Getting a Grip on the Saddle: Chasms or Cycles?

Abstract

The "saddle" is a sudden, sustained, and deep drop in sales of a new product, after a period of rapid growth following takeoff, followed by a gradual recovery to the former peak. The authors test for the generalizability of the saddle across products and countries and for three rival explanations: chasms in adopter segments, business cycles, and technological cycles. They model both boundary points of the saddle-start of the sales drop and recovery to the initial peak-using split-population models. Empirical analysis of historical sales data from ten products across 19 countries shows that the saddle is fairly pervasive. The onset of the saddle occurs in 148 product-country combinations. On average, the saddle occurs nine years after takeoff, at a mean penetration of 30%, and it lasts for eight years with a 29% drop in sales at its depth. The results support explanations of chasms and technological cycles for information/entertainment products and business cycles and technological cycles for kitchen/laundry products. The authors conclude with a discussion of the findings, contributions, and implications.

The Impact of Entrepreneurial Orientation, Technological Capability, and Consumer Attitude on Firm Performance: A Multi‐Theory Perspective

“…These markets are substantially different from one another, which necessitates different sets of product market strategies (Goldenberg, Libai, and Muller ; Mahajan and Muller ). Further, sales in the main market often increase concurrently with a sales decline among the early adopters in the early market (Chandrasekaran and Tellis ). The immediate strategic implication of this phenomenon for the firms in the early market is the following: either they should smoothly transition to the main market from the early market, or they should exit the early market when it enters the decline phase.…”

“…Finally, the nature of dual markets for new products is such that when the early market starts to decline, the main market concurrently strengthens (cf. Chandrasekaran and Tellis ; Goldenberg, Libai, and Muller ). Logically, entrepreneurially oriented firms are more likely to be in the early market.…”

The challenges of pre-launch forecasting of adoption time series for new durable products

“…Even after the incubation period many other factors are likely to affect adoption rates and may lead to departures from an Sshaped curve. Studies by Goldenberg, Libai, and Muller (2002) and Chandrasekaran and Tellis (2011) found that the adoption levels of many products experienced a significant drop after a period of rapid growth followed by a recovery to the former peak. Chandrasekaran and Tellis (2011) suggested that these 'saddles' or 'chasms', which tend to last for several years, appeared to result from discontinuities in the transition between early and late markets (i.e.…”

“…Studies by Goldenberg, Libai, and Muller (2002) and Chandrasekaran and Tellis (2011) found that the adoption levels of many products experienced a significant drop after a period of rapid growth followed by a recovery to the former peak. Chandrasekaran and Tellis (2011) suggested that these 'saddles' or 'chasms', which tend to last for several years, appeared to result from discontinuities in the transition between early and late markets (i.e. from customer heterogeneity), from economic slumps and from consumers' reluctance to purchase because they expected technologically superior products to be available in the near future.…”

“…Nevertheless, for the periods that follow the early stages of a product's life these types of formal models are more likely than the previous two approaches to be able to represent the complex, non-linear time-series patterns associated with diffusion (Chandrasekaran & Tellis, 2011;Goldenberg, et al, 2002;Kohli, et al, 1999) . In the case of the different forms of the Bass model, they are also based on a firm theoretical rationale.…”