Why do we spend so much time with straight line (linear) models?

In MGSC 1205, most of the course focussed on linear models describing relationships with straight lines. Are cost and revenue functions always linear? And now we are looking at building models from data, and once again we are assuming that relationships are linear. But look at finance models. Compound interest is the basis of investing. Compound interest is an exponential model. Many models of sales growth assume that initially growth is slow until consumer acceptance, then there is rapid growth until market saturation and then it slows again. Total sales over time follow an S shaped curve. If many relationships are curve, what are the reasons why we have such a strong preference for linear models? What are the drawbacks of assuming linearity?