🌊10. The Dynamics of Design Evolution
Marcos Antonio de Lima Filho, PhD.
10.1 A Dynamic Model of Design Evolution
As noted in the Literature Review, the correlation between technology and evolutionary theory has long captivated academic interest. Evolution and economics, for example, have a long history of reciprocal influence dating back to Darwin’s time (Beinhocker, 2006). This cross-pollination extends its reach into innovation research. The notable contribution of Abernathy and Utterback reflects this evolutionary influence, as their A-U model incorporates a punctuated model of industrial evolution.
Innovation scholars built on the theory of punctuated equilibrium to propose explanations for the phenomenon of disruption (Sood & Tellis, 2011). Unfortunately, as discussed in Section 6.2, this research tradition experienced a decline in popularity during the late 1990s, when a paradigm shift occurred. Christensen’s new theory quietly detached disruption research from its roots in evolutionary theory and punctuated equilibrium, a shift explored in Section 8.3.
This study, in contrast, aims to re-establish the evolutionary roots within disruptive innovation research. Previous attempts to weave evolutionary theory into innovation research and design theory have largely treated evolution as a general metaphor. This research, however, goes beyond this approach. It advances the evolutionary analogy by linking the market selection of innovations to specific patterns of natural selection, a connection often overlooked. This analogy is substantiated by empirical data documenting the evolution of smartphones and commercial aircraft. In conclusion, the evolution of designs can be broken down into four primary categories of innovation: disruptive, directional, stabilising, and purifying innovation (Figure 10.1.1). Together, these categories shape a punctuated model of industrial evolution, thereby reintegrating and enriching the original insights of Abernathy and Utterback’s model.
The Uncertainty of Nascent Industries
To augment our comprehension of innovation, it is imperative to examine the systematic relationships among different innovation patterns, rather than focusing on a single pattern in isolation. Indeed, the pioneers of disruption research have always recognised that industrial evolution is a multi-patterned process. Abernathy and Utterback (1978) were among the first researchers to observe that technological evolution typically involves periods of extensive experimentation, followed by phases of consolidation and stability.
Abernathy and Utterback (1978) also observed that technologies do not emerge fully developed at the outset of their commercial lives. As a result, the rate of product innovation in a particular industry or product category is at its peak during its formative years. This “fluid phase” is characterised by a high rate of product innovation, paving the way for diverse technologies and design concepts to emerge. The early days of the automobile industry provide an excellent example of this fluid phase, where a plethora of automotive concepts, including electric and steam-driven cars, were produced by numerous manufacturers (Utterback, 1994). This uncertainty and learning behaviour is fairly typical in new industries of assembled products:
At this embryonic stage, no firm has a “lock” on the market. No one’s product is really perfected. No single firm has mastered the processes of manufacturing, or achieved unassailable control of the distribution channels. Customers have not yet developed their own sense of the ideal product design or what they want in terms of features or functions. The market and the industry are in a fluid stage of development. Everyone — producers and customers — is learning as they move along (Utterback, 1994, p. 23).
This fluid phase impacts design and innovative activity in several ways. The lack of a dominant design encourages intense technological competition among firms. New features and design concepts are continually introduced, tested, and either adopted or discarded, akin to the trial and error process that fuels the evolution of species.
In nascent industries, consumer preferences are not yet established and constantly changing, rendering incremental improvements insufficient to meet the market’s dynamic demands. Rather, this dynamic and uncertain environment fosters disruptive selection, leading companies to explore a vast array of diverse designs in a quest to discover the “fittest” solution. Consider, for example, the diversity of competing architectures that emerged during the early stages of the Jet Age (Figure 10.1.2).
As the aviation industry matured and technology advanced, a notable shift occurred in the architectural designs of commercial aircraft. The twin-engine design has gradually established itself as the dominant model across various types of aircraft, including turboprops, regional jets, narrow-body, and wide-body aircraft (as illustrated in Figure 10.1.3). This transition from diverse architectural options towards a single dominant design is also evident in the evolution of smartphones, as discussed in Section 6.4. In the early 2000s, the handset market was populated by different architectures, such as flip phones, sliders, and candy bar designs. However, following the iPhone’s introduction, the touchscreen slate became the most popular architecture due to its superior user experience.
The parallel with evolution becomes evident, as both industries underwent a process of testing numerous architectures against their respective environments until one emerged as dominant. The less-efficient and poorly adapted designs, unable to compete in markets, gradually became obsolete.
Therefore, in emerging industries, the market is particularly open to product innovations, both of disruptive and purifying nature. Similar to the role of mutations in natural selection, disruptive innovations infuse new “genes” into a “product’s DNA”, which can lead to the emergence of new dominant designs. Following this, a process analogous to purifying selection takes place, weeding out the “deleterious mutations” from markets — In this context, the poorly adapted designs that no longer meet the industry’s evolving demands.
The model presented in Figure 10.1.1 bridges the fluid stage of industrial evolution with disruptive innovation. This linkage is supported by historical evidence, which indicates that as technologies mature, the window for exploration tends to narrow. As product designs stabilise, industries often transition from a state of high variability and experimentation to one of consolidation and refinement, limiting the scope for radical and disruptive innovation.
The Emergence of a Dominant Design
The disruptive era lasts until a dominant design emerges. The dominant design usually takes the form of a new product or set of features (Utterback & Suárez, 1993). With this the focus of innovation shifts from meeting emerging needs with new concepts, to refining, improving and strengthening the dominant design and its appeal in the market (Abernathy & Clark, 1985). This process of consolidation establishes order for future progress and changes the character of innovation, making it less disruptive and more incremental and directional.
The results of this research highlight the relevance and applicability of the older disruption paradigm, firmly rooted in the tradition established by Abernathy and Utterback. Their model, characterised by a punctuated model of industrial evolution, provides a generalisable and parsimonious explanation for the emergence of dominant designs and their influence on innovation.
Abernathy and Utterback introduced the concept of a dominant product design, defining it as a product configuration that achieves a high degree of market acceptance. This dominance sets a standard that affects production processes and subsequent innovation efforts. As Utterback (1994) explains:
A dominant design has the effect of enforcing or encouraging standardisation so that production or other complementary economies can be sought and perfected. Effective competition then shifts from innovative approaches to product design and features, to competition based on cost and scale as well as on product performance (Utterback, 1994, p. 32).
The emergence of a dominant design represents a milestone or transition point in the life of an industry (Suárez & Utterback, 1995). This shift initiates a period of increasing stability or “equilibrium” as defined by the punctuated equilibrium model. This directional phase (Figure 10.1.1) is characterised by a change in innovation strategy. Instead of pursuing novel design possibilities, the focus shifts towards refining and perfecting the established design. This represents a transition from generating new design qualities to enhancing existing quantities.
The shift towards “new quantities” can be seen in the introduction of incremental innovations — subtle modifications and improvements aimed at refining the dominant design, increasing its efficiency, and reducing its cost. They typically do not revolutionise the product’s basic architecture or introduce completely new technological foundations. This consolidation, in turn, enables companies to “shift their focus from innovative products to larger-scale production of standardised offerings” (Utterback, 1994, p. 83).
In summary, the emergence of a dominant design ushers in a new phase of industrial evolution, marked by a shift from disruptive to directional/incremental innovation, increased emphasis on productivity and process improvement, and opportunities for market expansion.
The A-U model was first proposed in the late 1970s by Utterback and Abernathy (1975), and subsequently expanded upon by Abernathy and Utterback (1978). Despite its age, the A-U model retains its pertinence in explaining the evolutionary trajectories of a wide array of industries, including commercial aviation and smartphones. The A-U model had a profound influence on the development of innovation research in the 1980s and 1990s. In common, these studies recognised industrial evolution as a punctuated process, with periods of incremental progress disrupted by intervals of rapid, discontinuous change (Chesbrough, 2001).
The Abernathy-Utterback (A-U) model’s relevance is demonstrated in its ability to shed light on the evolution of industries as diverse as commercial aviation and smartphones. When this model was presented, the commercial aviation industry was characterised by a rich variety of competing designs, with no clear dominant design on the horizon. Various aircraft configurations, including twin-turbofan, trijet, and quad-turbofan, were competing for market dominance. The introduction of the supersonic Concorde in 1976 added a further layer of diversity to the technological landscape (Figure 10.1.2).
Before the eventual consolidation of the aviation industry was evident, Abernathy and Utterback had already theorised about the likely emergence of a dominant design during periods of technological diversity. They postulated:
When a product is first introduced, diversity of product technology is fairly easy to spot, for differences among the several extant versions of the product will often reflect differences in core concepts. (…) But after design hierarchies have crystallised, it becomes increasingly difficult to observe technical diversity (Abernathy et al., 1983, p. 104).
Indeed, as the industry matured, a particular design started to become prominent. By the 1980s, twin-engine designs began to rise as the dominant architecture across all aircraft types, including turboprops, regional jets, narrow-bodies, and wide-bodies (Figure 10.1.3).
Abernathy and Utterback were three decades away from witnessing the rise of touchscreen smartphones (Figure 6.4.1) and the co-dominance of Google’s and Apple’s operating systems (Figure 6.5.1). However, their A-U model accurately predicted the trajectory that this industry would take. The model forecasts that industries evolve from an era of fragmentation and instability, characterised by diverse products and design concepts, to a stage of consolidation and stability, featuring predominantly undifferentiated, commodity-like products (Utterback, 1994, p. 91). This progression closely mirrors the trajectory of most consumer electronics segments.
The A-U model also foresees that industries “will be populated by fewer firms commanding larger shares of the market than was true before the dominant design crystallised” (Abernathy et al., 1983, p. 132). When this model was proposed, the Airbus-Boeing duopoly was far from materialisation, as Airbus had not even delivered its first A320. Key players in the industry were numerous and varied, with firms such as McDonnell Douglas, Lockheed Martin, Fokker, Convair, and Douglas Aircraft offering unique aircraft models. However, as predicted by the A-U model, this industry would gradually transform “from a variety of firms with unique products to an oligopoly of firms with similar products” (Utterback, 1994). As the twin-engine design emerged as the industry standard, these traditional manufacturers succumbed to the dominance of the Airbus-Boeing duopoly.
Overall, the original A-U model continues to generate predictions that are independent of markets, industries, and time. As a final example, consider the following: “the emergence of a dominant design tends to be associated with overall market growth and the appearance of new customers” (Abernathy et al., 1983, p. 132). The evolution of the smartphone industry is a literal replication of this 40-year-old statement. The emergence of touchscreen smartphones led to a massive expansion of this market, represented by a thirteen-fold increase in global smartphone shipments (Figure 10.1.4).
In conclusion, the dominant design paradigm, as proposed by Abernathy and Utterback, provides not just a description of a phenomena, but also a robust explanation for the evolution of industries. The model’s theoretical parsimony facilitates its generalisation across different contexts and industries, an aspect that Christensen’s disruption paradigm falls short of. By presenting a punctuated model of industrial evolution, the A-U model reinstates the foundational evolutionary principles that originally anchored disruptive innovation research.
Maturity and Decline
As a dominant design emerges, the basis of competition shifts to refinements in product features, reliability, and cost (Utterback, 1994). This involves a transition from low to higher volume production, and from less to more emphasis on cost (Clark, 1985). Eventually, technological diversity gives way to technological stasis, which directly affects design activity:
Particular design approaches achieve dominance, production volumes increase, and performance criteria and processes are more clearly specified. The transition to a “specific” stage of development entails a change in the nature of innovation. In contrast to the fundamental changes introduced in the “fluid” phase, innovation in the “specific” stage is likely to alter only a small aspect of the basic product, and any changes introduced serve to refine the established design (Clark, 1985).
When a product reaches its mature stage of evolution, technological change is often all but invisible (Abernathy et al., 1983). This period of relative technological stasis alters the competitive balance in favour of companies with stronger process innovation expertise. When this happens, many firms are unable to compete and effectively fail (Utterback, 1994).
Hence, even though the emergence of a dominant design tends to be associated with overall market growth and the appearance of new customers, it is also associated with higher levels of industry concentration (Abernathy et al., 1983, p. 132). These industries become extremely focused on cost, volume, and capacity; product and process innovation appears in small, incremental steps (Utterback, 1994). Thus, once an industry reaches maturity, it cannot regain the competitive characteristics of its youth (Abernathy et al., 1983).
These cycles of development, growth, and maturity can eventually result in decline. The length of these phases can differ depending on the industry and the pace of technological advancements in that area. For instance, digital cameras and iPods experienced short periods of stability before rapidly declining due to the emergence of smartphones, which combined the capabilities of both devices into a single product (Figure 10.1.5). In contrast, the automobile sector has demonstrated a longer period of technological stability. The last century was marked by incremental improvements in safety, performance, and comfort, but the core concept of a petrol-powered, human-controlled vehicle has endured.
To conclude, this dynamic model of design evolution underscores that disruptive innovations do not exist in isolation. Rather, the evolution of industries is an intricate dance of adoption, adaptation, and ultimately, extinction of previous ideas and creations. Historically, there has been an overemphasis on the genesis and adoption of new technologies while neglecting the equally critical phenomena of their discard and extinction (Basalla, 1988). Such bias has led to an oversight of the reasons and processes through which societies divest of once-useful technologies.
Therefore, the study of disruptive innovations should not neglect their subsequent selection, replication, and eventual extinction. As noted long ago by material-culture theorist George Kubler, invention, replication, and discard are of equal importance in reaching a better understanding of the made world and how it evolves (Kubler, 1962/2008). Invention breaks stale routine, replication makes the invention widely available, and discard assures that there will be room for newly invented things in the future (Basalla, 1998). Hence, the discard phase (purifying innovation) is as much a part of evolution as disruption and replication, freeing up resources from obsolete technologies and opening space for new paradigms.
Last updated
