Innovation & product
Disruptive Innovation
A theory of why well-run incumbents lose to entrants with simpler, cheaper offers: serving their best customers pulls incumbents upmarket past what mainstream buyers need, leaving a foothold from which an entrant improves until the mainstream defects.
Also known as Disruption theory, Christensen's theory of disruption. First set out by Clayton M. Christensen, with Joseph L. Bower in 1995; the primary source is cited in full below.
- Format
- Structural model
- Level
- Corporate · Business unit
- Best for
- Analyse the environment · Position against competitors · Assess risk
- Decision stage
- Diagnose · Explore options
- Difficulty
- Intermediate
- Time to apply
- Half a day to classify a specific threat rigorously; the trajectory then needs monitoring over quarters, not meetings.
Plate · The model
The components
Sustaining trajectory overshoots
Incumbents improve their products along dimensions their most demanding customers value, and do so faster than mainstream customers can absorb. Performance oversupply turns yesterday's differentiation into unpriced surplus.
Signals of strength
Mainstream customers use a shrinking fraction of the product's capability · Price premiums for new releases are eroding · Buying decisions shift from performance to price and convenience · Sales teams struggle to articulate why the upgrade matters
Entrant takes a low-end or new-market foothold
A newcomer offers something worse on traditional metrics but simpler, cheaper or more convenient, serving overserved customers or nonconsumers. Incumbents can see it and still rationally ignore it, because its margins and customers look unattractive.
Signals of strength
The entrant's product is dismissed internally as a toy · Its customers are ones the incumbent is not sorry to lose · Its business model would be margin-dilutive if adopted by the incumbent · It is growing among people who previously bought nothing
Incumbents retreat upmarket
Faced with low-margin competition below, incumbents move toward their most profitable, most demanding customers. Each step up is locally rational and reported as strategic focus, and each concedes territory the entrant then occupies.
Signals of strength
Exit from entry-level products framed as portfolio rationalisation · Average margins rising while unit share falls · Investment cases for down-market defence repeatedly failing internal hurdle rates · Best people assigned to the highest-tier accounts
Disruption completes as the entrant improves
The entrant, improving along its own trajectory, eventually meets the performance mainstream customers require while keeping its cost advantage. Mainstream buyers defect, and the incumbent has run out of market to retreat into.
Signals of strength
The entrant wins head-to-head evaluations it used to be excluded from · Incumbent churn spreads from the bottom tier to the middle · The entrant begins raising prices without losing momentum · Incumbent responses shift from dismissal to imitation
When it earns its keep
- A cheaper, simpler competitor has appeared at the bottom of your market or among people you do not currently serve, and you need to judge whether it is a nuisance or a trajectory.
- You are the entrant, and you want to test whether your foothold has the structural properties that make incumbent retaliation unlikely rather than merely delayed.
- Your own product roadmap keeps adding capability your mainstream customers barely use, and you want to check whether you have overshot and opened the door beneath you.
- You are allocating innovation investment and need to distinguish sustaining bets, which incumbents usually win, from disruptive ones, which they usually lose.
And when it doesn't
- The new competitor entered at the top of the market with a better, dearer product. That is a sustaining attack, and the theory explicitly predicts a different contest; Christensen's own 2015 restatement rules Uber out of the definition on these grounds.
- You need a full competitive analysis of an industry. Disruption theory explains one failure mechanism; pair it with Five Forces or Seven Powers for the structural picture.
- You are using 'disruptive' as a synonym for 'new and successful'. The theory makes a specific prediction about footholds and incumbent incentives; stretched beyond that, it predicts nothing.
- The buying decision is dominated by regulation or long-term contracts that freeze the trajectory. The mechanism needs customers who are free to defect as the entrant improves.
How to run it
Before starting, gather the inputs the analysis depends on:
- A performance trajectory for your product against the performance mainstream customers can actually absorb, ideally with evidence of features going unused or price premiums shrinking.
- A map of overserved segments and of nonconsumers, the people priced or skilled out of the market entirely.
- The entrant's cost structure and business model, and whether its margins would look unattractive inside the incumbent's profit model.
- An honest account of your own resource-allocation process: which customers and margins get investment approved, and which proposals die.
- 1
Classify the innovation
Ask whether the new offer improves performance on the dimensions mainstream customers already value (sustaining) or trades away that performance for simplicity, convenience or price (disruptive). Most attacks on incumbents are sustaining, and incumbents usually win those. The classification does real work; do it first.
- 2
Identify the foothold
A disruptive entrant starts in one of two places: a low-end foothold among customers the incumbents overserve, or a new-market foothold among nonconsumers. Name the foothold precisely. If you cannot find one, the theory says you are looking at something else.
- 3
Plot the two trajectories
Draw the incumbent's rate of improvement against the rate at which customer needs grow. Overshoot happens because companies improve faster than customers can absorb. Then plot the entrant's improvement path and estimate where it intersects mainstream requirements.
- 4
Examine the incumbent's incentives
The heart of the theory is asymmetric motivation, rooted in Bower's work on resource allocation. Ask whether defending the foothold would mean the incumbent accepting lower margins and disappointing its best customers. If retreat upmarket is the profit-maximising response each quarter, expect the incumbent to retreat.
- 5
Watch the upmarket migration
Track whether the entrant is actually improving toward mainstream needs and whether the incumbent is ceding tiers of the market while reporting healthy margins on what remains. Rising average margins during share loss is the classic tell of an incumbent being disrupted comfortably.
- 6
Decide the response
Incumbents that survive disruption typically do it through a separate unit with its own profit model, free to cannibalise the core, as Christensen prescribed. Entrants should protect the foothold economics rather than accepting a flattering invitation to compete upmarket too early. Either way, the analysis must end in a resourcing decision.
Reading the result
A judgement about whether a given innovation is disruptive in the theory's strict sense, which foothold it occupies, how the incumbent's own incentives will shape its response, and what the trajectory implies for investment, defence or entry timing.
- The classification carries the prediction. Sustaining attacks favour incumbents; genuine disruptive footholds favour entrants, but only over the time it takes the trajectory to reach mainstream needs.
- Read incumbent behaviour through resource allocation rather than rhetoric. What gets funded reveals whether the retreat upmarket has already begun.
- Disruption is a process rather than an event. An entrant that has stopped improving toward mainstream needs is a niche player, however cheap it is.
A worked example
A high-street opticians chain assesses online prescription eyewear
A UK opticians chain with roughly sixty branches watches online retailers sell prescription glasses at a third of its prices, using uploaded prescriptions, home try-on kits and virtual fitting tools. Branch revenue depends on bundling eye tests with high-margin frame and lens sales. The board asks whether this is disruption in the strict sense, and what it implies.
- Sustaining trajectory overshoots
- Present. The chain has moved steadily toward designer frames, premium lens coatings and branch refits aimed at its most affluent customers. For a mainstream buyer replacing single-vision glasses, much of that capability is unpriced surplus, and the price gap to online is now large enough to overcome inertia.
- Entrant takes a low-end or new-market foothold
- Both footholds are visible. Low-end: price-sensitive customers with stable, simple prescriptions who resent paying for the branch experience. New market: people who deferred replacing glasses at all at high-street prices, and buyers of second and third pairs that the chain never sold. Online margins on these sales would dilute branch economics, so ignoring them has felt rational.
- Incumbents retreat upmarket
- Early signs in the chain's own plans: the draft strategy emphasises premium eyewear, audiology add-ons and flagship branches, and quietly deprioritises the value range. Each move is defensible on branch profitability, and each concedes the simple-prescription volume that funds branch overheads.
- Disruption completes as the entrant improves
- Partially blocked, and this matters. Online fitting is improving, but the eye test itself is a regulated, physical service that online retailers cannot yet absorb, and complex prescriptions, varifocals and children's fitting still favour the branch. The entrant's trajectory meets mainstream needs for frame retail, and stops short of the clinical core.
The read. This is genuine disruption of the frame-retail half of the business, and only that half. The theory says the chain should stop using frame margin to subsidise cheap eye tests, price the clinical service on its own feet, and launch or acquire an online arm run as a separate unit with its own economics, free to cannibalise branch frame sales before someone else does. Defending the bundle everywhere would repeat the incumbent's classic error of retreating upmarket one profitable step at a time.
Pitfalls
- Calling every successful entrant disruptive. The 2015 restatement by Christensen, Raynor and McDonald exists precisely because the label had drifted; an entrant without a low-end or new-market foothold is running a different play.
- Treating disruption as fast. The mechanism often takes a decade or more, and premature panic can be as costly as complacency.
- Assuming the disrupter always wins. The theory predicts incumbent difficulty, and says nothing about which of many entrants will capture the market.
- Reading incumbent retreat as stupidity. The uncomfortable core of the theory is that retreat is what good management, listening to its best customers, rationally chooses.
- Using the theory to justify self-disruption without separating the unit. Christensen's evidence is that disruptive ventures suffocate inside the core profit model.
What the critics say
Lepore's historical critique argued the theory rests on handpicked, circular case studies, that several celebrated disrupters in the disk-drive and other industries fared poorly, and that sustaining innovation succeeds far more often than the theory admits.
Lepore, J. (2014) 'The Disruption Machine: What the gospel of innovation gets wrong', The New Yorker, 23 June 2014.
King and Baatartogtokh surveyed 79 industry experts on the 77 cases cited by Christensen and Raynor and found that most did not fit the theory's four key elements well, concluding it has limited predictive power and should be treated as a warning rather than a law.
King, A. A. and Baatartogtokh, B. (2015) 'How Useful Is the Theory of Disruptive Innovation?', MIT Sloan Management Review, 57(1), Fall 2015.
Danneels argued from within innovation scholarship that the theory's key constructs, including what counts as a disruptive technology and when the prediction applies, were defined too loosely to test, a gap the 2015 HBR restatement only partly closed.
Danneels, E. (2004) 'Disruptive Technology Reconsidered: A Critique and Research Agenda', Journal of Product Innovation Management, 21(4).
Markides argued the theory conflates fundamentally different phenomena: technological disruption, business-model innovation and radical new-to-the-world products follow different competitive dynamics, are pursued by different kinds of firms and demand different incumbent responses, so a single theory of 'disruption' generates misleading advice. Business-model innovators, on his evidence, typically grow the market and coexist with incumbents rather than displacing them.
Markides, C. (2006) 'Disruptive Innovation: In Need of Better Theory', Journal of Product Innovation Management, 23(1), pp. 19-25.
Sood and Tellis tested the theory's core claims against 36 technologies in seven markets and found them largely unsupported: technologies attacking from below were introduced as often by incumbents as by entrants, were not systematically cheaper, rarely disrupted anyone, and performance trajectories crossed repeatedly rather than once, so 'disruption' is neither permanent nor the near-inevitable fate of attacked incumbents.
Sood, A. and Tellis, G. J. (2011) 'Demystifying Disruption: A New Model for Understanding and Predicting Disruptive Technologies', Marketing Science, 30(2), pp. 339-354.
Even where incumbents did fail, the causal attribution is contested: Tellis argued the disk-drive and related histories are better explained by incumbents' lack of visionary leadership and willingness to cannibalise their own products than by any property of the technologies themselves, which makes the theory's central mechanism, good management as the cause of failure, an artefact of case selection.
Tellis, G. J. (2006) 'Disruptive Technology or Visionary Leadership?', Journal of Product Innovation Management, 23(1), pp. 34-38.
Sources and further reading
- Bower, J. L. and Christensen, C. M. (1995) 'Disruptive Technologies: Catching the Wave', Harvard Business Review, 73(1), January–February 1995. ↗
- Christensen, C. M. (1997) The Innovator's Dilemma: When New Technologies Cause Great Firms to Fail. Boston: Harvard Business School Press.
- Christensen, C. M., Raynor, M. E. and McDonald, R. (2015) 'What Is Disruptive Innovation?', Harvard Business Review, 93(12), December 2015. ↗
- King, A. A. and Baatartogtokh, B. (2015) 'How Useful Is the Theory of Disruptive Innovation?', MIT Sloan Management Review, 57(1). ↗