Managing the Trade‐off between Ambiguity and Volatility in New Product Development

Abstract

Uncertainty is widely recognized to be a key contingency influencing optimal new product development processes. However, extant new product research treats uncertainty in a largely unidimensional manner. Even in studies where multiple dimensions of uncertainty are measured, they are hypothesized to behave in a similar manner. This paper advances a theory describing how ambiguity and volatility place different and conflicting demands on new product development processes. Drawing on ideas from the organization theory and learning literatures, the theory suggests that more ambiguous environments favor slower development processes based on larger samples of data and interpretations whereas more volatile environments favor faster and more flexible development processes. Five hypotheses are advanced suggesting that as the level of ambiguity increases (decreases) relative to the level of volatility, firms should respond by decreasing (increasing) the role of top management in new product development, increasing (decreasing) front‐end and back‐end participation, allocating more (less) time and effort to ideation, and using more (less) aggressive screening processes. Some of these hypotheses are highly intuitive (participation and screening) while others are counterintuitive (top management involvement and screening aggressiveness). The theory is tested using primary data on the development of 120 new products and their subsequent performance collected from firms in U.S. manufacturing industries. The data largely confirm the hypotheses and provide strong evidence of a trade‐off between ambiguity and volatility that can be managed to improve new product performance. The exception is back‐end participation, which should be lower in relatively more ambiguous environments, in contrast to the hypothesis. A comparison with three alternative models shows that ambiguity generally has a larger impact on appropriate new product development processes and success than either the level of volatility or the overall level of environmental uncertainty. A detailed examination of the estimated relationships shows that for ideation and screening, the relative level of ambiguity and volatility determines optimal new product development processes, whereas for top management involvement and participation, processes can be designed based primarily on the level of ambiguity alone, with less concern over the degree of volatility. The study highlights the importance of considering the composition of uncertainty in addition to its overall level in contingent approaches to new product development. Real‐world examples of highly ambiguous and highly volatile new product development contexts are discussed.

Perspective: A Review of Marketing Research on Product Design with Directions for Future Research

“…Other approaches highlight leveraging interdependencies between design and manufacturing phases in NPD (Bajaj, Kekre, and Srinivasan, 2004) or identifying conditions in which design activity coupled with strategy minimizes lead time but maximizes other performance measures (Joglekar, Yassine, Eppinger, and Whitney, 2001). Finally, top management involvement and participation can allow processes to successfully account for the level of ambiguity alone, with less concern for the degree of volatility (Carson, Wu, and Moore, 2012).…”

Analysis of the effect of risk management practices on the performance of new product development programs

“…Drawing on ideas from the organization theory and learning literatures, Carson et al (2012) proposed a theory describing how uncertainty due to ambiguity and volatility affect NPD process. They suggest that more ambiguous environments favor slow development processes wherea more volatile environments favor faster and more flexible PD processes.…”

Project Evaluation: A Mixed-Method Study into Dilemmas in Continuation Decisions

“…Conceivably, these indicators show the extent of equivocality in the project evaluations. Next, we adapted items from [25], [26], [27] and [28] to align our indicators with the extant studies.…”