules that govern
the decision process. Three dimensions of
influences are proposed: organization struc-
ture; elements of the buying process; and the
characteristics of individuals. However, in this
model no reference is made to the exchanges
and negotiation occurring between parties.
We can criticize these models for adopting
only a partial approach, and even the more
complex models fail to cover many of the
important points observed by Webster and
Wind. Moreover, many are theoretical rather
than realistic. They are of limited practical use
to the marketing or purchasing practitioner.
In an attempt to be more realistic, Robin-
sonet al. (1967) proposed the buygrid model.This
is based on empirical observations of buyer
behaviour in companies. In Table 7.3, we see
that decisions are thought to vary on two
dimensions. First, the stage of the decision (or
‘buyphase’) and, second, the nature of the
decision itself (the ‘buyclass’).
The buygrid model offers certain improve-
ments over earlier models. For example, it
recognizes that buying is often repetitive and
has a more normative basis for practising
managers. However, Ferguson (1979) found
that the buygrid model had only a limited
capacity to predict outcomes when he applied it
to examples of decision making. In most situa-
tions he observed that the proposed systematic
decision process was often ‘short-circuited’. To
overcome these objectives, we must turn to the
more comprehensive models which incorporate
a wider range of variables.Complex multidisciplinary models
Clearly, the models described previously are
inadequate both as descriptive representations
of reality and as predictive tools. Several
models have been developed which attempt toTable 7.3 The buygrid model
Buyphases BuyclassesNew task Modified re-buy Straight re-buyIdentification of need X X X
Determination of requirement X n/a n/a
Specific description of requirement X n/a n/a
Search for potential sources X n/a n/a
Examination of sources X n/a n/a
Selection of source X X X
Order routine established X X n/a
Evaluation of performance feedback X X Xn/a = not applicable
Source: Robinson et al.(1967).