that customer/technology interaction in terms of usage frequency contains both rational and
emotional elements. As such, the result aligns with the conclusions of the Portuguese
neuroscientist Damasio (1995,1999) about the role of emotions and feelings in human
behavior.
Usually, decisions on high involvement situations are theorized to depend on
cognitive factors. In this paper, we consider the usage frequency decision concerning a bank
delivery channel as a high involvement decision, due to the risks and security concerns that
remote channels provoke and the assets involved - monetary funds of the customer. Results
illustrate that this decision also depends on affective reasons, which are usually only
researched for impulse buying or in low involvement buying/choice decisions
The analysis also shows that the specific affective and cognitive determinants of each
delivery channel usage depend on the channel. Concerning financial services, the core service
is basically the same between channels. The differences come from the interface considered
to provide the service. Additionally, the results show that technology-based channels differ
from the branch channel, in terms of cognitive vs. affective usage frequency determinants.
In terms of research limitations, this research is of an exploratory nature and of
limited scope. One of the most common problems lies in the extent to which the empirical
results obtained can be generalized to other samples and contexts. The analysis concentrated
in one firm from one service industry (a bank). One question arises about the generalization
of the conclusions to other banks or other service industries. An interesting avenue for future
research would be to replicate the model with other banks and with other service industries
characterized by multiple delivery channels (technology-based and human-based). This
extension or replication could enlighten the generalization of the findings.
Further, the cognitive and affective variables presented in the model are, by model
design, not influencing each other. It is acknowledged that important interactions between
cognitive and/or affective variables might exist: however, the intent was to explore channel
usage incorporating the variables that represent the most critical influences. These efforts
comprised an initial qualitative stage, with in-depth interviews with both bank managers and
customers, and extensive literature research. The model presented could be explored further,
by integrating other variables as predictors of usage. For example, ‘familiarity with
technology’ or ‘familiarity with the channel’ could be explored as an additional influence on
usage. The model could also be extended to include relationships and interactions between
the variables. Additionally, there still might be confusion between the need for a banking
27