The North American auto industry squanders more than $10 billion each year because of poor planning and faulty coordination between carmakers and parts suppliers when products are developed and assembled, McKinsey research indicates. Yet the level of waste varies widely for players in both roles, suggesting a significant opportunity for improvement. If the entire industry were as efficient as its top performers, it could save $8 billion.

Indeed, we think this estimate is conservative. The study included only those costs borne by suppliers, such as additional engineering, last-minute modifications to production tools, and added costs for the parts themselves. It didn't take into account extra engineering by manufacturers, increases in the final cost of vehicles because of production issues, or the expense of warranty repairs, which often stem from design problems and cost the North American auto industry around $12 billion a year.

Our study evaluated the recent product-development efforts of eight parts manufacturers and six Japanese and US carmakers operating in North America¹ and found that 80 percent of the waste is concentrated in three areas. The largest one involved changes to product specifications during a new model's two- to three-year development process. We estimate that these modifications cost the industry almost $6 billion a year; that's some $360 for every car and light truck produced in North America in 2004. About two-thirds of this waste was the result of engineering changes made late in the product-development phase, after designs were frozen and manufacturing tools, such as dies and molds, had been produced.

The best automakers defined product specifications jointly with suppliers early in the development process. This collaboration resulted in significantly fewer changes and considerable savings, since in most cases the modifications were relatively minor. In the case of the worst automakers, ill-defined product specifications delayed agreement on a final design for parts and led to a higher number of last-minute changes, which are up to 20 times more expensive to make than early changes. Rush modifications also tended to inflate the final cost of the component because there wasn't enough time to optimize the design. Worse still, these alterations triggered additional negotiations between the carmaker and the supplier, eating up the valuable time of senior executives.

Product complexity—mostly the result of unnecessary variations in a component's design—is a second key source of waste and costs the auto industry about $1.5 billion a year. One manufacturer designed, developed, and tested more than 30 different versions of a key part for an instrument panel, for example; ultimately, it used only two-thirds of these. Meanwhile, a competitor needed fewer than ten alternatives, despite offering similar models in the same market segment. Not- withstanding this potential for waste, we found that few carmakers routinely discuss the economic trade-offs of additional complexity with their suppliers at the concept stage for new vehicles.

Overly optimistic estimates of production capacity for individual parts, plus wide swings in weekly production, added a further $1 billion in waste annually. The best programs made conservative estimates of production volumes and added capacity if demand for a part outstripped initial expectations. By comparison, poor programs assumed unrealistically high volumes, leaving suppliers' factories operating below capacity when orders were scaled back. To compound matters, once parts production began, these programs often experienced volatile demand that added costs for rush shipping and the like.

The wide discrepancies among companies in these areas show the potential for improvement. The best automakers wasted about 1 percent, on average, of the total cost of developing a new part; the worst squandered more than 10 percent. The gap was narrower among suppliers (exhibit). We also found large differences depending on the type of a part. Components that are part of a system or highly visible to consumers (such as those found in a vehicle's interior) generated additional costs of 10 percent, on average. At the other end of the spectrum, parts that were not integral to a larger system or were invisible to customers tended to generate significantly less waste.

For auto manufacturers and suppliers, the key to reducing waste is thorough planning early in the product-development process. During the initial phases of a program, the carmaker and its key suppliers should jointly set targets for the cost and performance of components. While this approach may seem self-evident, many North American automakers focus on the later phases of product development, when firefighting prowess is valued more than planning. Making fundamental changes to this approach—and replicating them across many product-development programs—will be major challenges. The potential savings make them well worth tackling.