Faculty Guest Post: Reasons for New Product Failure: A Behavioral Perspective

Reducing the persistently high failure rates of new product introductions remains one of the greatest challenges of innovation literature. To make progress on this front, a number of scholars have sought to identify and organize factors that contribute to new product success or failure in the marketplace. Although these studies have greatly expanded our understanding of new product performance antecedents, they share a notable limitation in that they follow closely the content and structure of the seminal SAPPHO and NewProd studies that pioneered this stream of research some 40 years ago.

As a result, only a relatively constant subset of factors has been explored, albeit quite extensively. On the downside, existing new product performance frameworks have not considered entire classes of potentially important factors, most notably those that pertain to managers’ incentive structures and cognitive biases. For example, recent research in economics, finance, and management shows that manager limitations and personal characteristics, such as overconfidence, narcissism, and cognitive style may affect firm strategies and performance.

I explore (in collaboration with Will Tracy from Rensselaer Polytechnic Institute and three colleagues from New York University, HEC-Paris and IIT-Kanpur in India) whether decision makers’ overconfidence, or excessive belief in their own abilities to generate superior performance, is associated with flawed intermediate decision inputs in the new product development process. Our general rationale is that flawed intermediate inputs are likely associated with a higher likelihood of poor new product performance in the marketplace once the product is launched.

We study graduate business students making a range of decisions over four rounds of a software-based management simulation exercise. As might be expected from an optimistic bias, we find overconfidence to be associated with a higher likelihood of over-forecasting new product demand.

Over-forecasts are problematic on two levels. First, they often lead to heightened expectations for new product performance. When unrealistic expectations are not met, the firm may be more likely to brand a new product a failure and treat it accordingly, which may produce a self-fulfilling prophecy.

Second, over-forecasting directly leads to higher production volumes, with more capital tied up in product inventory. Cash constraints may put pressure on managers to reduce inventory by discounting products that do not sell well. This may lead to a new product’s subpar performance on profit benchmarks the firm sets for its new products.

Dmitri Markovitch and William Tracy are assistant professors at the Rensselaer Lally School of Management.