6.2 Issues with Disruption Theory

Marcos Antonio de Lima Filho, PhD.

The publication of The Innovator’s Dilemma in 1997 marked a pivotal moment in the study of disruptive innovations. In this book, Christensen diverged significantly from previous research, giving rise to a new school of thought. Unlike his predecessors, who conceptualised disruption in a broader, more generalisable, and evolutionary sense, Christensen proposed a surgical definition of disruption. The concept was then narrowed to a set of very specific conditions, premises, and analytical tools.

In short, Christensen defines that an innovation becomes disruptive when its performance trajectory crosses the mainstream market’s performance threshold. A crossing signals whether a new technology is robust and productive enough to supplant the old one (Sood & Tellis, 2005). Then, at this point in time and competitive space—the intersection of the two performance trajectories—disruption occurs (Christensen et al., 2018). The analysis of performance trajectories is one of the major innovations that Christensen brought to the theory of disruption. Yet, as discussed in this section, the introduction of this cartesian instrument brought in a whole set of difficulties.

Before Christensen, the phenomenon of disruptive innovation could be explained by various but complementary frameworks, including dominant designs (Abernathy & Utterback, 1978), innovation transilience (Abernathy & Clark, 1985), technological discontinuities (Tushman & Anderson, 1986), architectural innovations (Henderson & Clark, 1990), and technology s-curves (Christensen, 1992b). These authors inherited much from the works of William J. Abernathy. His original contribution consisted of recognising that innovation is not a uniform phenomenon: “Some innovations disrupt, destroy and make obsolete established competence; others refine and improve” (Abernathy & Clark, 1985).

The concept of disruption was a prevalent theme among these theories. In fact, the overlap was so substantial that Christensen initially characterised his study of the disk drive industry as an example of architectural innovation (1992a). However, he later described it in terms of technology S-curves (Christensen, 1992b, 1992c), disruptive technologies (Christensen, 1993), and technological discontinuities (Rosenbloom & Christensen, 1994).

The release of The Innovator’s Dilemma, however, disrupted the continuity of this research tradition, creating a gap between theory and practice. Since then, disruptive innovations have been characterised in terms of intersecting trajectories of performance improvement. Disruption was also subject to a set of very strict conditions: disruptors should either establish a foothold among consumers at the bottom end of the market, or create new markets by targeting non-consumers (Figure 6.2.1).

Unlike previous scholars, Christensen introduced a “very specific definition of disruptiveness” (Christensen & Raynor, 2003). Tellis (2006) summarised Christensen’s thesis into five main premises, several of which can be further reduced into sub-premises, which are shown as letters in the following points:

1. A new disruptive technology initially underperforms the dominant one along the dimensions mainstream customers in major markets have historically valued.

2. But the disruptive technology (a) has other features a few fringe (and generally new) customers value. Products based on disruptive technologies are typically (b) cheaper, (c) simpler, (d) smaller, or (e) more convenient than those established on the dominant technology.

3. (a) The leading firms’ most profitable customers generally do not want and indeed initially cannot use products based on disruptive technologies. So (b) disruptive technologies are first commercialised in emerging or insignificant markets. (c) Incumbents conclude that investing in disruptive technologies is not a rational financial decision for them.

4. The new disruptive technology (a) steadily improves in performance until (b) it meets the standards of performance demanded by the mainstream market.

5. At that point, (a) the new (disruptive) technology displaces the dominant one and (b) the new entrant displaces the dominant incumbent(s) in the mainstream market.

As such, what Christensen meant by disruption was a very specific phenomenon and business strategy. This new paradigm of disruption became focused on “a specific type of technological change, which operates through a specific mechanism, and has specific consequences” (Danneels, 2004). Unlike his forerunners, “Christensen used it to mean something very specific” (Gans, 2016). The Innovator’s Dilemma established a very rigid and specific model for disruption. With such developments, Christensen claims to have evolved disruption into “a more broadly explanatory causal theory of innovation and competitive response” (Christensen et al., 2018), albeit at the sacrifice of simplicity.

From ancient to modern times, theologians, philosophers, and scientists have always regarded simplicity as a theoretical virtue. Aristotle wrote in Posterior Analytics, “We may assume the superiority ceteris paribus of the demonstration which derives from fewer postulates or hypotheses” (Baker, 2022). For Kant,“rudiments or principles must not be unnecessarily multiplied”. That is, other things being equal, simpler theories are better (Baker, 2022). “The grand aim of all science”, notes Einstein, “is to cover the greatest possible number of empirical facts by logical deductions from the smallest possible number of hypotheses or axioms” (Baker, 2022).

Therefore, given competing and equally plausible definitions of disruptive innovation, researchers should favour the “simpler” one (Hopp et al., 2018). It is certain that for any set of data there will always be more than one possible explanation, although there may be one that stands out based on its parsimony or potential explanatory power (Bryant, 2017). This does not seem to be the case for disruptive innovation theory, as I discuss in this chapter.

By focusing solely on “low-end disruptions”, Christensen ignored other discontinuous patterns of change, which may be of equal or greater importance (Utterback & Acee, 2005). For decades, scholars have advocated for a broader definition of disruption (Acee, 2001; Govindarajan & Kopalle, 2006; Markides, 2006; Sood & Tellis, 2011). One of the main reasons is that disruptive innovations can be offered initially at a higher price than existing products but still pose the ‘‘innovator’s dilemma’’ for incumbents (Govindarajan & Kopalle, 2006). Had Christensen adhered to the principles of simplicity and parsimony, he might have characterised disruptive innovation as a phenomenon that could emerge in low, mid, or high-end market segments.

However, instead of theoretical parsimony and integration with prior studies, The Innovator’s Dilemma contributed to fragment disruption research, and the effects persist to this day. The new theory was so radical that it effectively disrupted the research tradition on which it was founded. Since the publication of The Innovator’s Dilemma in 1997, Christensen’s concept of disruption has emerged as the “dominant” paradigm of disruption (Figure 6.2.2).

Indeed, Christensen established a new school of thought on the issue of disruptive innovation. Disruption entered business jargon and became a cultural phenomenon, possibly contributing to the demise of previous research traditions. Since the early 2000s, Google Scholar citations of disruptive innovation terminology in academic literature have far surpassed the citations of previous research traditions. However, Christensen evaluated that most references to disruptive terminology consist of pro-forma references, indicating an “underuse of disruptive innovation as a coherent theoretical concept” (Christensen, 2018). Danneels warned about this mismatch between theory and practice decades ago, and attributed it to the complexity of Christensen’s argument: “One can see from a search for disruptive technology on the web how loosely the term has come to be used and how it has become separated from its theoretical basis” (Danneels, 2004).

Another problematic development is Christensen’s conflation of business model strategies with performance trajectories. This departs from previous research in multiple ways, as disruptiveness was not tied to any particular analytical tool, competitive strategy, or business model. Christensen (2006) explained this decision, justifying that disruption was modelled as a relative phenomenon for predictive purposes. Accordingly, “no innovation is inherently disruptive”; thus “disruptive innovations must be evaluated relative to a firm’s business model” (Christensen et al., 2018).However, a large group of scholars manifested uneasiness with this logic (Acee, 2001; Danneels, 2004; Sood & Tellis, 2005; Tellis, 2006; Markides, 2006). For them, defining disruptive innovation as a disruptive business model that causes market disruptions appears tautological. To avoid such circularity, academics should define the types of technologies or innovations independent of their effects on a market (Sood & Tellis, 2005).

Scholars have also tried for decades to change Christensen’s stance on high-end disruptions. A high-end disruption is one in which a higher-performing and higher-priced innovation is introduced into leading established market segments and later moves towards the mass market (Utterback & Acee, 2005). It is a familiar strategy deployed by Apple, Boeing, and many other companies. Yet, these cases conflict with Christensen’s theory of disruption:

When initially introduced, disruptive innovations are inferior to incumbent products on accepted performance dimensions, but they offer a novel mix of attributes that appeals to fringe customer groups, notably those near the bottom of the market (Christensen et al., 2018).

Despite calls for change, until the very end of his career, the author insisted that “the notion of high-end disruption is incompatible with several important premises of the theory of disruptive innovation” (Christensen et al., 2018). However, even a cursory tour of discontinuous innovations reveals instances that defy Christensen’s premises:

These include consumer goods such as the compact disc, tools such as the electronic calculator, components such as the fuel injector and commodities such as wafer board and oriented strand board construction panels. (…) None of the examples cited above, at least initially, was a case in which cost and traditional performance were lower (…) The electronic calculator provides an almost purely opposed case in which cost was much greater but performance was also greatly enhanced (Utterback & Acee, 2005).

Critics argue that while the traits identified by Christensen may be common in his studies, they do not necessarily apply to all disruptive innovations (Danneels, 2004). Michael Porter, who is widely regarded as the father of modern business strategy, agrees that disruptions can also originate from high-end segments: “We don’t have good evidence on which form is most prevalent, but both exist” (Magretta, 2012). If disruptions can originate from high-end segments, then the assumption of low-end upstarts being necessary for disruption is flawed, and the concept of disruptive innovation should not be limited to this definition. By expanding the definition to encompass all market segments, we can characterise this phenomenon with enhanced parsimony and conceptual validity. As William of Ockham (1287-1347) said, “Entities must not be multiplied beyond necessity”. This principle of parsimony aligns with Newton’s first principle of reasoning, which states that “we are to admit no more causes of natural things than such as are both true and sufficient to explain their appearances”.

The Innovator’s Dilemma was highly successful in raising awareness within the management community about the phenomenon of disruption. The analysis of performance trajectories and the emphasis on low-end technologies led Christensen to model this phenomenon in a rather unique approach, which diverged significantly from prior research. This new model, along with the extensive editorial acclaim it received, played a key role in establishing a new paradigm for disruptive innovation.

Paradigms, as characterised by Thomas Kuhn, are “universally recognised scientific achievements that, for a time, provide model problems and solutions to a community of practitioners” (Kuhn, 1962/1996). However, every paradigm has its limitations. When a scientist adheres to a theoretical paradigm, a set of intellectual blinders prevents him or her from seeing certain aspects of reality (Rogers, 2003). Thus, as with any field of research, Christensen’s disruptive paradigm has its blind spots. They necessarily accompany a dominant paradigm (Rogers, 2003). Furthermore, Khun emphasises the risk of knowledge loss when a paradigm begins to dominate over competing paradigms:

(...) shifts in paradigms can take place even when the new theory is worse than the old. In short, knowledge can be and is lost as well as gained, and science often proceeds in a zig-zag rather than linear manner. We might add that this would be particularly true in the social or humane sciences. As a result, paradigms and basic truths get lost (...) The years may well bring retrogression as well as progress (Rothbard, 2006, p. 438).

This implies that knowledge can be lost, particularly when “the newer paradigm does not include parts of the older paradigm that were well established” (Marcum, 2015, p. 180). Unfortunately, it seems that this situation is applicable to the field of disruptive innovation research.

As seen in Figure 6.2.2, Christensen’s disruption theory emerged as the dominant paradigm of disruption in the early 2000s. In 2021 alone, scholars released 14,000 articles referencing disruptive terminology, compared to only 2,017 articles referencing prior research traditions. Together, these earlier frameworks account for 13% of the total, while Christensen’s disruption theory dominates the research landscape with an 87% share.

Such dominance may have contributed to the dissolution of previous research traditions, such as dominant designs, architectural innovation, and technological discontinuities. These alternative perspectives clashed with the new dominant paradigm, and despite the quality of their arguments, these works lacked the editorial backing necessary to spread ideas among practitioners. Rebecca Henderson, for instance, has always had a significant influence inside academic circles; nonetheless, she has never had a public reputation as a guru, maybe because she has not authored popular management treatises (Gans, 2016). Indeed, Christensen’s PhD dissertation draws significantly on Henderson and Clark’s (1990) study of dominant designs and architectural innovation:

This thesis relies extensively upon Henderson & Clark’s definitions, and provides rather unambiguous support for their predictions that established firms will be successful at points of change in component-level technologies (Christensen, 1992a, p. 19).

However, in later works, Christensen abandoned the concept of dominant designs and architectural innovation. These concepts were commonly discussed within the research tradition upon which disruption was founded. These ideas can be traced back to Abernathy and Utterback’s pioneering effort to link innovation to a model of industrial evolution (Abernathy & Utterback, 1978; Abernathy et al., 1983). The studies of this era incorporated a punctuated equilibrium model, in which periods of relative stability are interrupted by periods of discontinuous change. This equilibrium could be disrupted by the emergence of a new dominant design (Abernathy & Utterback, 1978), which would create the conditions for a technology-based reversal in the process of industrial maturity (Abernathy et al., 1983). This process would “disrupt, destroy, and make obsolete established competence” (Abernathy & Clark, 1985), requiring “fundamentally new skills and competences” (Tushman & Anderson, 1986). The new dominant design would establish a new set of core design concepts embodied in a new architecture (Henderson & Clark, 1990).

In prior studies, the direction of causality focused on the emergence of the dominant design, which shifted the basis of industry competition from product-based to process-based competition (Chesbrough, 2001). Back then, disruptive innovation was heavily influenced by Gould’s theory of punctuated equilibrium, which posited that evolutionary change occurs in sudden bursts rather than through gradual, incremental progress. This theory resonated with Christensen, who noted the similarities between Gould’s concept and the processes of technological change observed in the disk drive industry: “This process seems remarkably parallel to the processes of technological change observed in the disc drive industry” (Christensen, 1992a, p. 260).

Despite relying “extensively upon” this research tradition, Christensen broke with this link in subsequent works. Christensen and Bower (1996) introduced a novel causal mechanism for the failure of leading firms. Christensen argued that managers are doing an excellent job satisfying their high-margin customers in their existing business but are myopic about the threats posed by low-end customers or new markets (King & Baatartogtokh, 2015). Since this change, disruption has become a resource allocation problem, partly based on Bower’s 1970 book Managing the Resource Allocation Process. The Innovator’s Dilemma also abandons the punctuated model of industrial evolution and the concept of dominant designs. Christensen replaced them with a new disruptive model based on the analysis of performance trajectories, as well as a specific business strategy meant to solve the resource allocation problem (the “dilemma”).

As a result, The Innovator’s Dilemma was more than just a simple terminological shift that could be reconciled with previous studies; it represented a true revolution in the field. Christensen reinterpreted disruption by introducing a new definition and causal mechanism, both of which were incompatible with earlier research. These substantial departures, coupled with the popularity and dominance of his theory in popular press, may have contributed to the decline of the previous research tradition. Consequently, it is reasonable to assume that some explanatory power was lost during this transition.

Knowledge loss can occur as a result of paradigm shifts, particularly when successful explanatory accounts from the predecessor theory are not preserved in the successor theory, as Kuhn frequently emphasised. To mitigate this potential loss of valuable insights, researchers and practitioners must remain mindful of the contributions from earlier paradigms and strive to integrate diverse perspectives into their understanding of disruptive innovation.

Indeed, as an indication of this loss, the current disruption theory struggles to account for major historical events, such as the Jet Age. Within the new framework, the jet engine has been reclassified as “a radical but sustaining innovation” (Christensen & Raynor, 2003, p. 69). This stands in contrast to the interpretations of earlier scholars, who frequently cited the transition from piston to jet aircraft as a prime example of disruption. However, Christensen was unable to identify the same resource allocation process within the Jet Age that would align with his disruption model. The rise of Boeing and Airbus was founded on superior, rather than inferior, technologies, which contradicts the premise of low-end market entry. That is, in his pursuit of defining the phenomenon more precisely, Christensen inadvertently excluded some technological revolutions from the disruptive category. This “downgrade” signifies a Kuhnian “loss of explanatory power”, as prior research was able to readily acknowledge and explain the disruptive nature of such events.

Kuhn also showed that an accumulation of anomalies often precedes a paradigm shift, but only after a period of intense struggle or crisis (Marcum, 2015). Though scientists may begin to lose faith and then to consider alternative approaches, they do not renounce the paradigm that has led them into crisis (Kuhn, 1962/1996). It is quite the contrary:

(…) increasingly desperate attempts are made to modify the particulars of the basic theory so as to fit the unpleasant facts and to preserve the framework provided by the paradigm. Only when anomalies pile up to such an extent that the paradigm itself is brought into question do we have a “crisis situation” in science. And even here, the paradigm is never simply discarded until it can be replaced by a new, competing paradigm which appears to close the loopholes and liquidate the anomalies (Rothbard, 1971).

Hence, the accumulation of anomalies eventually leads to a crisis, during which a research community realises that these anomalies indicate a serious problem with the prevailing paradigm (Marcum, 2015). In the context of disruptive innovation research, Christensen has been addressing such anomalies since at least 2006. He claims that disruption theory has been improved “by making definitions and measures more precise, by refining the categories, and by clarifying the mechanism of causality” (Christensen, 2006). Since then, “the process of anomaly-seeking research has extended and refined the theory” (Christensen et al., 2018).

Despite these extensions and refinements, major theoretical difficulties remain, and the gap between theory and practice has become even wider over the past decades. Consider, for instance, how Christensen’s disruption model has led its own creator to make a series of inaccurate predictions:

• When asked about Apple’s first smartphone, Christensen stated that “the iPhone is a sustaining technology (…) the prediction of the theory would be that Apple won’t succeed with the iPhone” (McGregor, 2007). Contrary to this prediction, Apple’s stock price has since increased by over 5,500%, and the iPhone has become a defining product for its company and the entire industry.

• Christensen advised that regional jet manufacturers would continue to move upmarket and eventually become a threat to Boeing and Airbus: “The theories of innovation suggest that Boeing and Airbus will continue to miss much of the industry's real growth” (Christensen et al., 2004, p. 133). Ironically, regional jet sales began to decline soon after this prediction was made. In the years that followed, Bombardier’s share price dropped by as much as 80%, while Boeing’s increased by as much as 800%.

The number of anomalies might have been reduced if Christensen had incorporated the contributions of Danneels (2004), Sood and Tellis (2005), Utterback and Acee (2005), Markides (2006), and others. Instead, his subsequent extensions and refinements have primarily served to reinforce his paradigm’s framework. And despite the growing number of anomalies and academic objections to the theory’s fundamental assumptions, the 2016 edition of The Innovator’s Dilemma claims that “the theory of disruption continues to yield predictions that are quite accurate, in an astounding range of industries” (Christensen, 2016). One of his co-authors has even suggested that Christensen’s model could be used to predict stock performance: “There are some data to suggest that Disruption can be used successfully in this way” (Raynor, 2011).

Finally, the way most practitioners comprehend disruption is an anomaly in itself. Christensen’s analytical tools, premises, and specific conditions constitute a highly sophisticated system of thought. Christensen (2006) argues that these contributions have “greatly improved our understanding of the causal mechanism and the precision of the categorisation scheme.” In reality, however, such complexity may have encouraged practitioners to embrace the theory’s terminology while overlooking its fundamental principles. Despite becoming part of the business lexicon, Christensen acknowledges that “the theoryʼs core concepts have been widely misunderstood and its basic tenets frequently misapplied” (Christensen et al., 2015). This problem has persisted for decades:

(…) many people have equated our use of the term sustaining innovation with their preexisting frame of “incremental” innovation, and they have equated the term disruptive technology with the words radical, breakthrough, out-of-the-box, or different. (…) We regret that this happens, because our findings relate to a very specific definition of disruptiveness (Christensen & Raynor, 2003).

Christensen and his research collaborators have made numerous attempts to clarify his original thesis via research papers, management publications, interviews, conferences, and even a think tank. The Clayton Christensen Institute was established in 2007 to promote the theories of the Harvard professor. Despite these efforts, “over the past two decades, the theory of disruption has been interpreted and misapplied so broadly as to mean anything that is clever, new, and ambitious” (Christensen et al., 2016).

In addition to this gap between theory and practice, there is also a growing divide within academia itself. In 2018, Christensen and co-authors conducted a review of studies on disruptive innovation published in academic journals between 1993 and 2016. The authors identified a “decrease in direct engagement with disruption theory arguments”, an “overly broad application of the terms disruption/disruptive innovation”, an “indiscriminate use of its terminology”, and “frequent pro-forma references”, all of which indicate an “underuse of disruptive innovation as a coherent theoretical concept” (Christensen et al., 2018). Google Scholar data can confirm these conclusions (Figure 6.2.3).

This enduring mismatch between theory and practice, as well as within academia, indicates that it may be time to reevaluate the theoretical foundation of disruption. Christensen’s disruption theory has been a subject of debate for decades, with critiques appearing in top journals related to management and technological innovation. This prestigious group includes the Strategic Management Journal (Adner, 2002), Journal of Marketing (Sood & Tellis, 2005), Marketing Science (Sood & Tellis, 2011), International Journal of Innovation Management (Utterback & Acee, 2005), and MIT Sloan Management Review (King & Baatartogtokh, 2015). In 2006, The Journal of Product Innovation Management devoted an entire issue to the topic. Unfortunately, this constructive influx did not lead to any significant revisions.

In conclusion, this section has highlighted several issues with disruption theory, revealing a growing mismatch between theory and practice, as well as inconsistencies and debates within academia. While Christensen’s theory has generated substantial interest, it has also faced mounting critiques and a growing accumulation of anomalies. Despite these challenges, Christensen’s orthodoxy has persisted, bolstered by modest theoretical advancements that have not adequately addressed the concerns raised by critics. Moving forward, it is essential for scholars and practitioners alike to critically evaluate the theoretical foundation of disruption and engage in open dialogue to refine, build upon, or even challenge the existing framework. By doing so, the academic community can foster a more cohesive understanding of disruptive innovation, better inform practice, and ultimately contribute to the evolution of disruptive innovation theory.