Modeling generally implies “macro-modeling” so it is always important to specify when we are talking about “micro-modeling”. Most models in the past have been macro-models, and in marketing, SMTs were no exception. All SMT models created in the 1970s are “macro-models” for three major reasons:
a) In the 1970s macro-modeling almost completely dominated the world of modeling: economists work only with large aggregates that they combine using complex mathematical equations. Modeling in marketing naturally drew its inspiration from these models, which gave it the legitimacy that psychology at the time was still unable to give it.
b) Computer power at the time was extremely limited. In the 1990s, the “computer revolution” transformed many fields by hugely increasing the power and speed with which information could be processed, at a fraction of the cost. What was done slowly by the big expensive mainframe computers of the 1970s is now done in a flash by a small desktop PC that costs less than 500 Euros at a superstore. There is no other recent example of such a dramatic change in such a short time.
c) In the 1970s, only psychologists were interested in consumer behavior. These include the pioneer in the field, Arnold E. Amstutz, who in 1967 developed a micro-model that simulated the early theories of consumer behavior in the form of a mathematical mod. But this test was long confined to the field of experimental psychology.
One obvious question is why the “old” models like Bases or Designor did not tried to capitalize on this revolution to leverage all these virtually free new capabilities. The answer is that these models didn’t need to. They have basically made no changes for 40 years, although there have been some minor adjustments here and there so they could say they were keeping abreast of the latest trends in IT.
