Although companies say they want breakthrough products, most are far more adept at making incremental improvements to existing lines. Several divisions at 3M are overcoming the bias toward incrementalism by using a process that systematizes the development of breakthroughs.

by Eric von Hippel, Stefan Thomke and Mary Sonnack

Eric von Hippel is a professor at the Massachusetts Institute of Technology’s Sloan School of Management in Cambridge. Stefan Thomke is an assistant professor at Harvard Business School in Boston. Mary Sonnack is a Division Scientist at 3M Corporation. Sonnack and von Hippel and Joan Churchill are coauthors of a handbook on the lead user process to be published in 2000 by Oxford University Press.

When senior managers think of product development, they all dream of the same thing: a steady stream of breakthrough products-the kind that will allow their companies to grow rapidly and maintain high margins. And they set ambitious goals to that end, demanding, for example, that a high percentage of sales come from products that did not exist a few years ago. Unfortunately, the development groups of many companies don’t deliver the goods. Instead of breakthroughs, they produce mainly line extensions and incremental improvements to existing products and services. And given the pace of change in today’s markets, that’s a recipe for decline, not growth.

Given the imperative to grow, why can’t product developers come up with breakthroughs more regularly? They fail primarily for two reasons. First, companies face strong incentives to focus on the short term. To put it simply: although very new products and services may be essential to future growth and profit, companies must first survive to get to the future. That necessity tends to focus companies strongly on making incremental improvements in order to keep sales up and current customers-and Wall Street analysts-happy. Second, developers simply don’t know how to achieve breakthroughs, because there is usually no effective system in place to guide and support their efforts.

The lack of a system to guide product developers who are seeking to create breakthroughs is a problem even for a company like 3M, long known for its success with innovation. Traditionally, the company’s management has fostered innovation by taking a get-out-of-the-way attitude toward product developers who, in turn, have worked according to the aphorism "It’s better to seek forgiveness that to ask for permission." This relationship between managers and developers has seen the creation of a long line of profitable products, from waterproof sandpaper and Scotch tape in the 1920s to Post-it Notes and Thinsulate in the 1970s.

But by the mid-1990s, the company’s top managers were very concerned that too much of the company’s growth was coming from incremental improvements to existing products. Breakthroughs were fewer and farther between. The demands for-and the rewards from-incremental improvements provided the company with a strong incentive to focus on current products. To begin to counter this trend, management set a bold objective: that 30% of sales would come from products that had not existed four years earlier.

For the company to meet that goal, many people within it-senior managers, marketers, product developers, scientists-would have to change how they approached their work. Accordingly, some 3M people began to become acquainted with a new method for developing breakthrough products: the lead user process. The process-which makes the generation of breakthrough strategies, products, and services systematic-is based on two major findings by innovation researchers.

First, the researchers found that many commercially important products are initially thought of and even prototyped by product users rather than manufacturers. Second, they discovered that such products tend to be developed by "lead users"-companies, organizations or individuals that are well ahead of market trends and have needs that go far beyond those of the average user. Those discoveries transformed the difficult job of inventing breakthroughs into a systematic task of identifying lead users- firms or people that have already developed elements of commercially attractive breakthroughs-and learning from them.

Consider how an automobile manufacturer would apply the lead user process. If the company wanted to design an innovative braking system, it might start by trying to find out if any innovations had been developed by drivers with a strong need for better brakes, such as auto racers. It wouldn’t stop there, however. Next, it would look to a related but technologically advanced field where people had an even higher need to stop quickly, such as aerospace. And, in fact, aerospace is where innovations such as ABS braking were first developed: military and commercial aircraft pilots have a very high incentive to stop their vehicles before running out of runway.

In September 1996, a product development team in 3M’s Medical-Surgical Markets Division became one of the first groups in 3M to test the merits of the lead user process. The team was charged with creating a breakthrough in the area of surgical drapes-the material that prevents infections from spreading during surgery. By November 1997, the team had come up with a proposal for two new line extensions, one breakthrough product, and a breakthrough strategy that would take a revolutionary approach to treating infection. And the team may have done even more for 3M’s long-term health: it persuaded senior managers that the lead user process could indeed systematize the company’s development of breakthroughs. Before we turn to that story, however, we must first explain how the process differs from other methods of product development.

Learning from Lead Users
All processes designed to generate new ideas for products begin with information collected from users. What separates companies is the kind of information they collect and who they collect it from.Teams are usually taught to collect information from users at the center of their target market. They try to learn what people at the center need by conducting focus groups and analyzing sales data, reports from the field, customer complaints and requests, and so on. Then they rely on their own creative powers to brainstorm their way to new ideas. Teams following this approach assume that it is the role of users to provide information about what they need and the job of in-house developers to use that information to create new product ideas.
The lead user process takes a fundamentally different approach. It is designed to collect information about both needs and solutions from the leading edges of the target market and from markets that face similar problems in a more extreme form. Teams using this approach assume that lead users outside the company have already generated innovations and that their job is to track down especially promising lead users and adapt their ideas to the business’s needs.

True lead users are very rare. To find them quickly, project teams use telephone interviews to network their way into contact with experts on the leading edge of the target market. Networking is effective because people with a serious interest in any topic tend to know of others who know even more than they do. Team members begin by briefly explaining their problem to individuals they believe to have some expertise - for example, authors who have written about the topic. Then, they ask for a referral to someone who has even more relevant knowledge. It’s usually not long before a team reaches lead users at the front of the target market. The next step is to continue networking until lead users are found in markets and fields that face similar problems but in different and often more extreme forms. Those people can help teams discover truly novel solutions to needs in the target market. (See the exhibit to come.)

Consider how a team focused on medical imaging carried out its work. Its members knew that a major trend in this field was the development of capabilities to detect smaller and smaller features-very early-stage tumors, for instance. The team networked to the leading edge of the target market and identified a few radiologists who were working on the most challenging medical-imaging problems. They discovered that some of those researchers had developed imaging innovations that were in advance of commercially available products.
Team members then asked the radiologists for the names of people in any field who were even further ahead in any important aspect of the imaging problem they were working on. The radiologists identified, among others, specialists in pattern recognition and people working on images showing the fine detail in semiconductor chips.

Lead users in pattern recognition proved especially valuable to the team. Specialists in the military had long worked on computerized pattern-recognition methods, because military reconnaissance experts had a strong need to answer questions such as, "Is that a rock lying under that tree, or is it the tip of a ballistic missile?" These lead users had developed ways to enhance the resolution of the best images they could get by innovating with pattern-recognition software.Lead users often help project teams improve their understanding of the nature of the breakthrough they are seeking. For example, the medical imaging team started out with the goal of developing ways to create better high-resolution images. Their discovery of military specialists using pattern recognition led them to a new goal: to find enhanced methods for recognizing medically significant patterns in images, whether by better image resolution or by other means.

After identifying lead users in the target market and in fields with analogous problems, teams collect information from lead users and applying it to create a breakthrough commercial product or service. Some information from lead users is transferred in the course of telephone interviews or through on-site visits. More is transferred when the team hosts a workshop that includes several lead users with a range of expertise as well as a number of people from within the company-product developers, marketing specialists, manufacturing people.

A lead user workshop typically lasts for two or three days. During that time, the assembled group combines their insights and experience to design product concepts that precisely fit the sponsoring company’s needs. In the medical imaging example, lead users with a variety of experience were brought together: people on the leading edge of medical imaging, people who were ahead of the trend with ultrahigh resolution images, and experts on pattern recognition. Working together, they created a solution that best suited the needs of the medical imaging market and represented a breakthrough for the company. (See the insert "Manufacturers Build on Lead User Innovations")

Manufacturers build on lead user innovations
It is rare for a manufacturer to simply adopt a lead user innovation "as is." Instead, a new product concept that suits a manufacturer’s needs and market is based on insights gained from a number of lead users and from manufacturer developers as well.

For example, ABS braking systems developed for autos were based on the principles pioneered in a number of aerospace applications. However, auto firm developers did a great deal to adapt the basic concept to the cost and performance needs of the auto marketplace.

Similarly, a lead user study of home banking services uncovered innovations by lead users that showed significant advanced functionality that would be useful in the target market. However, the lead users were considerably more "computer literate" than ordinary users, and so adaptation was required to bring the functionality that lead users had pioneered to more ordinary users in a user-friendly form.

Diving In the Deep End
In 1996 Rita Shor, a senior product specialist in 3M’s Medical-Surgical Markets Division, heard an in-house lecture on the lead user process. Shor had been charged with the task of developing a breakthrough product for the division’s surgical-drapes unit, and she needed help. Traditional market research was providing abundant data but could not point developers toward a breakthrough. Shor called Mary Sonnack, a 3M division scientist and internal consultant. Sonnack, sponsored by Chuck Harstad, vice president of corporate marketing, and William Coyne, senior vice president, research and development, had spent the 1994-1995 academic year studying the lead user process with Eric von Hippel at MIT. Shor put the problem to Sonnack in stark terms: "Our business unit has been going nowhere. We’re number one in the surgical drapes market, but we’re stagnating. We need to identify new customer needs. If we don’t bring in radically new ways of looking for products, management may have little choice but to sell off the business." Sonnack, after warning Shor about the high level of commitment that would be needed-both from team members and from senior management-agreed to work with her.

At 3M, surgical drapes were an extension of the company’s innovations with masking tape. The drapes work by isolating the area being operated on from most potential sources of infection-the rest of the patient’s body, the operating table, members of the surgical team. But the diversity of the microbial world constantly challenged this fortress of fabric, which couldn’t cover, for example, catheters or tubes being inserted into the patient.
By the mid-1990s, surgical drapes were bringing 3M more than $100 million in annual sales. But the unit in charge of the draping business had not had a breakthrough product for almost a decade. Technological excellence was not the issue. In the early 1990s, the division had spent three years developing technologically advanced disposable gowns for surgeons that would safeguard them and their patients-and keep them more comfortable-by allowing water vapor but not viruses to pass through microscopic pinholes in the fabric. This technological and manufacturing feat, however, came to the market just as managed care was taking hold in the United States. Surgeons loved the fabric, but insurers wouldn’t pay for it, and sales were disappointing.
In short, the division saw little room for growth in existing markets, declining margins on existing products and, because of the drapes’ cost, few opportunities to penetrate less developed countries. Under those circumstances, Rita Shor was able to persuade senior management of the merits of trying the lead user process. A few weeks later, she and her project co-leader Susan Hiestand had assembled a team of six people from R&D, marketing, and manufacturing. They all agreed to commit half their time to the project until it was completed.

Looking for Lead Users
The team’s initial goal was, in essence, "Find a better type of disposable surgical draping." That was admittedly not a very creative statement of the problem, but the quality of the first framing of the problem is not critical to the project’s success. Experts and lead users are never shy about suggesting better ideas, and the evolutionary improvement of goals is an expected and desirable part of the lead user process. The group spent the first month and a half of the project learning more about the cause and prevention of infections by researching the literature and by interviewing experts in the field. They then held a workshop with management in which they discussed all that they had learned and set parameters for acceptable types of breakthrough products. (This work constituted phase 1 of the lead user process; see the insert "Step by Step Through the Process.")

Step by Step Through the Process
The lead user process gets under way when a cross-disciplinary team is formed. Teams typically consists of four to six people from marketing and technical departments, with one member serving as project leader. Team members usually will spend 12 to 15 hours per week on the project for its duration. That high level of immersion fosters creative thought and sustains the project’s momentum.
Lead user projects proceed through four phases. The duration of each phase can vary quite a bit-the 3M team spent six months alone on phase 3, when it researched surgical conditions in developing countries through on-site visits-but for planning purposes, a team should figure on five to six weeks for each phase.

Phase 1: Laying the Foundation. During this initial period, the team identifies the kinds of markets they want to target, as well as the type and level of innovations desired by key stakeholders within the company. If the team’s ultimate recommendations are to be credibly received, these stakeholders must be on board early in the project.

Phase 2: Determining the Trends. It’s an axiom of the process that lead users are ahead of the trend. But what is the trend? To find out, the team must talk to experts in the field they are exploring; that is, to people who have a broad view of emerging technologies and leading-edge applications in the area under study.

Phase 3: Identifying Lead Users. The team now begins a networking process to identify and learn from users at the leading edge of both the target market and related markets. The group’s members focus on gathering information that will help them to identify especially promising innovations and ideas that might contribute to the development of breakthrough products. Based on what they learn, teams also begin to shape preliminary product concepts and to assess the concepts’ business potential and fit with company interests.

Phase 4: Developing the Breakthroughs. The goal is to move the preliminary concepts toward completion. To that end, the team begins this phase by hosting a workshop with several lead users (the 3M team invited 11), as well as half a dozen in-house marketing and technical people and the lead user group itself. Such workshops may run from two or three days. During that time, the participants join together first in small groups and then as a whole to design final concepts that precisely fit the company’s needs.

After the workshop, the project team further hones the concepts, determines whether they fit the needs of target-market users, and eventually presents its recommendations to senior managers. By that point, its proposals will be backed by solid evidence that explains why customers will be willing to pay for the new products. Although the project team may now disband, at least one member stays involved with any concepts that are chosen for commercialization. In that way, the rich body of knowledge that was collected during the process continues to be useful during the remaining steps of product development and marketing.

For the next six weeks or so, team members focused on getting a better understanding of important trends in infection control. One cannot specify what the leading edge of a target market might be without first understanding the major trends in the heart of that market. Much of the team’s research at this early stage was directed at understanding what doctors in developed countries might need. However, as the group’s members asked more and more questions and talked to more and more experts, they came to realize that they didn’t know enough about the needs of surgeons and hospitals outside the developed world, where infectious diseases are still major killers. The team broke up into groups of two and traveled to hospitals in Malaysia, Indonesia, Korea, and India. They learned how people in less than ideal environments attempted to keep infections from spreading in the operating room. They especially noted how some surgeons combated infection by using cheap antibiotics as a substitute for disposable drapes and other, more expensive asepsis measures.

As a result of their field observations, the team concluded that a crisis was germinating in the surgical wards of developing countries. Doctors’ reliance on cheap antibiotics to prevent the spread of infection would not work in the long run-bacteria would develop resistance to the drugs. The team also realized that even if 3M could radically cut the cost of surgical drapes, most hospitals in developing countries simply would not be able to afford them. These insights led the team to redefine its goal: find a much cheaper, much more effective way to prevent infections from starting or spreading that does not depend on antibiotics-or even on surgical drapes. The team then networked their way into contact with innovators at the leading edge of the trend toward much cheaper, more effective infection control. As is usually the case, some of the most valuable lead users turned up in surprising places. For example, the team learned that specialists in some leading veterinary hospitals were able to keep infection rates very low despite facing difficult conditions and cost constraints. As one of the country’s foremost veterinary surgeons explained to them, "Our patients are covered with hair, they don’t bathe, and they don’t have medical insurance, so the infection controls that we use can’t cost much." Another surprising source of ideas was Hollywood. One of the team members learned that make-up artists are experts in applying materials to skin that are non-irritating and are easy to remove when no longer needed. These attributes are very important to the design of infection control materials applied to the skin.

As a final step in the project, the team invited several lead users to a two and a half day workshop. (As the insert "Why Lead Users Will Talk to Your Company" makes clear, the lead users’ reward for participating was purely intellectual; they all signed any property rights that might result from the workshop to 3M.) The bold central question, which had come out of the team’s research, was now this: "Can we find a revolutionary, low cost approach to infection control?" The participants met for several hours at a time in small groups; the composition of the groups was then changed and the process continued. Some groups floundered for a while before pulling ideas together toward the end of their sessions. In others, extroverted people at first dominated the discussion; later, the introverts warmed up and began contributing. All groups faced the challenge of navigating a sea of facts and trying to unite creative ideas with technical constraints.

Why Lead Users Will Talk to Your Company
We have found that most lead users are quite willing to give detailed information to manufacturers, and are generally willing to do so for free. Why? Consider that lead users develop innovations because they must. Occasionally their results give them some kind of competitive advantage; in those cases, they will not be willing to share what they know. More often, however, they would much rather have someone else take the steps of producing and marketing their innovations on a broad scale, or they would suffer no financial loss by revealing what they know. In such instances, they are generally happy to share their knowledge with companies.

For example, in a lead user study devoted to improving credit-reporting services, a team found that at least two major users of such services had developed advanced, on-line credit-reporting processes. One of the users regarded the service it had developed as a significant source of competitive advantage and refused to discuss any details with the team. The other, however, welcomed the team with open arms and fully revealed its system, saying that "We only developed this in the first place because we desperately needed it-we would be happy if you developed a similar service we could buy."

It is always good practice for lead-user project teams to tell people they are interviewing right up front that their company may have a commercial interest in the ideas under discussion. When someone expresses hesitation about discussing his or her ideas, the interview comes to an end and the team member moves on to find others who do not have such concerns. In the end, the workshop generated concepts for six new product lines, as well as a radical new general approach to infection control. The lead user team chose the three product line concepts that they felt were strongest to present to senior management. The first recommendation was for an economy line of surgical drapes. These drapes could be made with existing 3M technology and thus would not constitute a breakthrough product; nevertheless, they would be welcomed in the increasingly cost-conscious developing world.

The second recommendation was for a "skin doctor" line of hand-held devices. These devices would eventually be able to do two things: layer antimicrobial substances onto the skin during an operation and vacuum up blood and other liquids during surgery. These products provided a significant advance in user functionality, and could be based on existing 3M technologies.

The third new product proposal was for an "armor" line that would coat catheters and tubes with antimicrobial protection. This, too, could be created with existing 3M technology, and promised to open up major new market opportunities for 3M. The company had previously focused solely on products designed to prevent surface infections; the armor line would allow it to enter the $2 billion market aimed at controlling blood-borne, urinary tract, and respiratory infections.

Changing Strategy
When a lead user team probes until it finds innovators at the leading edge of target and other markets, the questions and answers it develops often point toward the need for strategic change. Indeed, that’s what happened at 3M. In addition to concepts for novel product families, the team had identified a revolutionary approach to infection control, but following it would require a change in the division’s strategy.

Until this point, the division had focused on products that were, in a sense, one size fits all. In other words, every patient, regardless of the circumstances that brought him or her there, would get the same degree of infection prevention from the same basic drapes. In the course of their research, the team members learned that some people entered the hospital with a greater risk of contracting infection-because of malnutrition, for example, or because they were diabetic. Doctors thus wanted a way of treating individual patients according to their needs through "upstream" containment of infections. That is, they wanted ways to treat people before surgery in order to reduce the likelihood of their contracting disease during an operation. Should 3M move in that direction? The project team debated among themselves the wisdom of proposing a strategic change to senior managers. According to one team member, "In thinking about challenging the entire business strategy, we were crossing boundaries. I think the lead user methodology had pushed us in that direction. It allowed us to gather and use information in a different way than we had before, and it also provided emotional support for change. Based on extensive research, we were suggesting a major change-but as a team. We didn’t feel like lone rangers."

But not everyone on the team wanted to make this last recommendation. One feared that senior management might reject all the team’s proposals if they went ahead with it. In the end, they decided to make the case for strategic change and successfully persuaded senior management to go along with it. The exact nature of the change is proprietary information, and we can’t reveal it here. But 3M’s senior management believes it will produce very positive and far-reaching bottom-line results for the company.

3M experimented with the lead user process in order to break free from incrementalism-a mind-set that many companies fall prey to because, in the short term, it keeps sales and profits up. The process offered 3M a systematic way to generate new products and strategies based on a deep understanding of the leading edge of rapidly moving markets. It brought cross-functional teams into close working relationships with leading-edge customers and other sources of expertise.

Thousands of Innovators
It’s often the case that innovative product users far outnumber a company’s developers. For example, many people believe that user-developed software products-Apache Web-server software, for example-are better than commercially developed products. That’s less surprising when you consider that more than half a million sites use Apache software, and that thousands of users participate in developing and supporting it. That is many times the number of people that a commercial software developer like Microsoft can afford to dedicate to server software development and support.
And consider video game development. Sony recently set up a Web site to support hackers interested in exploring and developing new types of games that could be played on the Sony Playstation. It quickly attracted 10,000 participants, a number that vastly exceeds the number of in-house and contract developers creating games for the Playstation. It’s possible that, taken individually, in-house developers are technically more skilled than most user-developers. But the user-developer community mobilized by Sony has a huge diversity of interests and skills. Sony’s vice president of third-party R&D, Phil Harrison, thinks that some of them will come up with "some radically new forms of creativity that will break the conventions that are holding back the business today."

3M believes that the process is providing the systematic approach to the generation of breakthroughs that had been missing. The company’s head of R&D, William Coyne, reports that "we have now tested the lead user method in eight of our 55 divisions. Corporate management is very enthusiastic about the process, and the line of 3M people interested in learning the method from Mary Sonnack’s group [3M’s Lead User Process Center of Excellence] extends out her office door and around the block."

Most companies can happily make room for the lead user process. That’s not to say that the process will crowd out all other ways of working, or that incremental improvements have no value-obviously, they do. But they’re not sufficient anymore, if they ever were. Does the lead user process always guarantee success? Of course not; nothing can. But by giving companies a way of finding out who the people on the cutting edge are-the people who are so impatient and so much in need of the next best thing that they are willing to make it for themselves-the lead user method opens up new avenues. It takes teams and companies in directions they wouldn’t have imagined during the day-to-day crush of business.