Resources
Management of technology: how technologies and the firms behind them actually evolve, compete, and win, each idea tied to the literature and, where we have one, a case.
The economics of transitions page asks who bears a cost, who captures a value, and what a market will price. The human side of technology page asks what happens when a real person meets the technology in a real place. This page asks a different and complementary question: how does a technology itself evolve, how do firms compete to shape and own it, and what decides whether an innovation displaces the incumbent or dies in the attempt? These are the questions of the management-of-technology and innovation-studies literatures, and they sit underneath every transition the Lab studies.
The distinction is worth stating plainly. Economics is the lens of allocation and welfare: prices, incidence, market failure, finance. Innovation dynamics is the lens of evolution and strategy: how a technology matures, how a dominant design emerges, how firms build the capabilities to ride or resist change, how an innovation diffuses through a population. The two overlap, path dependence and network effects live in both, but they answer different questions, and keeping them separate makes each sharper.
Each concept is defined, anchored to its canonical source, and, where the Lab has written about it, linked to the applied case. Full references are at the end.
A technology improves slowly at first, then rapidly as effort compounds, then slowly again as it approaches physical limits. Performance plotted against effort or time traces an S. Foster (1986) made the S-curve a tool of technology strategy: the incumbent riding a maturing curve is most vulnerable exactly when its technology looks most refined, because a new technology on a younger, steeper curve is about to overtake it. Reading which curve a technology sits on, and how much room it has left, is the first question of technology management.
Early in a technology's life, many rival designs compete; then the market converges on one configuration that becomes the reference every competitor must adopt. Utterback and Abernathy (1975); Abernathy and Utterback (1978) named this the dominant design, and showed that its emergence flips the basis of competition from product innovation to process and cost. Before the dominant design, the game is invention; after it, the game is efficiency. Much of a transition's turbulence is the fight to set the dominant design, because whoever sets it shapes the market that forms around it.
The Abernathy-Utterback model divides an industry's life into phases: a fluid phase of rapid, uncertain product experimentation; a transitional phase as a dominant design emerges; and a specific phase of incremental, cost-focused process innovation. Knowing which phase a technology is in tells you what kind of competitor wins, the agile experimenter early, the efficient scaler late, and warns when the rules are about to change.
Innovation ranges from incremental improvements that reinforce existing competences to radical breaks that render them obsolete. The distinction matters because incumbents are structurally good at the first and structurally bad at the second: their skills, suppliers, and customers all pull toward the familiar. Most failed transitions are radical innovations judged by incremental standards.
An innovation can leave every component unchanged and still destroy an incumbent, if it changes the way the components are linked. Henderson and Clark (1990) showed that this architectural innovation is peculiarly deadly because incumbents, whose knowledge is organised around the old architecture, often fail even to perceive it as a threat. It looks like a minor rearrangement and is in fact a redefinition. Many transitions are architectural: the pieces are familiar, the system is new.
The trajectory a technology takes is shaped by the small, early, sometimes accidental choices that got locked in, so history, not just merit, determines which design prevails. This is the innovation-dynamics reading of Arthur (1989) and David (1985) (treated as an economic result on the economics page): here the emphasis is strategic, that an early lead in a path-dependent technology is worth fighting for out of proportion to its immediate value, because the path, once set, is hard to leave.
Technical progress does not explore every possible direction; it follows a paradigm, a shared sense of what the relevant problems are and how to solve them, and advances along the trajectories that paradigm defines. Dosi (1982) drew the analogy to Kuhn's scientific paradigms: a paradigm acts as a lens that focuses effort on some avenues and makes others nearly invisible, which is why breakthroughs so often come from outside the incumbent's field of view. For a transition, the paradigm is the deep reason an industry keeps improving the wrong thing, refining the combustion engine while the problem quietly becomes the powertrain, and why genuine change usually requires a paradigm shift, not just harder work along the existing trajectory.
Some technologies, steam, electricity, the computer, are not single innovations but platforms that reshape every sector they touch, and their gains show up slowly, because the economy has to reorganise around them first. Bresnahan and Trajtenberg (1995) named these general purpose technologies; David (1990) showed, through the example of the electric dynamo, that the productivity payoff can lag the invention by decades while factories, skills, and institutions are rebuilt around it. The lesson for a transition is patience with a twist: the absence of measured gains in the early years is not evidence the technology has failed, it is evidence the complementary reorganisation has not happened yet.
An innovation that is worse on the metrics incumbents value, but better on a new dimension (cheaper, simpler, more accessible), can take root in a low-end or new market the incumbent ignores, then improve until it displaces them from above. Christensen (1997) formalised this, and the key insight is counter-intuitive: incumbents fail not because they are badly managed but because they are well managed, listening to their best customers is exactly what makes them miss the disruption. The Lab reads this pattern in emerging-market technology, where the "worse" product that serves the underserved is often the one that scales.
Not every threat is disruptive. Sustaining innovations, even dramatic ones, improve a product for existing customers, and incumbents almost always win those. Disruptive innovations change who the customer is and what "good" means. Misclassifying a threat, treating a disruption as a sustaining challenge, or panicking over a sustaining one, is a common and costly strategic error.
When a technology's value to each user rises with the number of other users, competition tends toward a single winner, and the contest becomes a standards battle fought over installed base and compatibility rather than product quality. Katz and Shapiro (1985) formalised network externalities (treated on the economics page as a market-structure result); in innovation strategy the lesson is that timing, sponsorship, and compatibility choices can matter more than being best. We examine who controls a standard, and therefore the market around it, in Who Holds the Pen on the Standard.
New combinations do not add to an economy so much as replace what came before, so innovation and disruption are one process seen from two sides. Schumpeter (1942) named the "perennial gale of creative destruction." For technology management it is the founding fact: competitive advantage is temporary by construction, and a firm's task is not to defend a position but to keep generating the next one.
Moving first can secure a lead through learning, network effects, and pre-emption, or it can mean bearing the cost of educating a market that a fast follower then captures. Whether the pioneer or the follower wins depends on the appropriability regime and the pace of change, which is why "first" is a strategy, not a guarantee.
A firm's own innovation can be excellent and still fail commercially if the other innovations it depends on, the complements, the infrastructure, the partners upstream and downstream, are not ready. Adner (2006) called this the "wide lens": the innovator's real risk is often not execution but the readiness of the ecosystem around it, and the more partners a promising technology needs, the more ways it can be held up by the slowest of them. Electric mobility is the textbook case: a good vehicle is not enough if the charging, the financing, the mechanics, and the grid are not ready together. Reading the whole ecosystem, not just the product, is what separates a technology that could work from one that will.
Inventing something valuable and capturing the value from it are different problems, and the second is usually harder. Teece (1986) showed that whether the innovator profits, rather than an imitator or a supplier, depends on the strength of the appropriability regime and on who controls the complementary assets: the manufacturing, distribution, brand, and qualification needed to bring the innovation to market. This is the single most useful frame in technology strategy, and it explains why the inventor is so often not the one who gets rich. We apply the make-or-buy version to network architecture in Own the Battery, Rent the Shopfront.
In a fast-changing environment, what matters is not a firm's current resources but its ability to sense change, seize opportunities, and reconfigure itself in response. Teece, Pisano, and Shuen (1997) called this bundle of higher-order abilities dynamic capabilities, and it is why some firms survive successive transitions while others, equally strong at any single moment, do not. A transition is precisely the environment where dynamic capability, not static advantage, decides who is left standing.
A firm's ability to recognise the value of new external knowledge, assimilate it, and apply it depends on how much related knowledge it already has, so the capacity to learn is itself cumulative. Cohen and Levinthal (1990) named this absorptive capacity, and it explains why a country or firm cannot simply buy its way into a frontier technology: without the prior base, the new knowledge cannot be absorbed. It is the deep reason capability is slow to build, explored in the field in Capability Is the Slow Part.
Which parts of a technology a firm develops in-house and which it sources is a strategic choice about where its distinctive capability should sit, not only a transaction-cost calculation. Coase (1937) and Williamson (1985) gave the economics (see the economics page); in technology strategy the added consideration is that outsourcing a component can hollow out the capability a firm will need for the next generation. We read this in battery-swap architecture in Own the Battery, Rent the Shopfront and in what an incumbent carries into a new regime in The Incumbent's Second Life.
To sell into a demanding buyer's supply chain, a producer's output must be qualified, certified as fit, and qualification, not the technology or the capital, is often the true gate to a market. This is where appropriability meets the field: the complementary asset the innovator most often lacks is acceptance by the buyer. Drawn out for battery-precursor material in The Lock-In Runs Both Ways and for export policy in The Ban Is Not the Policy.
When a transition arrives, the incumbent is not always the loser. A firm that carries transferable capability, distribution, relationships, or process skill into the new regime can win it, which is why some petrol-era players thrive in electric mobility. Reading what actually transfers, and what does not, is the strategic question. Set out in The Incumbent's Second Life.
How a product is divided into components shapes how the industry that makes it is divided into firms. Baldwin and Clark (2000) showed that modularity, splitting a system into parts that connect through standard interfaces, lets different firms innovate on different modules independently, which speeds progress but also lets the value migrate to whoever controls the parts that matter. The related mirroring hypothesis holds that the structure of an organisation tends to mirror the architecture of the product it builds. For a transition this is decisive: modularising a technology (battery, motor, controller, chassis as separable parts) opens it to new entrants and reshuffles who captures the value, which is often the real contest behind an apparently technical standards choice.
Firms once did their research behind closed doors and commercialised only what they invented; increasingly, valuable knowledge flows across the firm's boundary in both directions, and the task is to manage those flows rather than seal them. Chesbrough (2003) called this open innovation. It matters for transitions because the knowledge a niche needs is usually scattered across many actors, universities, suppliers, users, rivals, and the players who assemble it fastest, rather than the ones who guard their own, tend to set the pace. Closed strategies that worked in a stable regime often fail in the open, fast-moving early phase of a transition.
An innovation spreads through a population not all at once but in a predictable sequence, innovators, early adopters, early and late majority, laggards, each group persuaded by different things. Rogers (1962) mapped this, and the practical lesson is that the tactics that win the early adopters are not the ones that win the majority; many innovations stall precisely at that handover. Reading where an innovation sits in its diffusion curve tells you who to convince next, and how. The adopter-behaviour reading of the same curve, why the enthusiast is not representative of the majority, is on the human side of technology page.
The gap between the visionaries who will tolerate a rough product and the pragmatists who will not is where many technologies die. The named form, Moore's chasm, has been the language of technology marketing for three decades. The early market and the mainstream market want different things, and a product tuned to the first can fail to cross to the second. In a transition, this chasm is often the real barrier, not the technology and not the price, but the leap from the enthusiasts to the ordinary user.
Some technologies get more valuable the more they are adopted: each user improves the infrastructure, deepens the skill base, and raises the option's credibility for the next user. Arthur (1989) formalised this (see the economics page for the economic treatment); in innovation dynamics it is the engine that lets a niche cross the chasm and eventually overtake an incumbent, and the reason the early going is so slow.
Real adoption is rarely a clean replacement. Users add the new technology alongside the old, keep the incumbent as backup, and shift gradually, so a diffusion curve hides a messier reality of stacking and partial use. We read this pattern in African solar in Stacking, Not Switching.
Once a technology is proven, further pilots do not advance it; the binding constraint moves from demonstrating that it works to scaling it, and scaling is a different discipline from proving. Mistaking the two keeps a technology stuck in a permanent pilot phase. Argued for European agriculture in Europe Has Enough Demonstrations.
The innovation-dynamics literature above studies technologies and firms. The transitions literature scales the same questions up to whole systems, energy, mobility, food, where the "technology" is a regime of infrastructure, institutions, and behaviour, and the "firm" is a society. The bridge is worth naming, because the Lab works at exactly this join.
Transitions are read across three levels: protected niches where novelty incubates, the incumbent regime that resists it, and a broad landscape of prices, politics, and crises that can shake the regime loose. Geels (2002), building on Rip and Kemp (1998), gave this the multi-level perspective, now the dominant frame in transition studies. It is the systems-level version of the niche-versus-incumbent contest that innovation dynamics studies at the firm level.
New technologies need protected spaces to mature before they can face the regime, and those spaces can be deliberately created and managed, through pilots, subsidies, and shielded markets, to build the networks, expectations, and learning a niche needs. Kemp, Schot, and Hoogma (1998) called this strategic niche management. It is the transitions-scale version of nurturing an innovation through its fragile early phase.
A transition can unfold in several shapes depending on the timing of landscape pressure and the readiness of the niche: substitution, transformation, reconfiguration, or de-alignment and re-alignment. Geels and Schot (2007) set out these pathways. Reading which pathway a transition is on changes what a niche should do, and the Lab's own four-ways-a-transition-lands diagnostic is a field-oriented cousin of this map.
The Lab's own Bypass, Repurpose, Weaken framework sits in this tradition. Where the multi-level perspective describes the levels and the pathways describe the shapes, BRW classifies the mechanism by which a specific niche strategy engages a specific regime barrier, the operational, field-testable layer beneath the broader theory. It is innovation dynamics made diagnostic.
The concepts above draw on the following canonical sources in the management-of-technology, innovation-studies, and socio-technical-transitions literatures.
Abernathy, W. J., and Utterback, J. M. (1978). Patterns of industrial innovation. Technology Review, 80(7), 40 to 47.
Adner, R. (2006). Match your innovation strategy to your innovation ecosystem. Harvard Business Review, 84(4), 98 to 107.
Arthur, W. B. (1989). Competing technologies, increasing returns, and lock-in by historical events. The Economic Journal, 99(394), 116 to 131.
Baldwin, C. Y., and Clark, K. B. (2000). Design Rules, Volume 1: The Power of Modularity. Cambridge, MA: MIT Press.
Bresnahan, T. F., and Trajtenberg, M. (1995). General purpose technologies: 'Engines of growth'? Journal of Econometrics, 65(1), 83 to 108.
Chesbrough, H. W. (2003). Open Innovation: The New Imperative for Creating and Profiting from Technology. Boston: Harvard Business School Press.
Christensen, C. M. (1997). The Innovator's Dilemma: When New Technologies Cause Great Firms to Fail. Boston: Harvard Business School Press.
Coase, R. H. (1937). The nature of the firm. Economica, 4(16), 386 to 405.
Cohen, W. M., and Levinthal, D. A. (1990). Absorptive capacity: A new perspective on learning and innovation. Administrative Science Quarterly, 35(1), 128 to 152.
David, P. A. (1985). Clio and the economics of QWERTY. American Economic Review, 75(2), 332 to 337.
David, P. A. (1990). The dynamo and the computer: An historical perspective on the modern productivity paradox. American Economic Review, 80(2), 355 to 361.
Dosi, G. (1982). Technological paradigms and technological trajectories. Research Policy, 11(3), 147 to 162.
Foster, R. N. (1986). Innovation: The Attacker's Advantage. New York: Summit Books.
Geels, F. W. (2002). Technological transitions as evolutionary reconfiguration processes: A multi-level perspective and a case-study. Research Policy, 31(8-9), 1257 to 1274.
Geels, F. W., and Schot, J. (2007). Typology of sociotechnical transition pathways. Research Policy, 36(3), 399 to 417.
Henderson, R. M., and Clark, K. B. (1990). Architectural innovation: The reconfiguration of existing product technologies and the failure of established firms. Administrative Science Quarterly, 35(1), 9 to 30.
Katz, M. L., and Shapiro, C. (1985). Network externalities, competition, and compatibility. American Economic Review, 75(3), 424 to 440.
Kemp, R., Schot, J., and Hoogma, R. (1998). Regime shifts to sustainability through processes of niche formation: The approach of strategic niche management. Technology Analysis & Strategic Management, 10(2), 175 to 198.
Rip, A., and Kemp, R. (1998). Technological change. In S. Rayner and E. L. Malone (Eds.), Human Choice and Climate Change (Vol. 2, pp. 327 to 399). Columbus: Battelle Press.
Rogers, E. M. (1962). Diffusion of Innovations. New York: Free Press.
Schumpeter, J. A. (1942). Capitalism, Socialism and Democracy. New York: Harper & Brothers.
Teece, D. J. (1986). Profiting from technological innovation: Implications for integration, collaboration, licensing and public policy. Research Policy, 15(6), 285 to 305.
Teece, D. J., Pisano, G., and Shuen, A. (1997). Dynamic capabilities and strategic management. Strategic Management Journal, 18(7), 509 to 533.
Utterback, J. M., and Abernathy, W. J. (1975). A dynamic model of process and product innovation. Omega, 3(6), 639 to 656.
Williamson, O. E. (1985). The Economic Institutions of Capitalism. New York: Free Press.
Read this page alongside the economics of transitions, which takes the allocation-and-welfare view of the same terrain, and the transitions primer, which sets out the socio-technical frame in full. A handful of foundational ideas, path dependence, network effects, increasing returns, appropriability, deliberately appear on more than one page, because they genuinely belong to more than one literature; each page treats them through its own lens and points to the others.
Innovation dynamics is where the Lab's technical literacy and its social-science method meet: understanding how a technology evolves is what lets us read, in the field, whether a specific deployment is riding its curve or about to be overtaken, setting the dominant design or chasing someone else's, building the capability to last or hollowing it out. For how we apply this to a real case, see What We Do.