Key Principle
Customers and investors effectively control a firm's resource allocation, not its executives. The systems that well-managed companies develop to kill ideas their customers do not want -- rigorous planning processes, ROI screens, customer validation gates -- are the same systems that kill disruptive projects. The trap is self-reinforcing: good practices (listening to customers, investing in high-return projects, analyzing market trends) succeed with sustaining technologies and systematically fail with disruptive ones.
Why This Matters
Resource allocation is the real strategy of any organization. What projects get funded, what engineers get assigned, what salespeople pursue -- these decisions, not executive speeches or strategic plans, determine what a company actually does. In well-managed firms, this allocation process is optimized to serve existing customers and pursue high-margin opportunities. This optimization is precisely what makes disruptive response structurally impossible from within the mainstream organization.
Without understanding resource dependence as a structural force, firms misdiagnose disruption failures as leadership failures and try to fix them by hiring visionary executives. But the actual bottleneck is the incentive structure governing which proposals survive to reach any executive at all. The resource allocation machine operates correctly -- it is solving the wrong problem.
The trap is self-reinforcing in a specific way: the usual remedies -- planning better, working harder, becoming more customer-driven, total quality management, process reengineering -- all deepen the trap because they optimize for the sustaining context. Disruptive products promise lower margins, serve insignificant or nonexistent markets, and are rejected by a firm's most profitable customers. Standard investment processes screen them out correctly by their own logic. This is what makes the dilemma structural rather than a failure of vision.
Good Examples
The six-step decision pattern in disk drives. Engineers at established firms built working prototypes of disruptive drives before entrants did (Seagate had roughly 80 3.5-inch prototypes). Marketing showed prototypes to existing customers, who rejected them. Resource allocation processes redirected investment to sustaining projects. Frustrated engineers left and founded startups. The firm was rationally pulled along a trajectory that proved fatal. At each step, the rational decision from within the firm's value network led deeper into the trap. (Chapter 2)
Minimill disruption in steel. Minimills entered at rebar -- the lowest quality, lowest margin tier. Integrated mills were "almost relieved" to cede it because rebar contributed the least to their bottom line. Each successive retreat (bars, rods, structural beams, sheet steel) was individually rational: managers allocated resources to higher-margin products. But collectively, they ceded the entire industry from the bottom up. Bethlehem Steel invested $1.3B in R&D and plant improvement, all directed at conventional steelmaking -- the resource allocation system working exactly as designed. (Chapter 4)
The mid-level filtering mechanism. The real gatekeeper of resource allocation is not the executive who signs off on investments but the mid-level employee who decides which proposals reach the executive's desk. Proposing a high-margin sustaining project is career-safe; proposing a low-margin disruptive project with no proven market is career-dangerous. Disruptive opportunities are killed before senior leadership ever sees them -- not by deliberate strategy but by the rational self-interest of the people who assemble the pipeline. (Book Group Guide)
Counterpoints
The CEO override problem. Even when a CEO personally champions a disruptive project, the hundreds of daily resource allocation micro-decisions made by employees throughout the organization collectively redirect effort back toward mainstream customers. A major disk drive CEO shepherded four generations of 1.8-inch drives; none sold. Honda was buying 1.8-inch drives from a Colorado startup. The CEO's salespeople had no incentive to pursue an $80M market when quotas were tied to the multibillion-dollar computer industry. (Chapter 4)
Career asymmetry as hidden filter. Projects that fail because the market was not there are career-ending for the champion. Projects that fail technically are forgiven as part of R&D's probabilistic nature. This asymmetry systematically filters out disruptive proposals before they reach senior decision-makers. The kill mechanism is not a conscious decision but a structural feature of how careers are evaluated. (Chapter 4)
Seagate's flash memory withdrawal. Seagate and Quantum both built competitive flash memory products -- proving the capability existed. But flash created value in different networks (palmtop computers, cameras, cash registers) while resources were absorbed defending mainstream disk drive positions. Both firms withdrew with under 1% market share by 1995. The resource allocation system chose correctly within the value network and catastrophically from outside it. (Chapter 2)
Key Quotes
"Sustaining projects addressing the needs of the firms' most powerful customers almost always preempted resources from disruptive technologies with small markets and poorly defined customer needs." -- Clayton M. Christensen, Chapter 2
"It is very difficult for a manager to motivate competent people to energetically and persistently pursue a course of action that they think makes no sense." -- Clayton M. Christensen, Chapter 4
"There are times at which it is right not to listen to customers, right to invest in developing lower-performance products that promise lower margins, and right to aggressively pursue small, rather than substantial, markets." -- Clayton M. Christensen, Introduction
"The most formidable barrier the established firms faced is that they did not want to do this." -- Clayton M. Christensen, Chapter 2
Rules of Thumb
- Treat resource allocation as the real strategy. What gets funded is what gets done, regardless of what the strategic plan says.
- Do not assume executive vision can override a resource allocation system optimized for sustaining innovation. The system's hundreds of daily micro-decisions will redirect effort back toward mainstream customers.
- Watch for middle-management filtering: if disruptive proposals never reach the executive level, the problem is structural incentives, not individual cowardice.
- When your best customers reject a new technology, that rejection is evidence for disruption potential, not evidence against investment.
- The firm's "well-developed systems for killing ideas customers don't want" will kill disruptive projects with equal efficiency. The kill system does not distinguish between bad ideas and disruptive ideas.
- Cost structure lock-in drives asymmetric mobility: firms move upmarket easily (higher margins, bigger customers) but cannot move downmarket. The most deadly attacks come from the direction where the least profit appears to be. (Chapter 4)
- Customer co-migration is invisible: 8-inch drive makers did not notice their northeasterly drift because their customers (DEC, Prime, Wang) were all moving toward mainframe territory simultaneously. (Chapter 4)
Related References
- Creating Independent Organizations for Disruption - Independent organizations are the structural countermeasure to resource dependence, placing disruptive projects outside the mainstream allocation system
- Unknowable Markets and Discovery-Driven Planning - Resource allocation systems demand the quantified forecasts that cannot be produced for disruptive markets
- Performance Oversupply and the Basis-of-Competition Shift - Resource allocation optimizes for the current basis of competition, making firms blind to the approaching shift