Content uploaded by Daniel A. Levinthal
Author content
All content in this area was uploaded by Daniel A. Levinthal
Content may be subject to copyright.
The Emergence of Emerging Technologies
Ron Adner
INSEAD
Boulevard de Constance
77305 Fontainebleau Cedex, France
ron.adner@insead.edu
Daniel A. Levinthal
Wharton Business School
2000 Steinberg-Dietrich Hall
Philadelphia, PA 19104
levinthal@wharton.upenn.edu
(Forthcoming, California Management Review)
September 2002
*INSEAD Fontainebleau, France **The Wharton School, University of Pennsylvania. We thank
the editors, three anonymous reviewers and Robert Gunther for their comments on a prior draft.
2
The Emergence of Emerging Technologies
Ron Adner and Daniel Levinthal
What is discontinuous about the moment of radical technological change? We suggest that the
discontinuity typically does not lie in a radical advancement in technology itself; rather, the
discontinuity stems from a shift of an existing technical lineage to a new domain of application.
Seeming revolutions such as wireless communication and the internet did not stem from an isolated
technical breakthrough. Rather, the spectacular commercial impact was achieved when an existing
technology was re-applied in a new application domain. We use the biological notion of speciation
events, which form the basis for the theory of punctuated equilibrium, to reconcile the process of
incremental change within a given line of technical development with the radical change associated
with the shift of an existing technology to a new application domain. We then use this lens to explore
how managers can cope with, and potentially exploit, such change processes.
3
Discussions of technology evolution have offered sharply contrasting perspectives of the
pace and mechanisms of technological change. On the one hand, we have arguments regarding the
gradual, incremental nature of technological change (Dosi, 1983; Basalla, 1988; Rosenbloom and
Cusumano, 1987). In contrast, others have offered the image of technological change as being
rapid, even discontinuous (Tushman and Anderson, 1986; D’Aveni, 1994). Indeed, the locus
classicus of evolutionary perspectives of technological change (Schumpeter, 1934) offers the
dramatic imagery of “waves of creative destruction”. How can these contrasting perspectives be
reconciled?1
The theory of punctuated equilibrium (Gould and Eldridge, 1997), developed in the context
of evolutionary biology, provides a powerful framework to integrate ideas of gradual change in
underlying science with apparent discontinuities in the commercial application of technologies.2
Gould and Eldridge confronted a fossil record that seemed inconsistent with the gradualist
interpretation of Darwin’s ideas. Contrary to the expectation of an incremental process of descent
with modification, they identified periods in which there seemed to be bursts of evolutionary activity.
Their resolution of empirical evidence and Darwin’s Theory was to note the importance of
speciation events --- the separation of one evolving population from its antecedent population, which
in turn allows populations to follow different evolutionary paths.
There are two critical features of speciation. One is that it is genetically conservative ---that
is, speciation is not triggered by a transformation of the population. Second, the speciation event
1 The arguments in this paper draw on our previous research into the nature of technology evolution and
technology competition (Adner and Levinthal, 2000; 2001a; 2001b; Levinthal 1998; and Adner 2002).
2 Mokyr (1990) applied the notion of punctuated equilibria to the process of technical change. However, his
argument hinges on the presence of occasional, dramatic mutation events and not the notion of speciation
events that are critical in the underlying theory of punctuated equilibrium as developed by Gould and Eldrigde
(1977) and as applied to the process of technical change by Levinthal (1998).
4
allows the two populations to grow quite distinct as a result of their now different selection
environments.
What can this framework tell us about the evolution of new technologies? How can it help
resolve the discordant images of both gradual and radical technical change? In particular, how can it
help us identify those critical transition points when emerging technologies realize commercial
importance?
The analogue of speciation in technological development is the application of existing
technologies to a new domain of application. Technological discontinuities are generally not the
product of singular events in the development of the technology itself. As in the biological context,
the critical factor is often a speciation event, transplanting the existing technological know-how to a
new application domain where it evolves in new directions. The technological change associated
with the shift in domain can be quite minor; indeed, in some instances, there is no change in
technology.
While the speciation event is, in an immediate sense, technologically conservative, it may
have significant commercial impact. In the new application domain, the resources available and
selection forces present may result in rapid subsequent technical development. Consider these ideas
in the context of the development of wireless communication technology. Wireless communication
technology has, at many junctures, been heralded as revolutionary, including the introduction of
wireless telegraphy, radio broadcasting, and wireless telephony. However, beneath these seemingly
radical changes was a gradual technological evolution within a lineage in which dramatic changes
were initiated by the application of existing technology to new domains of application:
5
• Laboratory device: Wireless communication technology started as a laboratory device
used by German physicist Heinrich Rudolph Hertz to test Maxwell’s theories on
electromagnetic waves. The critical functionality in this domain was the reliable
measurement of electromagnetic waves.
• Wireless telegraphy: The development of wireless telegraphy was driven by the ability
of Marconi to generate financial backing for a corporation to pursue the commercial
application of electromagnetic waves to serve remote locations such as ships and
lighthouses for which wired telegraphy was not an alternative (Garrat, 1994). For this
second application domain of wireless telegraphy, a new functionality of distance was
required. Researchers focused on enhancing the power of transmitters and increasing
the sensitivity of receivers. The effort to develop superior receivers for wireless
telegraphy (and an effective repeater for wired telephony) ultimately led to the
development of the vacuum tube (Aitken, 1985).
• Wireless telephony and broadcast radio: The vacuum tube provided the basis for a
continuous wave transmitter, a technology that allowed for the transmission of voices
and was readily applied in the new application domains of radiotelephony and
broadcasting. Wireless telephony was initially used for public safety purposes, such as
police and emergency services. Only in recent years has wireless technology penetrated
more mainstream, mass consumer markets. The application of the vacuum tube to
broadcast radio first emerged with ham (wireless telegraphy) operators and then was
rapidly refined by the already established corporate entities of Westinghouse, RCA, and
General Electric.
The initial prototype of the technology that entered each new domain, whether it was Hertz’s
laboratory equipment, Marconni’s early wireless, or broadcast radio, was readily derived from the
existing state of knowledge. The shifts in application domains, however, were significant
6
breakpoints in the technology’s development because they signaled a shift in selection criteria --- a
shift in the critical functionality by which the technology would be evaluated. In addition, the shifts
radically changed the resources available to support the development of the technology. Contrast
Hertz’s assembly of components laying about the laboratory he took over in Karlsruhe, Germany
with Marconni’s ability to generate financial backing for a corporation to pursue the commercial
application of electromagnetic waves as an alternative to wired telegraphy (Garratt, 1994) and the
commitment of resources by the already established corporate entities of Westinghouse, RCA3, and
General Electric to its refinement.
Yes, wireless communication technology has undergone extraordinary change in the hundred
years since Hertz’s experiments. However, the dramatic breakthroughs that set the technology on a
new course were as much discoveries of new domains of application, as advances in the underlying
technology. These “speciation” events were, of course, made possible by wonderfully creative
development efforts that commanded tremendous amounts of time and financial resources.4
However, these efforts were supported within the existing application domains. Broadcast radio
and wireless telephony could not have been possible in the absence of continuous wave transmitters,
but the impetus to develop that technology and the resources to do so came from efforts to enhance
3 The name Radio Corporation of America (RCA) is potentially confusing in this context. RCA was founded by
General Electric, AT&T, and United Fruit (joined latter by Westinghouse) in order to pool their patents in the
pursuit of wireless telegraphy (Aitken, 1985). RCA was not founded in anticipation of broadcast radio and
indeed the emergence of broadcast radio shortly after the founding of RCA caused considerable conflicts among
the parent companies.
4 Indeed, more generally the role of intentionality and choice in technology development clearly distinguishes it
from processes of biological change. We apply the biological framework to understand the nature of the
selection environment acting on possible technologies and, in particular, the niche structure of the resource
space, but we are not assuming a process of blind variation generation. Indeed, we assume, and argue that
actors should try to anticipate the structure of the possible selection environments in which they can develop a
given technological initiative.
7
the distance and clarity of wireless telegraphy and AT&T’s interest in developing an effective
repeater for long-distance, wired phone service.
Thus, a technology undergoes a process of evolutionary development within a given domain
of application. At some juncture, that technology, or possibly set of technologies, maybe applied to
a new domain of application. The technological shift necessitated by this event is modest. Just as
biological speciation is not a genetic revolution – the DNA of the organism doesn’t suddenly mutate
– technological speciation is not usually the result of a sudden technological revolution. The
revolution is in the shift of application domain. The distinct selection criteria and new resources
available in the new application domain can result in a technology quite distinct from its technological
lineage. Framing technology evolution in terms of speciation leads us to differentiate between a
technology’s technical development and a technology’s market application. This distinction is
useful in understanding broad patterns of technological change, and leads to specific strategic
implications for technology management.
__________________________________________________________________________
Insert Figure 1 here
___________________________________________________________________________
Certainly, not all emerging technologies exhibit this pattern. Some development efforts take
place in the context of research laboratories and have no commercial precursor prior to their initial,
dramatic commercial applications. Examples of such innovation include genetic engineering,
automated DNA sequencing, and chemical based photography. Our purpose is not to claim that all
innovation follows the incremental path we are detailing here. However, we would argue that the
pattern of speciation that we characterize here is far more common than is widely realized. Indeed,
8
many emerging technologies that are viewed as having appeared dramatically and rapidly in their
mainstream markets, such as xerography (Dessauer, 1971), home video recording (Rosenbloom
and Freeze, 1985), and the Internet (Berners-Lee, 1999), actually have a long pre-history of
technical development occurring in relative small and peripheral market segments.
Lineage Development: Selection Criteria and Resource Abundance
Given the speciation event, why might we observe a radically divergent technology emerge
and why might this lead to a rapid pace of technological change? Within our framework, the nature
and pace of technological change are driven by two elements of the selection process. One is a
process of adaptation. The technology becomes adapted to the particular needs of the new niche
that it is exploiting. The second element corresponds to the resource abundance of the niche. As a
result, the mode of development is influenced by the particular features of the niche, while the pace
of development is driven by the resources that this niche is able to provide.
A technology naturally adapts to the niche to which it is being applied (Basalla, 1988). This
adaptation reflects the distinctive needs of the niche regarding functionality. The new application
domain may value particular elements of functionality that were largely irrelevant to the prior domain
to which the technology was applied. In the disk drive industry, we see that the attributes of size,
weight, and power requirements become relevant in the new niche of portable computers
(Christensen and Rosenbloom, 1995). These same attributes had had little relevance for
manufacturers of desktop machines.
Needs should be viewed both in terms of the relative importance of various attributes, such
as different price/performance tradeoffs among potential consumers, but also the minimal threshold
9
of functionality for a technology to be viable in a given application domain (Adner and Levinthal,
2001). Thus, for example, a horse-less carriage that is likely to breakdown after a quarter of a mile
is a novelty, not a substitute for a horse.
The other class of factors is the resources available to sustain the innovative activity.5 While
a new application domain may have a distinct set of selection criteria, if the resources in this niche
are quite limited then we should not expect to observe the rapid development of new technological
forms. It is the combination of distinct selection criteria and the availability of substantial resources
to support innovative efforts associated with the new application domain that results in a speciation
event with dramatic consequences for subsequent technological development. The pace of
development becomes much more rapid if the technology is able to satisfy the needs of not only the
possibly peripheral niche to which it may have first entered but, as it develops in functionality or cost
is reduced, the technology is able to penetrate larger, more mainstream niches. A key to this
transition is the degree to which resources (scientific, managerial, organizational, complementary
capabilities) that were of value in one market niche can be leveraged in the new market niche. In this
regard, consider the market journey of Global Positioning Systems (GPS) technologies, from strictly
5 In this sense, the framework relates to the important work on the population ecology of organizations (Hannan
and Freeman, 1989). Hannan and Freeman, and a now long line of significant subsequent research (Carroll and
Hannan (2000) provide a recent summary of this area), highlight the role of the selection environment in
determining the demography of organizational populations. In particular, the early work of Hannan and Freeman
(1984) and subsequent work on resource partitioning (Carroll, 1985) demonstrate the importance of the niche
structure of the resource environment, most explicitly through the concept of carrying capacity. While sharing
this common footing, our work differs from ecological analysis in two fundamental ways. First, where population
ecology examines organizational forms, our work explores the evolution of technological forms, which can both
cross and coexist within organizational boundaries. Second, whereas population ecology assumes that
organizational forms are fixed relative to their environments, we examine the ways in which technological forms
change within and across distinct selection environments and resource spaces.
10
military applications, to geoscience and surveying applications, to trucking fleet management, to the
current penetration attempts in private automobiles.6
Creative Destruction --- the New Displacing the Old
“Creative destruction” occurs when the technology that emerges from the speciation event is
ultimately able to successfully invade other niches, possibly including the original domain of
application. For example, the 3.5-inch computer disk drives that were initially developed for the
niche of portable computers ultimately became viable for the mainstream desktop market
(Christensen and Rosenbloom, 1995). Radial tires were initially developed in the distinct niche of
high-performance sports cars (Sull, Tedlow and Rosenbloom, 1997; Foster, 1986) that valued the
high-performance of the radial tires. The resources made available from the success of radials in this
niche lead to increased efficiency in the production process. That reduction in cost, in conjunction
with a different attribute of radials --- their greater longevity relative to bias-ply tires --- allowed
radial tires to penetrate the mainstream niches of replacement tires and ultimately the original
equipment market of automobile manufacturers.
This successful “invasion” of the mainstream niche is the dramatic event on which
commentators tend to focus. However, that dramatic invasion is the outcome of a substantial period
of development in a relatively isolated niche. Prior to any lineage development, there is little
possibility that the new technological form can out-compete the refined version of an established
technology in its primary domain of application. As Rosenberg (1976) argues in his analysis of the
6 We note that resource availability is a function not only of actual penetration of larger markets, but also of
actors’ expectations of this penetration. For example, the US Congress’s 1994 decision to open the Internet to
commercial transactions released a wave of resources in the expectation that the internet would penetrate a
number of large, new application domains.
11
development of the machine tool industry, even those technologies that ultimately became widely
diffused general purpose technologies initially focused on the needs of a particular application
domain.
What permits the new technology to have some basis of viability is the existence of niches,
or peripheral elements of existing niches, that exhibit a somewhat different set of selection criteria.
Such peripheral niches may also serve as safe havens for technologies whose primary application
domains have been invaded. For instance, the teletype endured even with the full development of
telephone technology, because it had the attribute of providing a written record that was valuable in
business transactions, as well as allowing for asynchronous communication. Its demise awaited the
development of an alternative form of written networked communication --- improved facsimile
technology and the development of large scale computer networks. A final component in the
persistence of a technology in a given domain of application are various forms of switching costs.
These may be costs incurred by individual actors, or more tellingly, costs associated with network
externalities (Arthur, 1989; David, 1985).
New Combinations and New Application Domains
In our discussion of speciation, we have highlighted a particular form of recombination ---
the combination of an existing technology with a new context. Clearly, other sorts of recombination
occur as well. The development of the CAT scanner for medical imaging, for example, illustrates the
linking of two formerly disparate technologies. CAT scanning (Trajtenberg, 1990) drew upon X-
ray technology, which was already applied in the medical imaging domain, and computer technology,
12
which had been applied to data processing (see Figure 2). Of course, to make these new
combination effective in the context of medical imagining, computer science work on appropriate
imaging algorithms had to be developed as well as significant design changes in X-ray machinery to
allow for rotation and multiple scans.
Cases in which literally “off the shelf” technology can be applied to a new context are likely
to be rare. However, even in an example such as the CAT scanner, it is important to note that the
tremendous performance improvements and cost reductions in computer processing power that
made the CAT scanner possible occurred for reasons quite apart from any possible technological
opportunity of medical imaging. Note also that the expenditures that supported these advances
dwarf the relatively modest expense of developing imaging algorithms or data storage systems that
were specialized to the medical imaging niche (Trajtenberg, 1990).
Insert Figure 2 here
The notion of new combinations has been put forward most famously in the context of the
study of technical change by Schumpeter (1934). Schumpeter (1934: 66) posed the notion of
innovation as “carrying out new combinations”, where a new combination could take on one of five
different forms: (1) a new good, (2) new method of production, (3) new geographic market, (4)
new source of supply, (5) new organization of an industry.
A critical issue from our perspective is whether the new form, or combination, competes for
resources against the antecedent forms from which it was derived or whether it derives resources
from other niches in the environment. The former case may generate a process of creative
13
destruction whereby the new form substitutes for the prior form in that same application domain.
However, we suggest that new combinations are more likely to find their initial home in a new
application domain or in the peripheral realm of the existing niche.7 It is this property of not initially
competing in the niche space of the antecedent form that defines the property of speciation.
Thus, we are highlighting a particular type of Schumpeterian ‘combination” --- the
introduction of an existing technology into a new application domain, in contrast to the usual focus on
new to the world products or processes. Many of these creative recombinations produce new
forms that prove unviable in the market place. Witness the many variants of pen-based computing
and personal digit assistants (PDAs) that have been commercial failures (McGahan, Vadasz, and
Yoffie, 1997). Yet, unlike biological processes, technological evolution is not restricted to
processes of random variation and environmental selection. Agents of technical change can actively
monitor for hopeful new variants. Our encouragement for such Schumpeterian entrepreneurs
looking at existing emerging technologies is to pay particular attention to market niches that will
accept the technologies in their present form. Deploying emerging technologies in these early
markets can be a means of both realizing profits in the near term and providing valuable feedback
regarding the demand for possible attributes of technical functionality to support and guide their
further development.
Patterns of Technology Evolution
7 When speciation takes the form of occupying a wholly distinct niche it is referred to as allopatric speciation
(Mayr, 1963). More typically speciation involves the exploitation of the periphery of an existing niche (Bush,
1975).
14
At least initially, commercial discontinuities in wireless communication took the form of
providing communication services where none existed before rather than new technology for existing
services. A new technology will be viable if it out-competes existing technologies on some
performance criteria, whether an element of functionality or cost and thereby achieve a relative
advantage. It is unlikely that a new technology will initially dominate an established technology in its
primary domain of application.
For example, early wireless communications technology, despite high cost and poorer sound
quality, outperformed the more refined wired systems for mobility and flexibility. This allowed it to
build an initial niche in military and police communications where mobility was valued enough to
outweigh the other weaknesses.
In most cases, the early domains of application are ones that not only value the distinctive
functionalit y, but also ones that can tolerate relatively crude forms of the new technology. For
example, while minimally invasive surgery techniques were applied in ninety-five percent of
gallbladder procedures (a relatively simple surgical procedure) within two years of their introduction,
it took years of further development before the technologies were applied in heart surgeries.
Similarly, early pen-based computing could be used for structured forms and signature capture but
not for tasks that required handwriting recognition.
This general pattern of industry development from a small niche to an ever broader set of
niches through a series of speciation events has been observed in numerous settings. The history of
the video recorder, which characterizes the transition from a peripheral niche to the mass consumer
market, is illustrated in Figure 4. The technology was initially introduced for the distinct niche of
broadcasters. As the manufacturing process was refined and the product design simplified, it was
15
possible to penetrate a new niche of industrial and commercial users (Rosenbloom and Cusumano,
1987). Finally, this development continued to the point that the product was able to penetrate the
mass consumer electronic market. It is important to note that, at each point in its development,
video recording technology was commercially viable and profitable within the niche in which it was
operating.
_______________________________________________________________________
Insert Figure 3 here
________________________________________________________________________
In ecological terms, we might think of this as the artifact shifting from a specialist to a
generalist. Video recorder development, which previously drew resources from the narrow niche of
television broadcasters, could now draw on the mass consumer market to fund further advance. In
some cases, the technology never goes beyond the initial peripheral market, but remains an isolated
“island of application”. Gallium arsenide, for example, was heralded as a replacement for silicon in
semiconductors in the early 1980s based on the superior speed that it provided. However, the
technology has proved commercially viable only in the context of supercomputer applications and
communications devices (Wiegner, 1988). Recently, the demand for gallium arsenide has increased.
In line with our arguments, this new demand is coming from the application domain of
communication devices, not mainstream computing applications (Ristelhueber, 1993).
Implications for Firm Strategy
How can managers use this process of technological speciation to their advantage?
Biological speciation often occurs through external events that separate one population from another.
16
However, once managers understand the process of technological speciation, they can actively look
for ways to move the speciation process forward. Our recommendations do not eliminate the
uncertainty inherent in emerging technologies, but provide important insights for managers to cope
with and even exploit this uncertainty:
• Focus on the intersection of markets and applications: Technology speciation
distinguishes between a technology’s technical development and a technology’s
market application. Because of the belief that technological revolutions usually occur in
the lab, managers sometimes undervalue the importance of applications. Some writers
have focused on the impact of technological milestones on the rate of progress of
development (Sahal, 1985). We are suggesting that, while these “supply side”
considerations are important, it is critical to consider constraints and thresholds on the
demand side as well. The leap to new application domains affects the attributes of the
technology that are developed as well as the resources available for its development.
The implication is that there are probably many technological developments that could
take off, in the proper application domain.
Managers need to focus attention on the issue of potential application domains. There
may be a variety of technologies that are sitting in the laboratory that would begin to
emerge if transplanted into the right application domains. Discussions of the management
of emerging technologies emphasize long-term vision and patient investment as the key
to developing nascent technologies to the point that they can have a real impact on the
existing technological order. These discussions pay little attention to the market contexts
in which innovations are exploited and the impact these market interactions may have on
exploration activity. The wireless communications revolution came as a result of
recognizing the potential applications of technology. Which technologies in your labs
might have a similar potential if they were moved to the right domain?
17
• Focus on selecting market contexts for a technology, rather than selecting
technologies for a fixed market context: The question should be: Where can I find
an application domain in which this fledgling technology will thrive? There are many
examples of novel products based on relatively crude technologies which serve
intermediate markets while biding their time to enter the ‘mainstream’ market: solar
powered calculators were proving grounds for solar cells; digital watches and
calculators were platforms for early liquid crystal displays; inventory management
systems and simple signature capture devices were predecessors of pen based
computing. Even in these technologies, which have to date only partially fulfilled initial
expectations, firms such as Sanyo, Sharp, and Casio were able to introduce profitable
products into the market that allowed them to learn about and refine the technology.
Contrast the experience of those firms with companies that kept development in-house
until such time as they felt they could address their mainstream customers rather than
focusing on a small target segment to refine the technology. For example, ARCO’s
investments in solar energy power stations and Apple’s investments in the Newton pen
based computer were prematurely transplanted to the mainstream market in which they
were unprepared to thrive. Similarly, early success in speech recognition technologies
has been found not by those firms targeting the mass market of Language User Interface
(LUI), the holy grail that would free computer users from their keyboards, but rather in
the focused niches such as telephonic applications (e.g. directory assistance) and
medical and legal dictation, which were characterized by very well defined vocabulary
sets and by users’ high willingness to pay.
• Understand market heterogeneity: While increasing attention has been paid to the
influence of market feedback on technology management (Abernathy, 1978; Von
Hippel, 1988; Leonard Barton, 1995; Christensen, 1997; Moore, 1995), the
implications of the possible diversity of feedback on development strategies remain
relatively unexplored. Exploiting market opportunities at early stages of technology
18
development requires closer consideration of market heterogeneity. Different
consumers have different requirements for purchasing products and use different criteria
when evaluating their options. Different facets of the commercial market have different
thresholds of viability. These differences may be differences in magnitude, such as the
level of script recognition required of pen-based computers for inventory management
versus word processing applications; or they may be differences in kind, such as the
relative importance of price and performance of computers for space science application
versus home use.
An initiative that fails in one market subset may still be highly successful in another. For
example, the first users of the xerographic process were specialty printers who used it to
make offset masters. Early machines required a 14-step process to make a single copy,
which prevented them from penetrating many markets. The requirements of specialty
printers were sufficiently low however, and their complementary skills sufficiently high,
that they were able to derive benefits from the product despite the cumbersome
technology. Further, because of their understanding of the printing process, these
specialty printers aided Haloid (later renamed Xerox) in expanding the market for
xerography to other printing sub-fields, such as microfilm printing, that ultimately led to
the corporate mainstream (Dessauer, 1971; Pell, 1998).
• Expand your selection criteria: While introducing greater experimentation, the
company also needs to diversify the selection criteria it uses to evaluate initiatives.
Because firms do not have the same internal diversity as the broad market, they cannot
match the richness of the market selection environment in their own internal selection
processes. The internal selection environment of an organization, which is governed by a
hierarchical structure, cannot reflect the diversity of selection criteria of the independent
consumers that compose the market. This is why companies often tend to overlook
potential opportunities for applications of technology. Viewing the market as an
amalgamation of independent selection environments, the challenge is not to accurately
19
determine the needs of the market, but rather to recognize the variety of evaluation
criteria being applied in the market’s component segments.
Often times, new initiatives, even with positive feedback from the market, do not take
hold and develop within established firms. A common factor in such episodes is the fact
that the initiatives do not have great saliency within the operating unit in which they are
based. Management may tend to find the initiative to be a distraction from their primary
efforts. Initiatives in emerging markets are handicapped in two ways. First, they do not
fall within the existing strategic context and, as a result, do not present obvious pursuits
for the firm (Burgelman, 1991). Second, they often target markets of insufficient
magnitude to attract the attention of the larger organization. Therefore, in contrast to the
usual problem in experimentation of signal to noise ratios, we suggest that many business
initiatives suffer from a problem of signal-to-baseline. Existing business activities form
the baseline. If the current and near-term expectations of performance of the new
initiative are modest relative to this baseline, then the scarce resource of managerial
attention, as well as more conventional resources such as capital, will tend not to be
allocated to them.
The most basic managerial tool to modulate the signal to baseline ratio is organizational
structure. Not surprising, firms that have sustained high levels of innovation throughout
their history, such as Hewlett-Packard and Johnson & Johnson, have long-standing
commitments to narrow charters for their operating units. Even though many new
initiatives for these companies emerge from existing operating units, these initiatives
typically are ultimately pursued in the context of a new, dedicated operating unit.
• Be careful where you look for market insights: Because of the diversity of markets,
the lessons managers take away about the potential applications of new technology may,
in large part, depend upon where they look. As companies “probe and learn” about
20
markets, what they learn may be directly related to where they probe (Morone, 1993).
Generalizing market signals from a given niche to the broader market can lead to
dangerous distortions in expectations, which can lead to overly pessimistic assessments
of opportunities (e.g., hard disk manufacturers like Seagate and Quantum, by relying on
the early assessments of desktop computer users, missed enormous opportunities in
small, light, but low capacity disk drives (Christensen, 1997); similarly, incumbent
telecommunication equipment providers like Lucent and Siemens followed their lead
users and chose to limit their early involvement in packet based switching equipment).
Such false generalizations may also lead to overly optimistic assessments of
opportunities (e.g., satellite based mobile telephony operators like Iridium and
Globalstar who extrapolated the preferences of a small set of globe-trotting executives
to the broader market). Learning and adaptation are feedback-driven processes
(March and Simon, 1958; Nelson and Winter, 1982); as a result, decisions regarding
the sources of feedback have significant implications for learning and directing change.
• Learn by doing: Engage in exploration through exploitation. By engaging the market,
firms not only sell product and create revenues, they also gain information on market
size, preferences and requirements. Flexibility in market focus allows for a broad set of
alternative bases of market support, customer feedback, production experience, and
accompanying these, the increased capacity to learn and improve in subsequent
development attempts. Learning requires action. Arguably, in mature markets, firms can
learn about market preferences by observing consumer responses to other firms’
products. However, for emerging technologies that offer new functionalities and
functionality bundles, understanding consumer preferences strictly on the basis of this
kind of vicarious learning is less effective.
• Mainstreaming niche technologies: Given the pattern of evolution shown in Figure 3,
managers can look for opportunities to accelerate this evolution. One approach is to
21
target niches whose qualitative selection criteria overlap with the mass market’s but
whose absolute requirements are more accommodating along some dimension.
Selecting early niches on the basis of such preference overlap allows firms to transfer
experience acquired in the incubating niche to other market segments. Such overlap
characterized the early evolution of the United State’s semiconductor industry for whom
the early requirements of government sponsors, particularly the Department of Defense,
for smaller, lighter, and more reliable digital integrated circuits mapped well onto the
selection criteria employed in the emerging market for industrial computers and,
ultimately (in step with lower prices) to the mass market that was created by the
microprocessor. In Europe, by contrast, semiconductor firms were initially focused on
consumer electronics markets (stereos, televisions, and automobiles) and, to this end,
pursued analog integrated circuit technology which proved much more difficult to port to
other niches. Indeed, as the applicability of digital integrated circuits has expanded,
analog circuits have been displaced even from their home niches (Malerba, 1985).
In evaluating technology opportunities, managers can examine whether there are
relatively small technological changes or complementary technologies that will open the
technology to a whole new level of development, such as Web browsers that opened up
the latent opportunity of the already existing Internet protocols (Berners-Lee, 1999).
Indeed, just as Marconni’s reapplication of Hertz’s discoveries to wireless telegraphic
communication unleashed a wave of resources which served to accelerate the evolution
of wireless radio technology, the introduction of internet technology to the mass market
fueled a simultaneous boom in investment and innovation which would have been
impossible had the internet remained in the exclusive domain of government labs and
research institutions.
Clearly, the market feedback based approach suggested here raises important questions as
to the relationship between the sequencing of expenditures and the timing of feedback. An
22
investment process based on feedback is implicitly making assumptions about the level of initial
commitments relative to subsequent commitments and about the speed of feedback relative to the
pace at which these commitments must be made. To the extent that there are large fixed investments
that must be made prior to the realization of any market feedback, then the process suggested here
is not appropriate.
A further organizational challenge is raised by attempting a feedback-based approach to
technology management in the face of multiple application domains. In a heterogeneous demand
environment market rejection signals are not definitive – an innovation that is rejected in one
application domain may nonetheless find acceptance in another domain; an innovation that is
rejected at a certain state of development may find acceptance after further refinement. The open-
ended nature of this feedback can make abandoning projects quite difficult. Exploiting the inherent
flexibility of a sequential development process, however, requires that firms be able to exit
opportunity paths in a timely and efficient manner. Adopting efficient resource (re)allocation
processes is thus a critical component to implementing a feedback based approach.
Given these boundary conditions, one is left with the empirical question as to how
constraining the boundaries are. The development histories of such disparate innovations as
xerography (Dessauer, 1971; Pell, 1998), video recording (Rosenbloom and Freeze, 1985,
machine tools (Rosenberg, 1976), electric power utility (Hughes, 1983), speech recognition
technology (Forrester, 2000), and internet telephony (Mines and Delhagen, 1998) provide evidence
that development through sequential entry into markets is not an uncommon mode of technology
evolution.
23
While technologies may require a significant incubation period, the investments required
during the incubation stage are rather modest; the large scale investments associated with
technology ventures are related to scaling up for mass production and establishing distribution and
support systems (Rosenberg, 1976). The costs associated with mass production are indeed high,
but that does not negate the possibility and benefit of engaging intermediate markets before
committing to the pursuit of the mass market. The example of Sony, which produced its first
Betamax machines at a U-Matic plant which was producing video tape machines targeted at
corporate and institutional users, suggests that that even production capacity can be built
incrementally in parallel with activity in intermediate markets (Lyons, 1976).
The pattern of exploration through exploitation is borne out time and again in the evolution of
technology. At times it is a matter of deliberate strategy, in which firms actively direct development
through intermediate markets in their quest to reach their target market. This was the case in the
development of video technology, where a market for home video was suggested in 1956, twenty
years before a video machine was successfully introduced to the home market (Rosenbloom and
Cusumano, 1987). At other times, the market niche which propels an innovation to success
emerges through a relatively undirected stream of market experiments, such as Dupont’s discovery
of ballistic applications for Kevlar after failed initial attempts at serving tire and airframe
manufacturers (Rosenbloom and Hounshell, 1992).
Conclusions
Managerial interest in emerging technologies hinges on the promise these technologies hold
for their mature states. However, the path to technological maturity holds great uncertainty. As a
24
result, a primary challenge in managing technological emergence is how to structure development
activities before the full character of the technology and of its market relevance is established
The challenge of identifying applications in the early stage is driven not only by the limitations
in technology performance, but also by the fact that attention focused on a search for market
application is attention diverted from immediate development. Therefore, firms have incentives to
develop technologies in-house rather than attempt to exploit their possibilities in an elusive market.
This is, in a sense, the reverse of the more common criticism leveled at managers that a lack of
research focus is attributed to concern with the short term over the long term (Dertouzos, Lester,
and Solow, 1989). Early on, the market search process is likely to lead to blind alleys and, because
‘negative knowledge’ is not highly regarded, the outcomes of such search are not seen as being of
high value. As such, in the case of emerging technologies, short-term results are easier to show for
research and development activity than for market activity.
New technologies, like new genetic species, undergo periods of evolution and revolution.
They involve technological development and the transfer of the technology to new domains of
application. Beneath the revolutionary emergence of new technologies is often a process of shifting
application domains and rapid subsequent growth in the new domain. By understanding this
process, managers can better use it to their advantage.
25
Figure 1: Speciation in the Development of Technology
Shift in application domain
Lineage development
• Distinct selection
criteria
• Distinct resource
pool
Possible “invasion” of the
original niche or other niches
Antecedent application domain
Technology
development
26
Figure 2: Technological Convergence in CAT Scanning
Technology A
X-Ray Technology Technology B
Computer Technology
Application domain
Application domain
Medical Imaging
Data Processing
Convergent Technology
CAT Scanning
27
Figure 3: Technology Evolution and Penetration
of Application Domains by Video Recorders
Video Recording Technology
Application domain
Application domain
Application domain
Broadcasters
Industrial and Commercial
Mass Consumer Market
Product/Price Improvements
à
28
References
Abernathy, William. 1978. The productivity dilemma. Baltimore: Johns Hopkins University Press
Adner, Ron (2002). “When are technologies disruptive? A demand based view of the emergence of
competition.” forthcoming, Strategic Management Journal.
Adner, Ron and Daniel Levinthal (2000). “Technology speciation and the path of emerging
technologies.” in Wharton on Emerging Technologies, G. Day and P. Schoemaker (eds.) John Wiley
and Sons: New York, NY.
Adner, Ron and Daniel Levinthal. (2001a). “Demand heterogeneity and technology evolution:
Implications for product and process innovation”. Management Science, 47: 611-628.
Adner, Ron and Daniel Levinthal (2001b). “What is not a real option: Identifying boundaries for the
application of real options to business strategy.” Working paper.
Aitken, Hugh. (1985). The Continuous Wave: Technology and American Radio, 1900-1932.
Princeton University Press: Princeton, NJ.
Arthur, W. Brian. (1989). “Competing technologies, increasing returns, and lock-in by historical
events”. Economic Journal, 99: 116-131.
Basalla, George. (1988). The Evolution of Technology. Cambridge University Press:
Cambridge, England.
Berners-Lee, Tim. (1999). Weaving the Web. New York: HarperCollins.
Burgelman, Robert (1991) Intraorganizational Ecology of Strategy Making and Organizational
Adaptation: Theory and Field Research. Organization Science. August.
Bush, G. (1975), “Modes of animal speciation,” Annual Review of Ecological Systems, 6, 339-
364.
Carroll, Glenn (1985). “Concentration and specialization: Dynamics of niche width in populations of
organizations.” American Journal of Sociology, 90: 1262-83.
Carroll, Glenn and Michael Hannan (2000). The Demography of Corporations and Industries.
Princeton, NJ: Princeton University Press.
Christensen, Clayton and Richard Rosenbloom. (1995). “Explaining the attacker’s advantage:
Technological paradigms, organizational dynamics, and the value network”. Research Policy, 24:
233-257.
29
Christensen, Clayton M. 1997. The Innovator’s Dilemma. Boston, MA: Harvard Business School
Press.
D’Aveni, Richard. (1994). Hypercompetition. Free Press: New York, NY.
David, Paul. (1985). “Clio and the economics of QWERTY”. American Economic Review, 75:
Dessauer, John H. 1971. My years with Xerox; the billions nobody wanted. Garden City, NY:
Doubleday.
Dertouzos, M.L., Lester, R.K., and Solow, R.M. 1989. Made in America: Regaining the
Productive Edge. The MIT Commission on Industrial Productivity. Cambridge Mass: MIT Press.
Dosi, Giovanni. (1983). “Technological paradigms and technological trajectories”. Research
Policy, 11: 147-62.
Forrester (2000) The web’s speech impediments. The Forrester Report. April
Foster, Richard. (1986). Innovation: The Attacker’s Advantage. Summit Books: New York,
NY.
Garrett, G. (1994). The Early History of Radio. Institute of Electrical Engineers. London.
Gould, Stephen and Niles Eldredge. (1977). “Punctuated equilibria: The tempo and mode of
evolution reconsidered”. Paleobiology, 3: 115-151.
Hannan, Michael and John Freeman (1984). “Structural inertia and organizational change”.
American Sociological Review, 49: 149-64.
Hannan, Michael and John Freeman (1989). Organizational Ecology. Cambridge, Mass:
Harvard University Press.
Hippel, Eric von. 1988. The Sources of Innovation. New York: Oxford University Press.
Leonard-Barton, Dorothy. 1995. Wellsprings of knowledge : building and sustaining the sources of
innovation. Boston, Mass. : Harvard Business School Press.
Levinthal, Daniel A. (1998). “The slow pace of rapid technological change: Gradualism and
punctuation in technological change”. Industrial and Corporate Change, 7: 217-247.
Lyons, Nick. 1976. The Sony vision. New York : Crown Publishers.
30
Malerba, Franco (1985) “Demand structure and technological change: The case of the European
semiconductor industry”. Research Policy, 14: 283-297.
March, James G. and Simon, Herbert A. 1958. Organizations. New York: Wiley.
Mayr, Ernest (1963), Animal Species and Evolution. Harvard University Press: Cambridge,
MA.
McGahan, Anita, Leslie Vadasz, and David Yoffie. (1997). “Creating value and setting standards:
The lessons of consumer electronics for personal digital assistants”. In D. Yoffie (ed.), Competitng
in the Age of Digital Convergence. Harvard Business School Press: Boston, MA.
Mines, H. and Delhagen, K. (1998) “Internet phone service for consumers”. The Forrester Report.
Feb 13.
Moore, Geoffrey A. 1995. Inside the Tornado: Marketing strategies from silicon valley’s cutting
edge. Harper Business: NY.
Mokyr, Joel, (1990). “Punctuated equilibria and technological progress”. American Economic
Review, 80: 350-355.
Morone, Joseph G. 1993. Winning in high-tech markets: the role of general management : how
Motorola, Corning, and General Electric have built global leadership through technology. Boston,
Mass. : Harvard Business School Press.
Nelson, Richard and Winter, Sidney. 1982. An Evolutionary Theory of Economic Change.
Cambridge, MA: Harvard University Press.
Pell, Erik M. (1998). From Dream to Riches – The Story of Xerography, Carlson Circulating
Books, Rochester, NY.
Ristelhueber, Robert. (1993). “GaAS are making a comeback, but profits remain elusive”.
Electronic Business Buyer, 19: 27-28.
Rosenbloom, Richard and Michael Cusumano. (1987). “Technological pioneering and competitive
advantage: The birth of the VCR industry”. California Management Review, 51-76.
Rosenbloom, Richard S. and Freeze, Karen J. 1985. "Ampex Corporation and Video Innovation"
in Research on Technological Innovation, Management, and Policy, Richard Rosenbloom ed, Vol
2, pages 113-185, JAI Press.
Rosenbloom, Richard and David Hounshell. (1992) Du Pont Kevlar® Aramid Industrial Fiber.
Harvard Business School Case #9-391-146.
31
Rosenberg, Nathan. (1976). “Technological change in the machine tool industry, 1840-1910". In
N. Rosenberg (Ed.),Perspectives on Technology. M. E. Sharpe: London, England.
Sahal, Devendra. 1985. Technological guideposts and innovation avenues. Research Policy, 14, 61-
82.
Schumpeter, Joseph A. (1934). The Theory of Economic Development. Harvard University
Press: Cambridge, MA.
Sull, Donald, Richard Tedlow and Richard Rosenbloom. (1997) Managerial commitment and
technological change in the US tire industry. Industrial and Corporate Change.
Trajtenberg, Manuel. 1990. Economic analysis of product innovation : the case of CT scanners.
Cambridge, Mass. : Harvard University Press.
Tushman, Michael and Philip Anderson. (1986). “Technological discontinuities and organizational
environments”. Administrative Science Quarterly, 31: 439-465.
Wiegner, Kathleen. (1988). “Silicon Valley 1, Gallium Gulch 0". Forbes, 141: 270-272.
Yoffie, David. (1996). “Competing in the Age of Digitial Convergence”. California Management
Review, 38: 31-53.
