Evidence of pioneering by scientists in scientific instruments

A second study shows the same clear pattern of user pioneering of new markets and applications. In this case the contrast is not between household sector free innovators and producers; it is between scientists employed by universities and firms and producers of scientific instruments. But the motivational distinction is the same: Scientists developed and improved novel instruments in order to use them in their scientific work, whereas producers developed novel instruments in order to sell them to many users.

William Riggs and I studied the sources and the timing of important innovations affecting two related types of instruments used in electron spectroscopy (Riggs and von Hippel 1994). Electron spectroscopy for chemical analysis (ESCA) and auger electron spectroscopy (AES) are both used to analyze the chemical compositions of solid surfaces (Riggs and Parker 1975; Joshi, Davis, and Palmberg 1975). In our 1994 study, Riggs and I identified 64 innovations judged to be important by both users and producers expert in these instrument types. The period of development studied began with the initial inventions in about 1953 and extended to 1983.

As can be seen in figure 4.2, the pattern of important innovations in ESCA and AES is very similar to that in whitewater kayaking. Scientists

Figure 4.2

Source of important innovations in two types of scientific instruments over time. Graph a represents frequency of innovation; the first user innovations were developed around 1953, the first manufacturer innovations were commercialized around 1969. In graph b, the vertical axis represents millions of constant dollars, with a base period of 1982-84. Source: Riggs and von Hippel 1994, figure 2.

were the initial developers of both instrument types, and also of all early important improvements; producers only begin to innovate years later, with their first important innovations being commercialized in 1969. Also note that, just as in whitewater kayaking, the frequency with which both scientists and producers generated important innovations eventually declined, even though the combined sales of ESCA and AES instruments were rising (figure 4.2b).

The distinction between the innovation motives of scientists and producers is evidenced by a clear difference in the types of innovations they developed. As can be seen in table 4.2, scientists tended to develop innovations that enabled the instruments to do qualitatively new types of things for the first time. Such functions might have been of interest only to the innovators themselves, or they might also have been of interest to some additional fraction of the market. In contrast, manufacturers tended to develop innovations that made an instrument more convenient and more reliable in generalā€”attributes of at least some interest to all potential customers. For example, scientist users were the first to modify the instruments to enable them to image and analyze magnetic domains at sub-microscopic scales, a capability of interest to only some users. In contrast, producers were the first to computerize instrument adjustments to improve ease of operation, a matter of interest to all users. Sensitivity, resolution, and accuracy improvements fall somewhere in the middle, as the data show. These types of improvements can be driven by scientists seeking to do specific new things with their instruments, or by producers applying their technical expertise to improve the products along known general dimensions of merit, such as accuracy (von Hippel 2005).

Table 4.2

Sources of scientific equipment innovations by nature of improvements effected

Type of improvement provided by innovation

Innovation developed by

Total (n)



New functional capability




Sensitivity, resolution, or accuracy improvement




Convenience or reliability improvement




Source: Riggs and von Hippel 1994, table 3. Sample size 64.

The difference in focus between scientist innovators and producer innovators can also be seen in the scientific vs. commercial importance of the innovations the two types of innovators developed. Riggs and I found that the scientific importance of scientist-developed innovations was on average significantly higher than that of producer-developed innovations (p < .001). However, the commercial importance of producer-developed innovations was on average significantly higher than that of scientist-developed innovations (p < .01).

How can we understand these patterns? I propose that the logic is identical to that discussed for the kayaking innovation study described earlier. Scientists, with their rewards based on the research value of the innovations to their own work and on the "scientific importanceā€ of their developments, innovate first. They are not concerned with the potential size of a commercial market for their innovations. In contrast, producers wait until the nature, the scale, and the potential profitability of the market are clear before investing in innovation development. And when they do invest, producers tend to focus on developing innovations of interest to the entire market, such as convenience and reliability improvements, rather innovations of interest only to some segments of the market.

< Prev   CONTENTS   Source   Next >