Medical Device Link.

The discussion is organized around the two primary phases of a product development project's lifetime—feasibility and commercial development—and the critical concept that marks their division (see Figure 1). Companies have developed many models for tracking a project from inception to product release. ISO 9000– and FDA-mandated design controls have tended to introduce commonality to these systems. All of the models feature a decision point at which executives closely examine the project and determine whether it will support the investment required to bring the product to market. This decision is usually formalized when every company group involved signs off on a document variously called a design review, a product development plan, an approval for development, or something similar. Prior to this point, the project is in the feasibility phase, and its future is uncertain. Following the go decision, which might be termed the design freeze, all groups are committed to the project and it enters commercial development.

Project Leadership

Leadership is the foremost element of successful project execution. Medtech executives need to find and cultivate project managers who know how to move new products through the organization and who have an intimate understanding of prospective customers' needs. A great misconception to avoid is that a single person must manage a project through both the feasibility and commercial development phases. The skills required for project management vary dramatically from stage to stage of the project, and few individuals manifest them all (see sidebar above).

Feasibility is a time of exploration and discovery. It is also when unexpected problems, competitive challenges, and breakthroughs change the project landscape. Project leadership requires a tolerance for stops, starts, and changes in direction. Timelines will change frequently. In this phase, successful leaders are those with an intuitive ability to focus on what is important. Product development managers need to maintain a fix on the project goals and navigate a course through unpredictable new technology, volatile markets, and setbacks to progress, and reach the destination charted by upper management.

Commercial development requires a different type of leadership. Most of the project's technical uncertainties should be resolved and the path to commercial introduction clear. The project leader must keep the team focused on that single path, coordinating the various parts of the organization under pressure of an unforgiving timeline. Investment in the project now increases dramatically. Perhaps the most difficult thing for project managers to do during this phase is to shelve ideas that might ultimately make a better product but that would certainly derail the project from its current, carefully determined course. What is expected of the project leader is that the project be brought through commercial development on budget and on schedule.

It is a rare individual who can manage both the feasibility and commercial development phases of a complex project. Change in project leadership should at least be considered at the time of the design freeze, and medical technology executives should reinforce a company culture that views this transition as natural. Also, executives should reward both feasibility and commercial development leaders equitably. Some company senior managers appear to favor one set of skills over the other. But management failure to recognize the value of both types of leadership will likely drive off those who see themselves as less respected, resulting in an organizational inability to execute effectively and, ultimately, to failure of the program.

A difficult fact is that any new project represents risk and has only a limited probability of success. Project leaders at all stages need to be able to recognize when their options have run out and it is time to terminate the project so that the company can invest its resources elsewhere. In turn, one of the biggest challenges for senior management is to see that the project leaders deserve rewards for making this difficult but necessary decision.

New Product Design

A typical new product development project for a nonpharmaceutical medtech company will span three to seven years from inception to product introduction, and pharmaceutical projects may take much longer. This time frame is an eternity when regarded in light of the speed of change in both technology and the marketplace.

Project managers should develop a clear understanding of product requirements to serve as a guide throughout the life of the project. The requirements should be expected to evolve during the feasibility phase of the project, as understanding of the market changes, but they should continue to serve as the foundation for the new product program.

Requirements must be fixed at design freeze and remain constant during commercial development. Think of requirements as emerging and evolving during feasibility, and specifications as governing commercial development. This speaks to the importance of not beginning commercial development until the product requirements are clearly understood. The following steps can serve as a roadmap to successful development and use of product requirements.

Project Initiation. The first stage of feasibility is usually to scout the project. The scout group supplies the project team with insight into customer needs and an understanding of whether their project will be focused toward development of a totally new product or improvement of an existing product.

Project planners then should develop a comprehensive list of product attributes—characteristics such as performance, convenience, possible adverse effects, manufacturability, price, development costs, and time to market—and assign a minimum acceptable requirement and highly desirable goal to each attribute (see Table I). For this approach to be effective, setting minimum acceptable requirements has to mean that proj-ect leaders are willing to delay the program until all of these requirements are achieved—or to terminate the project if even one proves to be beyond reach. It should be expected that many of the desired goals will prove difficult to attain. Executives will need to decide how much time and investment to allow for their pursuit. This first step of project definition should be sufficient to guide initial R&D.

Market Research. The second step during feasibility is to test the product attributes and requirements against the needs of real customers in the market. This can be performed internally or contracted out to a market research specialist. However done, it is absolutely critical that the study be performed without bias and that it cover all segments of the target market. Results of the study should include a prioritization of the product attributes and a validation of the minimum acceptable and highly desirable benchmarks for each targeted market segment. This prioritization should provide project leaders with a basis for deciding on the level of investment, giving weight to product attribute measures deemed highly desirable. It may also guide company marketers in prioritizing the candidate market segments.

Two common pitfalls can hamper the advancement of this stage of the project. First, senior management may balk at the cost of the market research. Beware: penny wise is pound foolish. It is not unusual for a modest new medtech product effort to cost from $5 million to $10 million, sometimes even more. Investing 1% of the total project cost, $50,000–$100,000, in an effort to ensure that the project direction reflects the market need, ought to be considered prudent rather than extravagant. Too often, however, the latter judgment prevails and the company proceeds without this critical guidance. Perhaps the reason is that market research must be paid for in cash, whereas much of the rest of the product development project cost is hidden, being opportunity costs or an application of already existing resources (employees, laboratories, etc.).

A second management error is to confuse survey results for different market segments. Perhaps the market research determines that one group of customers will pay $50,000 for a product offering a moderate performance level and is not interested in paying more, while a second group of customers is willing to pay $150,000 for a high-performance product. Somehow this translates into management believing that the overall market will pay an average of $100,000 for a product with only moderate performance. A disastrous case of wishful thinking!

Alpha Testing. The final step in the feasibility phase is usually alpha testing. In an alpha test, the research prototype is taken to selected customers in each market segment so that they can try the product.

Alpha tests typically mix experimental protocols developed by the manufacturer and applications devised by the customer. The test is designed to determine whether the product prototype meets the proposed requirements. It also serves to check whether the stated requirements are what the customer really wants. It's not unusual to hear an alpha evaluator say, "It seemed like what I wanted, but when I actually tried it I realized that it wasn't quite what I need." At the conclusion of the alpha test, the sponsor should know whether or not customers find the product concept attractive.

Medical technology executives must be poised at this stage to make changes in both the product and the requirements for it. This is their last chance to be flexible before the requirements become frozen into specifications for the commercial development phase of the project.

Project managers also commonly make mistakes in the alpha testing stage of the project. One is to set up the alpha trial in an environment so artificial that potential real-world failures will not be detected. For example, if its actual customers will be expected to provide some of their own reagents, a diagnostic test manufacturer risks missing a potential failure mode during the trial by providing those reagents for them.

Another error is to succumb to the pressure to move the project along despite obvious problems. It is a true test of managerial courage to report a program delay when the organization is eagerly looking forward to the new product introduction. Executives of successful organizations allow for such setbacks during feasibility in order to prevent more-costly delays in commercial development—not to mention the even more undesirable consequences of a new product failing in the marketplace.

Design Freeze. The fulcrum of the shift from feasibility to commercial development, design freeze is a serious commitment. At this stage, company executives finalize the previously evolving product requirements and convert them into manufacturing and performance specifications. These specifications will be the basis for product release and ongoing quality assurance testing. ISO 9000 and FDA design control requirements require the manufacturer to conduct a project review, to maintain files describing the determination and validation of customer requirements, and to sign off on a product development plan that contains the specifications and agenda for the commercial development phase.

Beta Testing. Beta testing is conducted near the end of the commercial development phase and provides the final validation that the new product meets specifications and is ready for release. It is critical that beta tests be conducted replicating all of the conditions of manufacturing and use that would be characteristic of the commercial product. In addition, beta testing of medical devices may be part of the process of gathering data required for FDA or other regulatory submissions. Beta test protocols are generally more rigid than those used for alpha testing. The objective is to determine whether or not the product meets customer needs as defined by the specifications set at design freeze.

The last opportunity for the project leaders to make changes in the product prior to release comes with beta testing. There is tremendous pressure to move forward with product release at this stage. Unfortunately, rationalizing failed evaluations is the norm. Responsible executives have to display extreme courage to delay release to fix problems uncovered during beta testing.

Forecasting and Prioritization

Forecasting the success of new product introductions and prioritizing product development projects are endeavors that always seem to roil medical technology companies. Top executives believe that the project's champion ought to be able to give them an estimate of expected return prior to the commitment of thousands of person-hours and millions of dollars. Most companies see more opportunities than they have resources to develop. Their need to select and prioritize new product investments is critical. The manufacturing operation needs a unit forecast in order to determine scale-up requirements and begin production planning. Other departments have to know when to expect product release so they can incorporate that event into their plans.

Unfortunately, meaningful forecasts, especially for the most innovative projects, are extremely difficult to generate. Marketers find it virtually impossible to accurately predict the commercial performance of a product that will only be released in three to five years, and into a rapidly changing market. The best approach to coping with this paradox is to begin with broad assumptions that will be refined as the project proceeds. In most cases, a true forecast will not be possible until a project enters commercial development. However, a product development manager can provide company executives with the information they need to manage the project from inception through feasibility.

Project Selection. How do company leaders decide which proposed new product development projects deserve resources? Medical technology companies typically seek to support a portfolio of new projects that span a range of risk versus reward. Much has been written regarding project portfolio management. The following summarizes one model that has been suggested.¹

Executives considering candidate medtech projects can score them on the basis of their likelihood of being successfully developed and their commercial value if successful. Estimates based on predefined criteria and the judgment of the evaluators need only be given as high or low. Criteria with some specificity must be used in order for the qualitative rankings to be meaningful. For example, company managers might see high value in a project exceeding a three-year revenue projection of $10 million, and high risk in any product introduction with a probability of success under 75%.

Once scored in this simple fashion, projects can be assigned to a position within a project portfolio matrix (see Figure 2). Proposed projects with a low potential value and low likelihood of success are dead ends to be immediately rejected. Projects with both high potential value and high likelihood of success are rarities that should be placed high on the agenda and pursued aggressively.

Figure 2. The project portfolio matrix. Incentives and rewards for project teams should reflect the categorical difficulty and potential market value of each project.

Most of the portfolio will be made up of bread-and-butter projects, relatively low in potential value but offering high likelihood of success, or projects that have a high potential value but low likelihood of success. Bread-and-butter projects are commonly product upgrades, whereas their more-challenging counterparts tend to be innovative product concepts.

Senior medical technology executives should consult the company's strategic plan to determine the suitable allocation of resources between these two classes of projects. Winning strategies are often based on a series of successful bread-and-butter projects supplemented by a small number of risky but potentially very rewarding challenges to provide high-end value to the portfolio. Executives charged with responsibility for prioritization should identify and staff the most promising projects in each category. At this point, a scoring system more rigorous than that used to place projects in the portfolio matrix might be employed.

Financial Models. The financial models most commonly used to support new product forecasting are based on discounted cash flow (DCF). All DCF models reflect the time value of money. The chief model is net present value (NPV). NPV measures the potential value of a project against the return on a hypothetical safe investment over time. The NPV of a project can be calculated easily with commercial spreadsheet programs.

Again, executives must beware two pitfalls in making this calculation. First, NPV can be skewed by the terminal value of a project—that is, the amount that business related to a new product is estimated to be worth at the end of the analysis horizon, typically about 10 years for a medtech product. Terminal value can be quite large and thus can strongly influence the calculated NPV. The choice of a model for determining terminal value is arbitrary, however. And terminal value is of secondary importance to a company that intends to build long-term value from the new business growing out of the product introduction (rather than sell the business once its high terminal value is established). Overdependence on terminal value to justify a return on investment can be neutralized by arbitrarily setting the terminal value at zero in order to remove its influence from the calculation.

The second pitfall involves selection of the discount rate—the return on the hypothetical safe investment. Theoretically, the discount rate should be set to the cost of capital or the current return on U.S. treasury bonds. Some analysts alternatively establish a discount rate to simulate presumed risk or to match a corporate growth objective. But the cost of capital can vary over time. And risk can be oversimplified when made a component of the discount rate. Variances in the discount rate among different projects can complicate comparisons.

One way to deal with these problems is to set the discount rate by edict as the same for all comparisons and to simulate risk through a sensitivity analysis (see below). A better approach might be to use the investor's rate of return (IRR), a calculation that determines the compounded annual return in percentage terms for the money that has been invested in the project. IRR uses neither a terminal value nor a discount rate. Removing these sources of variation gives a cleaner calculation and facilitates project comparisons.

The common practice of using the discount rate as a way of simulating risk obscures the critical need to identify the major components of risk early in the project. This can be achieved through a sensitivity analysis based on the construction of multiple scenarios that simulate the impact of possible unfavorable outcomes (see sidebar, page 119). For example, top-line revenues can be adjusted to simulate loss of market share or price erosion resulting from competition. Changes in operating margin can be used to reflect higher-than-anticipated supply costs or an inability to reach the desired manufacturing scale. Moving revenues out in time while leaving costs unchanged can be an effective way to model delays in product release. The carefully prepared sensitivity analysis can be invaluable to executives for identifying critical project parameters and setting go/no-go decision points at critical milestones.

Market Opportunity. It is virtually impossible to develop a meaningful market forecast for a new product until the product's performance capabilities can be defined. For a marketing manager to develop a forecast for a new product based on what the project's champion thinks or hopes the product will be able to do is a classic career-limiting mistake. What is dreamed of may not become reality.

However, an estimate of the project's financial potential is genuinely necessary both for project prioritization and in order to justify investment in it or to assign resources. A systematic determination of the product's market opportunity can meet these needs during the feasibility phase of the project.

Market opportunity for these purposes of planning is the potential sales that could be achieved if the product fulfilled all the highly desirable expectations, particularly that it would be clearly superior to all of its competitors with no obstacles to customer adoption (see sidebar, page 120). Precision is not possible, so this opportunity should be expressed in broad ranges. Might the market opportunity for a proposed new device or diagnostic system be $10 million, $50 million, or $200 million per year? The number could be derived by taking the total number of facilities that use the type of device or system, assuming that 20% of them purchase a new one each year, and that all will choose the obviously superior new product.

Alternatively, the analysis can be made more realistic by assuming that the company's share of the new purchases will be that of the current market-share leader. It is also important to factor in growth rate. Rapidly growing markets tend to present more opportunity than markets that are growing slowly or declining. A large market opportunity does not guarantee success, but a market opportunity that is too small in terms of either size or growth to justify committing the resources necessary to realizing the product concept is grounds for project termination. Expression of market opportunity in terms of units sold can also help provide the company's operations group with the information it needs for early-stage production planning.

Product Forecast. As the feasibility phase nears its end, sufficient information should be available to allow the marketing department to generate a complete product forecast. The performance of the product in terms of its key attributes should be known. Market research should be completed, and customer impressions of the product gathered from alpha testing. This input provides the necessary raw material for the forecast.

What distinguishes the stage where development of a product forecast is possible from a mere understanding of market opportunity is the ability to accurately test how attractive the actual product attributes are to target customers. Medtech company executives can reasonably expect the forecast to be in place when the product design is frozen and the project enters commercial development.

Companies will set different standards, but the product forecast typically will provide top- and bottom-line financials, return on investment based on a DCF model, the rate and number of product placements, and the unit production forecast required by the manufacturing department. Major changes become progressively more expensive at this stage, and all departments require a higher level of certainty. Any significant uncertainties yet remaining should be clearly identified at this time and their resolution included in the commercial development plan.

Test Markets. Test marketing provides the last opportunity for the medtech company to determine whether its offering—the complete package of new product and associated services—is sufficiently attractive to its target customers. The use of test markets can be part of the beta testing program or it may constitute the first stage of postlaunch commercialization. It is usually reserved for the most innovative products or for cases in which issues involving support, distribution, or production capabilities remain to be resolved. Test markets typically consist of customers who receive final product selected on the basis of geography or some other useful criterion.

Conclusion

One final consideration, perhaps the most important, needs to be addressed. New product development is a tough challenge, as evidenced by low success rates for truly innovative projects. Through some combination of intuition, experience, and insight, certain individuals can be effective in a product development environment that is characterized by uncertainty and high risk. Companies that seek out and reward people who thrive on challenging assignments are likely to be successful.

Skills, as an expression of knowledge, can be taught, whereas talents are simply elements of personality. Writing a project plan, developing a Gantt chart, and creating a new product forecast are all skills that can be learned. The ability to make the right decision at the right time, under pressure and with incomplete information, is a talent. Equally important is the talent to motivate others to keep trying when all seems lost. Finally, the ability to deliver bad news regarding project setbacks and yet maintain the support of senior company management is a talent. Medical technology company executives have to find individuals with the talents essential for new product development and give them what they desire most: the opportunity to take the risks necessary to deliver innovative new products to customers who will be delighted with them.

Richard S. Schifreen, PhD, is business unit leader for molecular diagnostics at Promega Corp. (Madison, WI).

REFERENCE

1. M Menke, "Strategic R&D Portfolio Management: Improving Biomedical Research Quality and Productivity" (paper presented at the Executive Program in Biomedical Research Management, Harvard School of Public Health, Cambridge, MA, March 31, 1993).

Copyright ©2001 MX