A ®rm's ability to compete in new product markets is vital to its pro®t-ability and long-term survival. Therefore, it is important to understand the development and growth of these markets. Following a pioneering study by Bass (1969), diusion models have traditionally provided this understanding in marketing. The great appeal of the Bass model is that it is a simple one that ®ts the data very well and provides parameters that have an intuitive behavioural interpretation. The model suers from three wellknown limitations: (1) it does not include marketing variables that could in¯uence new product diusion and sales; (2) the model's parameters are unstable; and (3) the model's forecasts are inaccurate before the sales peak and especially prior to the point of in¯ection. Subsequent research has made progress especially in extending the Bass model to include marketing variables. However, the extensions have come at the cost of simplicity: the new models are far more complex than the simple Bass model. We propose an alternate simple model of new product growth for consumer durables, based on the concept of aordability rather than on diusion. We compare this model with the diusion model in terms of ®t, stability and validity of parameters, and forecasting ability. The alternate model is a little inferior to the diusion model in ®t, but superior in terms of the stability and validity of parameters and forecasting ability. We discuss the limitations and implications of our model. #