Disruption: Why Incumbents Fail and Challengers Win
Clayton Christensen's disruption theory explains why well-managed, customer-focused incumbents consistently lose to new entrants — and what both sides should do about it.
The Innovator's Dilemma
In 1997, Harvard Business School professor Clayton Christensen published "The Innovator's Dilemma," one of the most influential business books ever written. Its central puzzle: why do well-managed companies, focused on satisfying their best customers, fail to respond to new entrants who appear to offer an inferior product?
The companies that failed weren't badly managed. They surveyed their customers, who said they wanted better performance on existing dimensions. They invested accordingly. They maintained profitability. They did everything that good management practice suggested — and they lost.
Understanding disruption theory is essential for any entrepreneur looking to challenge an incumbent and for any executive at an established company trying to defend their position.
Two Types of Disruption
Christensen distinguishes two patterns of disruptive innovation:
Low-end disruption: A new entrant targets the least profitable customers of an incumbent — the customers the incumbent is happy to ignore because they're not particularly demanding and don't pay much. The entrant offers a "good enough" product at dramatically lower cost. Over time, the entrant improves its offering and moves up-market, eventually competing for the incumbent's core customers.
The classic example: steel mini-mills versus integrated steel producers in the 1970s and 1980s. Integrated mills produced high-quality steel for demanding applications (auto bodies, appliances) but had high overhead and couldn't profitably serve the lowest quality tiers. Mini-mills entered with electric arc furnaces, producing lower-quality steel (rebar — concrete reinforcing bar) at dramatically lower cost. Integrated mills didn't fight for rebar — margins were terrible and they had better uses for their capacity. Mini-mills won the rebar market completely. Then they improved their processes and moved up to structural steel, then bar, rod, and wire, and eventually began competing in flat-rolled steel. Each time, integrated mills retreated upmarket rather than defend low-margin segments. Eventually, the mini-mills had consumed virtually the entire commodity steel market.
New-market disruption: A new entrant doesn't compete with incumbents at all initially — it creates a market that didn't exist, converting non-consumers into consumers. The personal computer is the archetype. Minicomputers and mainframes were sold to corporations and research institutions. PCs were sold to individuals who couldn't afford or use a minicomputer — they were non-consumers of computing. The PC initially wasn't a threat to minicomputer companies, because PC buyers weren't minicomputer buyers. Over time, PCs became powerful enough to replace minicomputers for most applications, and minicomputer companies (DEC, Wang, Data General) effectively disappeared.
Why Incumbents Fail to Respond
The rational incumbent's response to a disruptor is often to do nothing — and this is frequently correct in the short run. The disruptor's product is inferior on the metrics the incumbent's customers care about. The market is small. The margins are low. Defending the low end would cannibalize profitable current business.
Christensen identifies three factors that make rational short-term decisions accumulate into strategic failure:
Resource dependence: Established companies allocate resources based on what current customers want and where current margins are best. Investment proposals targeting the disruptive new market fail the internal test — customers don't want it (their current best customers), margins are poor, market size is small. The company can't get resources allocated to disruption even when leadership intellectually understands the threat.
Processes: Established companies' processes are optimized for the current business model. The scale economics, quality systems, sales processes, and customer relationships that make the incumbent excellent at serving current customers are precisely wrong for serving the new market the disruptor is opening.
Values: What a company is willing to pursue shifts over time toward higher-margin, larger-market opportunities. A company that was once willing to pursue a $50M market opportunity requires $500M market opportunities to get management attention after it has grown. The disruptive market is too small to matter — until it's too large to ignore.
Jobs-to-Be-Done Theory
Christensen's later work developed the Jobs-to-Be-Done (JTBD) framework, which provides a different but complementary lens on disruption. The core idea: customers don't buy products; they "hire" products to do specific jobs in their lives. Understanding what job a product is hired to do — and what alternatives compete for that job — reveals disruption opportunities that traditional competitive analysis misses.
The milkshake example: a fast food chain wanted to improve milkshake sales. Customer research (what do you want in a milkshake?) was unhelpful. But when Christensen's team studied when and why people bought milkshakes, a clear pattern emerged: morning commuters bought milkshakes to make their commute more interesting and keep them satisfied until lunch. They were hiring the milkshake for the job of "entertainment + satiety during a boring commute." Their real competitors were bagels (messy, not satisfying long enough), bananas (too quick), and donuts (crumbly). A thicker, slower-to-drink, more interesting milkshake served the commuter job better.
The JTBD lens changes how you see competition. Netflix didn't just disrupt Blockbuster by offering the same movie-rental service more conveniently. Netflix changed the job it was hired to do from "get a specific movie" to "find something good to watch." That job was also competed for by books, other streaming services, video games, and social media. Understanding the job explains both who you're competing with and what product improvements will matter.
Nokia dominated mobile phones in the 2000s with 40% global market share. Their phones were excellent at making calls and sending texts — the core customer need they were focused on. The iPhone was, by Nokia's analysis, an inferior phone: shorter battery life, higher price, no tactile keyboard, limited cellular network support. Nokia's dismissal was rational based on their definition of the job. But the iPhone was hired for a different job — "a computer that fits in your pocket" — and it did that job dramatically better than Nokia's phones. Nokia's customers, surveyed about what they wanted from phones, didn't ask for the iPhone because they didn't know it was possible. By the time Nokia recognized the threat and tried to respond, the market had been defined in new terms where Nokia had no advantage. From 40% market share in 2007, Nokia's smartphone business was essentially gone by 2013.
Disruption Defense for Incumbents
Incumbents are not helpless against disruption, but defending effectively requires deliberate choices that cut against the normal management grain.
Create an independent organization: Christensen's prescription for incumbents is to create a fully separate unit to pursue the disruptive opportunity — with different processes, different cost structures, different performance metrics, and enough organizational independence to operate by the logic of the new market rather than the old one.
IBM created a separate unit in Boca Raton to develop the original PC — separate from its mainframe and minicomputer businesses, with different processes, different suppliers, and different metrics. The PC was technically and commercially successful. But IBM then failed to maintain proprietary control of the operating system (Microsoft's DOS) and the microprocessor architecture (Intel), ultimately ceding the value to suppliers.
Acquire the disruptor: Sometimes the right move is to buy the disruptive company before it grows into an existential threat. Google's acquisition of Android. Facebook's acquisition of Instagram. These were preemptive strategic acquisitions of potential disruptors.
Platform leadership: Becoming the platform on which disruptors build can turn disruption into an opportunity. Salesforce's AppExchange lets disruptive applications build on Salesforce's platform rather than competing with it. App stores allow Apple and Google to capture value from the very developers disrupting specific market categories.
The Disruptor's Strategy
For the entrepreneur seeking to disrupt an incumbent, Christensen's theory suggests a clear strategic path:
- Identify an overserved market or non-consumption: Look for customers the incumbent ignores (too unprofitable) or people who can't access the incumbent's product at all (non-consumers)
- Offer a simpler, more affordable, more accessible solution: Don't try to match incumbent performance immediately — win on different dimensions
- Improve relentlessly: Use the insights and economics from the initial market to improve the product and move up-market incrementally
- Avoid competing with the incumbent directly: Fight on the incumbent's terms only after you have the scale and capabilities to win
- Nokia dismissed the iPhone as an inferior phone — and by Nokia's definition of "phone," they were right. Apply the Jobs-to-Be-Done lens to an industry you know well: what job are customers actually hiring the incumbent's product to do, and where might a disruptor reframe that job entirely?
- Christensen's framework predicts that well-managed incumbents rationally ignore early disruptors. But Google, Amazon, and Netflix have all successfully disrupted adjacent markets from positions of incumbency. What makes these exceptions possible — and does it undermine the theory or clarify its boundaries?
- The advice to incumbents facing disruption is to create an independent organizational unit. But IBM's PC unit succeeded technically and then failed strategically (it ceded DOS to Microsoft and the chip architecture to Intel). At what point does organizational independence become organizational naivety — and how should incumbents think about the boundaries of separation?
- Disruption theory was developed primarily from industrial and technology case studies in the 1980s-90s. Does it apply with equal force to platform businesses, where network effects and data advantages may make incumbents harder to dislodge than Christensen's model predicts?