Product life cycles continue to shrink in "fast" industries like fashion and personal computers, where new products roll out the door every few months. Even "slow" industries like steel have quickened their pace. Speed is of the essence—but so are quality and cost. Companies in all sectors know they must deliver on all three dimensions simultaneously.

Yet world-class product development is still by far the exception. Even the most disciplined and creative product development processes can be plagued by schedule slippages, "firefighting," and inferior quality. Often, the problem is a failure to consider product development as a dynamic whole. Managers try to improve performance by breaking the process apart.

But isolated improvement efforts (such as the best practice themes listed in Exhibit 1) can actually have a negative effect on performance when they are implemented without regard to the multiple interrelationships that drive the product development process. Setting aggressive commercialization schedules, for example, often has just the opposite effect of what was intended: namely, longer delays with lower quality. And many other cherished management beliefs, such as "lean and mean" staffing, can also inadvertently contribute to product development problems.

Consider three company experiences from different industries:

• An electronics company bolstered testing procedures and management reviews of new product programs, but experienced chronic design delays and high rates of product returns due to subpar quality.
• A major automotive company created crossfunctional teams and highly developed project milestones, but suffered massive schedule and cost overruns on an early trial of their new system.
• A biotechnology company used state-of-the-art project scheduling software to plan development and rollout of a sequenced set of disease assays that had to be integrated into an overall system for hospitals, but saw 50 to 100 percent commercialization delays with habitual resource switching across programs.

What’s needed instead is an integrated, holistic approach that captures product development as a complex system. It must recognize and account for the cascading sequence of causes and effects that determine how well a company delivers against the key performance metrics of speed, quality, and cost. Using a business dynamics model, we have developed such an approach.

Product development as a complex system

The model was constructed to simulate the various steps most development projects pass through (concept generation, design, prototype, testing, pilot operations, and manufacturing ramp-up), along with the various resources needed at each step (such as design and process engineers). It also reflects a fact of life in most industries: market requirements are always rising. Above all, it was designed to capture the feedback loops within the product development process that are not usually addressed by one-off best practice efforts.

Exhibit 2 illustrates some of the major interrelationships that trap development organizations into schedule slippages, "firefighting," and inferior quality. Suppose that a particular project begins to fall behind schedule—the "design schedule slippage" at the center of Exhibit 2. A host of ripple effects ensue.

Toward the right of the exhibit, for example, the design schedule slippage causes manufacturing ramp-up to be accelerated in order to keep close to delivery targets. But the lack of manufacturing readiness creates manufacturing quality problems that have multiple repercussions. Within manufacturing, on the right of the exhibit, unscheduled downtime dictates personnel rotation that slows the learning curve and leads to still more quality problems. In addition, manufacturing problems ripple back into design on the left of the exhibit when engineering resources are sucked downstream into problem solving, compounding the original design slippage.

These kinds of cascading impacts are frequently seen in real-life product development projects. One company calls them "train wrecks" because of their huge schedule and budget overruns. The implication is that such overruns are an aberration—yet they are more often inherent in the product development process itself. Such problems can easily be repeated if the process is not changed (Exhibit 3).

Huge—but elusive—potential benefits

Modeling the project portfolio showed that the potential benefits of improving the product development process were tremendous, with multidimensional gains from lower cost, shorter commercialization time, and better quality (Exhibit 4).

However, it emerged that only a small part of this potential could be captured by applying one or two of the best practice themes. In other words, individual strategies like improving test quality or upgrading design for manufacturability (DFM) had relatively low impact (Exhibit 5). Not only is there no "silver bullet," but uncoordinated functional strategies may unintentionally conflict: for example, boosting resources to upgrade DFM can actually result in poorer test quality, or in faster manufacturing ramp-up that dilutes the average skill level of line personnel and thereby cuts productivity.

The low leverage afforded by functional strategies is unfortunately confirmed by many companies’ experience: they might use the "theme of the month" to try to get product development process improvements, but they rarely see much benefit. One company upgraded its test procedures to root out previously overlooked errors, but resources were soon choked by mounting rework piles, leading to new problems. The message is simple: trying to fix product development through functional routes yields only incremental benefits that may not be sustainable over time. "Single-point" solutions don’t work. To reach world class, an integrated approach is needed.

Why fixes fail

Why do so many of these individual fixes add little value? How should strategies be coordinated to achieve bigger payoffs? To find the answers, it is necessary to analyze the full gamut of factors influencing the product development process. Close examination reveals, for example, that the common management strategy of setting aggressive commercialization schedules has the contrary effect: longer delays with poorer quality. In Exhibit 6, management demands that a project be completed 40 days early. For a while, the strategy seems to work: the project gets through concept development 15 days ahead of time. But design is completed only 10 days ahead, which means that some of the concept development gains have been lost. Eventually, the project is released to manufacturing 35 days later than originally scheduled—or fully 75 days behind the more aggressive completion date.

Why the reversal? A number of forces are exerting unintended secondary effects. Schedule pressures lead to skimping on upfront DFM and test design, thereby causing more rework, poor design stability, and still more intense schedule pressure. The opposite strategy, "going slow to go fast," would actually accelerate product commercialization and raise both design and manufacturing quality.

In tandem, a strategy of adding more development resources and committing them early in the process can in fact yield lower costs with improved quality (Exhibit 7). This approach works, first, because overall design costs increase only marginally—by one-third of the direct resource addition—owing to a reduction in rework and testing (for example, design recycles shrink by 25 percent). Second, higher manufacturing productivity and lower warranty costs enable a net reduction in total life cycle costs. Adding strategic resources in a high-leverage part of the product development process thus saves money, while dramatically speeding time to market.

Another important factor is the use of "phase gates" as checks of product quality—a growing practice in many industries. A quality gate is a "hard" screening and rework point beyond which products cannot pass until they are upgraded to at least the gating quality level. Analysis reveals that a quality gate upstream in concept development can actually help boost profits by over 12 percent (Exhibit 8). However, downstream quality gates add progressively less value and can eventually destroy value. In the exhibit, a quality gate positioned behind the prototype stage or in integrity testing is worse than no quality gate at all, reducing total profits by 4 to 5 percent. Why?

First, if the quality gate is too late in the process, earlier stages are still worked ineffectively. This means that when the downstream quality gate is eventually triggered, considerable resources are inevitably required to bring product quality back up to standard. Commercialization slows as a result. The project itself may be helped, despite running late, but other projects will suffer because resources have had to be pulled away from them.

In addition, with several projects now running late in design, compounded schedule pressures yield adverse tradeoffs—such as an accelerated manufacturing ramp-up that negates design improvements. So design quality goes up, but manufacturing readiness and quality deteriorate.

What’s the bottom line? If problems are not detected and corrected at the earliest upstream gates, downstream quality gates end up pulling resources from one project to another, which leads to multiple schedule delinquencies, poorer design stability, and quality shortcuts prompted by intensified schedule pressures. It seems likely that companies utilizing multiple quality gates across design and manufacturing may be unintentionally reducing their profits and quality. The challenge for management is to design quality gates that are effective, efficient, and positioned as early as possible in the development process.

The biggest benefits are achieved with a broad, integrated improvement strategy (Exhibit 9). Such a strategy is multi-pronged: functional improvements include better testing and DFM; net design resources are added early in the process; firm but realistic schedules are adhered to; quality gates are instituted in both concept development and manufacturing ramp-up; and projects that fall behind schedule are more likely to be "killed off" than permitted to drag down resourcing on other projects. This breadth yields synergies with almost 50 percent of the impact of the sum of the individual strategies, and has the potential to double total company profitability.

Activating and sustaining virtuous performance cycles can create world-class product development

Product development synergies emerge from mutually reinforcing strategies (Exhibit 10). Increasing the engineering resources committed upstream, for example, accelerates design and testing and reduces the need for mid-course corrections to catch up with competitive technology. Early quality gates and test upgrades amplify the benefits: more stable design enables better project feedback, still more effective testing, and a reduction in manufacturing downtime due to quality "patches." Accelerated time-to-market itself leads to premium pricing and higher market share. Activating and sustaining these virtuous performance cycles can create world-class product development organizations that satisfy customers and surpass competition.

To capture these benefits, management needs both to put in place and to follow through an integrated plan like that illustrated in Exhibit 9. Without a clear management vision of the full scope and span of required improvements, satisfaction with "early wins" could lead to easing off so that the big, long-run synergies are never captured.

An integrated approach

Launching an integrated product development strategy involves three main elements:

• High-level sponsorship and "championing"
• An incisive view of the leverage points to attack
• Sustained implementation planning and performance tracking.

The product development champion may be the CEO or a business unit head. In most cases, however, the CEO must be a visible sponsor because project and portfolio evaluation and resourcing and expenditure decisions will invariably be reviewed by the CEO. The product development champion and the CEO would typically convene a product development board including key functional heads (such as senior managers in development, manufacturing, and marketing) who will be charged with implementation, arriving at solutions, and initiating improvement pilots.

Developing an incisive view of leverage points starts with a thorough diagnosis of product development performance. This diagnosis should rigorously dissect past performance and root causes. How have major projects performed in terms of cost, quality, and time? At what stage(s) are problems usually found? How stable are designs over time? Are manufacturing learning curves steep or shallow?

Next, interviews and workshops can be used to get input from both senior managers and those actually doing development work to build an understanding of key issues and needs, day-to-day work pressures, interactions between activities, potential "vicious cycles" limiting performance, and other causal relationships. These relationships can be captured first in a cause-effect diagram like that in Exhibit 2, and then in a quantitative business dynamics model that represents the "moving parts" of the overall process. The model must be tested for accuracy—for example, by tracking actual projects, as in Exhibit 3.

Running a range of alternatives will help catalyze thinking about causes, improvement potential, and possible leverage points

Once the management group gains confidence in the model’s portrayal of the actual product development process, running a range of alternative scenarios will help catalyze new thinking about the causes of problems, the potential for improvement, and possible leverage points. The process is highly interactive, and must involve product development managers and senior executives in challenging their beliefs and, eventually, in building consensus on what to do. If new products have chronically overrun their budgets and schedules or been prone to manufacturing quality problems, for example, managers need to understand why—and not on a project-by-project, "story-telling" basis, but at a fundamental process level: what aspects of the process are broken, how many cures are needed, and where do we start?

Senior executives should also ask themselves a critical question: "What are the levers by which I can influence product development success or failure?" Model scenarios can then be crafted to shed light on these influence points. If unrealistic schedules are part of the problem, for example, senior managers should come to understand how they contribute to these schedules through project reviews, incentive schemes, and so on.

More positively, senior managers should participate in shaping integrated strategies from their understanding of individual and combined leverage points. One company found that setting up crossfunctional teams was helpful, but insufficient; as one manager put it, "It simply changed the venue of our arguments."

Senior managers need to be involved in designing the total strategy package, not piecemeal approaches

Senior managers need to be involved in designing the total strategy package, not piecemeal approaches; who else is responsible for the whole process? What new skills are needed to manage effective quality gates upstream in design, say, or to facilitate earlier and more informed "kill" decisions before failed projects have siphoned off vital resources? How many projects can we realistically undertake? One company learned that when it added more development resources, it inadvertently shot itself in the foot by starting up more projects at the same time, so that the development organization was still spread extremely thin. Creative use of model scenarios can be a powerful forcing function in enabling these discussions.

Senior managers also need to recognize that leverage points will shift over time as improvements are achieved. Faster design completion, for example, will expose new bottlenecks in, say, a company’s ability to ramp up manufacturing quickly, or in the skills of manufacturing personnel. Identifying performance bottlenecks and the leverage points that allow them to be removed thus needs to be a "living" management process, not a one-time event. Proactively removing bottlenecks one after another helps create the conditions for ever-higher levels of performance.

Implementation planning and performance tracking starts with the selection of pilot projects and involves establishing clear process performance metrics that will act as leading indicators of change, as well as end-of-process objectives. The pilot should build on a prioritized sequence of initiatives, not follow the prevalent "rifle shot" approach. This year’s improvement targets and focus areas should be crystal clear to the organization and lead to next year’s targets, and so on. The CEO and product development champion should be heavily involved in tracking progress, rewarding success, and encouraging and disseminating emerging best practices.

Key implications

A dynamic view of product development sheds light on the value of the different levers available to management. In isolation, even well-executed functional strategies may have low payoffs, for example. Management actions may also unintentionally conflict—as when an increased focus on design for manufacturability squeezes out test quality—or turn out to be detrimental to speed, quality, or cost.

All the same, large, sustainable gains can be achieved through an integrated and tailored approach. World-class product development performance can be created by far-sighted managers; this holds out great promise for the vast majority of companies that were not born that way.