1. See, for example, G.S. Lynn and R.R. Reilly, “Blockbusters: The Five Keys to Developing Great New Products” (New York: HarperBusiness, 2002); E.E. Bobrow and D.W. Shafer, “Pioneering New Products: A Market Survival Guide” (New York: Irwin, 1987); and R.M. McMath and T. Forbes, “What Were They Thinking?” (New York: Crown Business, 1998).
2. See R.G. Cooper, “How New Product Strategies Impact On Performance,” Journal of Product Innovation Management 1, no. 1 (January 1984): 5–18.
3. See N.P. Trepanning, “Understanding Fire Fighting in New Product Development,” Journal of Product Innovation Management 18, no. 5 (September 2001): 285–300. See also C. Billington, H.L. Lee and C.S. Tang, “Successful Strategies For Product Rollovers,” Sloan Management Review 39, no. 3 (spring 1998): 23–30.
4. M.L Fisher, J.H. Hammond, W.H. Obermeyer and A. Raman, “Making Supply Meet Demand in an Uncertain World,” Harvard Business Review 72, no. 3 (May–June 1994): 83–93.
5. The Cisco Systems transition example is based on a 2001 white paper, “Strategizing for Success: Cisco Systems Overcomes a Product Transition Dilemma,” ZDNet UK, London, February 20, 2001, http:// whitepapers.zdnet.co.uk/0,39025945,60045032p-39000468q,00.htm.
6. Billington, Lee and Tang corroborate this finding and present a high-level process for managing new product transitions. They recommend dual-product rollovers (that is, introducing the new product before the end of life of the old one) for transitions with high demand and supply risks and solo-product rolls (the new product introduction concurring with the old product’s end of life) for low demand and supply risk environments. Oftentimes, however, the industry dictates the choice of solo versus dual roll. Dual-product roll is standard in the high-tech industry where product platforms are common, even for products with low demand and supply risks. Further, the process proposed by Billington, Lee and Tang does not provide much insight into tactical and operational decisions regarding pricing, capability, marketing budgets or product deployment, all of which can have a substantial impact in the success of a transition.
7. We tested the transition mapping process, particularly the factor analysis process, using a large-scale product transition at Intel. For this transition, Intel’s central business planning group felt that sales of the new product would come in fairly strong. Defining x as the realistic “whisper” estimate among forecasters, a figure of roughly 1.2x was circulated to drive supply. Meanwhile, estimates aggregated from the geographical sales organizations suggested lower sales, ranging over time from 0.65x to 0.9x. Based on the results of the factor analysis and historical sales in the same product family, the transition mapping team predicted that sales were unlikely to exceed 0.93x and would likely be lower. The drivers for this recommendation included solid evidence that component cost would reduce demand early in the transition and that the complexity of the new platform posed significant supply risk. Sales forecasts were revised downward from 1.2x prior to the launch to about 0.9x six weeks after launch and then dropped even lower. By the beginning of the second quarter after launch, the forecast, informed by the transition mapping process, was trimmed to 0.79x for the first two quarters’ total sales. This helped avoid overbuilding supply for the new product while maintaining sufficient stocks of the old product. The process also supported decisions, such as increasing the marketing budget, that helped drive product sales early in the life cycle.
8. For example, refer to H.L. Lee and C. Billington, “Managing Supply Chain Inventory: Pitfalls and Opportunities,” Sloan Management Review 33, no. 3 (spring 1992): 65–73; or G.A. Zsidisin, A. Panelli and R. Upton, “Purchasing Organization Involvement in Risk Assessments, Contingency Plans, and Risk Management: An Exploratory Study,” Supply Chain Management 5, no. 4 (2000): 187–198.